Files
libreoffice/connectivity/source/commontools/FValue.cxx

2540 lines
87 KiB
C++
Raw Normal View History

/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
2012-06-12 22:04:38 +01:00
/*
* This file is part of the LibreOffice project.
*
* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/.
*
* This file incorporates work covered by the following license notice:
*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed
* with this work for additional information regarding copyright
* ownership. The ASF licenses this file to you under the Apache
* License, Version 2.0 (the "License"); you may not use this file
* except in compliance with the License. You may obtain a copy of
* the License at http://www.apache.org/licenses/LICENSE-2.0 .
*/
2001-03-28 10:26:47 +00:00
#include <string.h>
2001-03-28 10:26:47 +00:00
#include <stdio.h>
2001-07-24 12:18:40 +00:00
#include "connectivity/FValue.hxx"
2001-03-28 10:26:47 +00:00
#include "connectivity/CommonTools.hxx"
#include <connectivity/dbconversion.hxx>
#include <comphelper/extract.hxx>
2001-07-24 12:18:40 +00:00
#include <com/sun/star/io/XInputStream.hpp>
CWS-TOOLING: integrate CWS dbaperf2 2009-06-22 11:04:36 +0200 msc r273202 : #100000# fix to run this test 2009-06-22 09:41:04 +0200 msc r273200 : #100000# correct case 2009-06-22 09:02:42 +0200 msc r273199 : new SRB 2009-06-22 08:53:26 +0200 msc r273198 : #100000 2009-06-15 12:36:32 +0200 oj r272976 : test 2009-06-12 14:23:23 +0200 msc r272918 : #101587# unused feature was removed 2009-06-12 06:45:17 +0200 oj r272903 : use GUI instead of OS 2009-06-11 09:12:30 +0200 oj r272845 : remove unused dep 2009-06-11 09:11:09 +0200 oj r272844 : ColumnAliasInOrderBy not supported 2009-06-11 08:27:00 +0200 oj r272843 : CWS-TOOLING: rebase CWS dbaperf2 to trunk@272827 (milestone: DEV300:m50) 2009-06-10 15:30:17 +0200 msc r272822 : #100000 2009-06-10 15:27:40 +0200 msc r272820 : #102515# new function for the database type list 2009-06-10 15:26:09 +0200 msc r272818 : move file to dbaccess/tools/dbcreatetools.inc 2009-06-10 15:20:10 +0200 msc r272816 : #102515# new function for the database type list 2009-06-10 12:34:46 +0200 oj r272803 : #i101587# check extension 2009-06-10 11:53:45 +0200 oj r272800 : remove unused code 2009-06-09 09:57:25 +0200 oj r272752 : #i101587# EscapeDateTime moved into features section 2009-06-09 09:56:26 +0200 oj r272751 : enable finish 2009-06-08 11:50:37 +0200 oj r272726 : #i102588# move convert tzo saveDataSource 2009-06-05 11:54:20 +0200 jsk r272676 : #i102515# - NEW: Functions to retrieve Database names from API 2009-06-05 10:52:22 +0200 jsk r272671 : NEW: Functions to retrieve Database names from API 2009-06-05 09:36:54 +0200 oj r272666 : handle win and unx differently 2009-06-05 09:36:28 +0200 oj r272665 : handle win and unx differently 2009-06-04 13:41:56 +0200 oj r272625 : #i101268# make use of stringbuffer instead of oustring 2009-06-04 10:18:14 +0200 oj r272612 : add SAL_DLLPUBLIC_EXPORT 2009-05-28 10:56:33 +0200 oj r272381 : #i101587# create langpacks 2009-05-28 10:41:59 +0200 oj r272380 : #i101587# create langpacks 2009-05-28 10:29:30 +0200 oj r272379 : #i101587# create langpacks 2009-05-07 10:11:46 +0200 oj r271635 : fix compile warnings 2009-05-07 07:01:31 +0200 oj r271619 : change type from int32 to PathID 2009-05-07 06:53:19 +0200 oj r271618 : add all 2009-05-06 14:57:02 +0200 oj r271595 : add missing lib 2009-05-06 14:56:54 +0200 oj r271594 : no used anymore 2009-05-06 14:56:08 +0200 oj r271593 : no used anymore 2009-05-06 14:54:25 +0200 oj r271592 : no used anymore 2009-05-06 14:45:59 +0200 oj r271591 : add missing lib 2009-05-06 14:41:54 +0200 oj r271590 : change wnt 2009-05-06 14:22:21 +0200 oj r271589 : #i101587# use config for the drivers 2009-05-06 14:21:39 +0200 oj r271588 : #i101587# add new set for metadata and config data for driver 2009-05-06 14:19:36 +0200 oj r271587 : #i101587# handle metadata and props of driver 2009-05-06 10:11:11 +0200 oj r271570 : #i101587# correct wrong use of reference 2009-05-06 09:24:33 +0200 oj r271567 : #i101587# add new set for metadata 2009-05-06 08:52:06 +0200 oj r271563 : #i101587# add driver configuration 2009-05-06 08:51:04 +0200 oj r271562 : #i101587# add driver configuration 2009-05-04 10:24:41 +0200 oj r271442 : CWS-TOOLING: rebase CWS dbaperf2 to trunk@271427 (milestone: DEV300:m47) 2009-04-28 07:06:22 +0200 oj r271303 : compile error 2009-04-27 14:12:48 +0200 oj r271270 : #i101268# make use of stringbuffer instead of oustring 2009-04-27 14:12:27 +0200 oj r271269 : #i101268# make use of stringbuffer instead of oustring 2009-04-27 09:30:10 +0200 oj r271255 : use string buffer when possible 2009-04-24 11:56:01 +0200 oj r271207 : #i101268# remove end() call from loop 2009-04-23 14:24:03 +0200 oj r271171 : add dep 2009-04-23 13:25:25 +0200 oj r271164 : #i101268# add config entries for database drivers 2009-04-23 13:07:13 +0200 oj r271161 : #i101268# add config entries for database drivers 2009-04-23 13:06:33 +0200 oj r271160 : #i101268# add parent url pattern entry for driver node 2009-04-23 13:05:39 +0200 oj r271159 : #i101268# add config entry for driver node 2009-04-23 13:04:09 +0200 oj r271157 : #i101268# add config entry for driver node 2009-04-23 06:41:32 +0200 oj r271142 : #i101268# add new configuration file Drivers 2009-04-22 13:49:19 +0200 oj r271103 : convert EOL 2009-04-22 13:29:17 +0200 oj r271101 : #i101268# use of driver configuration entries for features and properties 2009-04-22 13:28:08 +0200 oj r271100 : #i101268# use of driver configuration entries for features and properties 2009-04-22 13:15:26 +0200 oj r271098 : #i101268# add new configuration file Drivers 2009-04-22 13:14:43 +0200 oj r271097 : #i101268# use of driver configuration entries for features and properties
2009-07-03 12:24:35 +00:00
#include <rtl/ustrbuf.hxx>
#include <boost/type_traits.hpp>
#include <boost/static_assert.hpp>
2001-03-28 10:26:47 +00:00
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
using namespace ::dbtools;
2001-03-28 10:26:47 +00:00
using namespace ::com::sun::star::sdbc;
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
using namespace ::com::sun::star::sdb;
2001-03-28 10:26:47 +00:00
using namespace ::com::sun::star::uno;
using namespace ::com::sun::star::util;
2001-07-24 12:18:40 +00:00
using namespace ::com::sun::star::io;
2001-03-28 10:26:47 +00:00
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
namespace connectivity
{
namespace {
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
static bool isStorageCompatible(sal_Int32 _eType1, sal_Int32 _eType2)
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::isStorageCompatible" );
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bool bIsCompatible = sal_True;
2001-06-26 12:32:47 +00:00
if (_eType1 != _eType2)
{
SAL_INFO( "connectivity.commontools", "ORowSetValue::isStorageCompatible _eType1 != _eType2" );
2001-06-26 12:32:47 +00:00
switch (_eType1)
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
bIsCompatible = (DataType::CHAR == _eType2)
|| (DataType::VARCHAR == _eType2)
|| (DataType::DECIMAL == _eType2)
|| (DataType::NUMERIC == _eType2)
|| (DataType::LONGVARCHAR == _eType2);
2001-06-26 12:32:47 +00:00
break;
2001-07-24 12:18:40 +00:00
case DataType::DOUBLE:
2001-06-26 12:32:47 +00:00
case DataType::REAL:
bIsCompatible = (DataType::DOUBLE == _eType2)
|| (DataType::REAL == _eType2);
break;
2001-06-26 12:32:47 +00:00
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
bIsCompatible = (DataType::BINARY == _eType2)
|| (DataType::VARBINARY == _eType2)
|| (DataType::LONGVARBINARY == _eType2);
2001-06-26 12:32:47 +00:00
break;
2001-07-24 12:18:40 +00:00
case DataType::INTEGER:
bIsCompatible = (DataType::SMALLINT == _eType2)
|| (DataType::TINYINT == _eType2)
|| (DataType::BIT == _eType2)
|| (DataType::BOOLEAN == _eType2);
2001-07-24 12:18:40 +00:00
break;
case DataType::SMALLINT:
bIsCompatible = (DataType::TINYINT == _eType2)
|| (DataType::BIT == _eType2)
|| (DataType::BOOLEAN == _eType2);
2001-07-24 12:18:40 +00:00
break;
case DataType::TINYINT:
bIsCompatible = (DataType::BIT == _eType2)
|| (DataType::BOOLEAN == _eType2);
2001-07-24 12:18:40 +00:00
break;
case DataType::BLOB:
case DataType::CLOB:
case DataType::OBJECT:
bIsCompatible = (DataType::BLOB == _eType2)
|| (DataType::CLOB == _eType2)
|| (DataType::OBJECT == _eType2);
break;
2001-06-26 12:32:47 +00:00
default:
bIsCompatible = sal_False;
}
}
2001-06-26 12:32:47 +00:00
return bIsCompatible;
}
}
// -----------------------------------------------------------------------------
#ifdef DBG_UTIL
#include <vector>
#include <rtl/string.h>
namespace tracing
{
struct AllocationType
{
const sal_Char* pName;
sal_Int32 nAllocatedUnits;
AllocationType( ) : pName( NULL ), nAllocatedUnits( 0 ) { }
};
// =============================================================================
class AllocationTracer
{
public:
typedef ::std::vector< AllocationType > AllocationState;
static AllocationState s_aAllocated;
static ::osl::Mutex s_aMutex;
public:
static void registerUnit( const sal_Char* _pName );
static void revokeUnit( const sal_Char* _pName );
private:
static AllocationState::iterator getLocation( const sal_Char* _pName );
};
// =============================================================================
AllocationTracer::AllocationState::iterator AllocationTracer::getLocation( const sal_Char* _pName )
{
AllocationState::iterator aLookFor = s_aAllocated.begin();
for ( ;
aLookFor != s_aAllocated.end();
++aLookFor
)
{
if ( 0 == rtl_str_compare( aLookFor->pName, _pName ) )
// found
return aLookFor;
}
// not found
s_aAllocated.push_back( AllocationType() );
aLookFor = s_aAllocated.end(); --aLookFor;
aLookFor->pName = _pName; // note that this assumes that _pName is a constant string ....
return aLookFor;
}
// =============================================================================
AllocationTracer::AllocationState AllocationTracer::s_aAllocated;
::osl::Mutex AllocationTracer::s_aMutex;
// =============================================================================
void AllocationTracer::registerUnit( const sal_Char* _pName )
{
::osl::MutexGuard aGuard( s_aMutex );
AllocationState::iterator aPos = getLocation( _pName );
++aPos->nAllocatedUnits;
}
// =============================================================================
void AllocationTracer::revokeUnit( const sal_Char* _pName )
{
::osl::MutexGuard aGuard( s_aMutex );
AllocationState::iterator aPos = getLocation( _pName );
--aPos->nAllocatedUnits;
}
#define TRACE_ALLOC( type ) tracing::AllocationTracer::registerUnit( #type );
#define TRACE_FREE( type ) tracing::AllocationTracer::revokeUnit( #type );
}
#else
#define TRACE_ALLOC( type )
#define TRACE_FREE( type )
#endif
// -----------------------------------------------------------------------------
void ORowSetValue::setTypeKind(sal_Int32 _eType)
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::setTypeKind" );
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
if ( !m_bNull && !isStorageCompatible(_eType, m_eTypeKind) )
{
switch(_eType)
2001-07-24 12:18:40 +00:00
{
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
case DataType::VARCHAR:
case DataType::CHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
(*this) = getString();
break;
case DataType::BIGINT:
{
sal_Int64 nVal(getLong());
sal_uInt64 nuVal(getULong());
if (nVal == 0 && nuVal != 0)
(*this) = nuVal;
else
(*this) = nVal;
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
break;
}
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
case DataType::FLOAT:
(*this) = getFloat();
break;
case DataType::DOUBLE:
case DataType::REAL:
(*this) = getDouble();
break;
case DataType::TINYINT:
(*this) = getInt8();
break;
case DataType::SMALLINT:
(*this) = getInt16();
break;
case DataType::INTEGER:
{
sal_Int32 nVal(getInt32());
sal_uInt32 nuVal(getUInt32());
if (nVal == 0 && nuVal != 0)
(*this) = nuVal;
else
(*this) = nVal;
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
break;
}
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
case DataType::BIT:
case DataType::BOOLEAN:
(*this) = getBool();
break;
case DataType::DATE:
(*this) = getDate();
break;
case DataType::TIME:
(*this) = getTime();
break;
case DataType::TIMESTAMP:
(*this) = getDateTime();
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
(*this) = getSequence();
break;
case DataType::BLOB:
case DataType::CLOB:
case DataType::OBJECT:
2009-11-11 13:30:02 +01:00
case DataType::OTHER:
(*this) = makeAny();
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
break;
default:
(*this) = makeAny();
SAL_WARN( "connectivity.commontools","ORowSetValue::setTypeKind(): UNSUPPORTED TYPE!");
2001-07-24 12:18:40 +00:00
}
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
}
m_eTypeKind = _eType;
}
2001-03-28 10:26:47 +00:00
// -----------------------------------------------------------------------------
void ORowSetValue::free()
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::free" );
2001-03-28 10:26:47 +00:00
if(!m_bNull)
{
switch(m_eTypeKind)
{
2001-07-24 12:18:40 +00:00
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
2001-03-28 10:26:47 +00:00
OSL_ENSURE(m_aValue.m_pString,"String pointer is null!");
rtl_uString_release(m_aValue.m_pString);
m_aValue.m_pString = NULL;
break;
2001-07-24 12:18:40 +00:00
case DataType::DATE:
2001-03-28 10:26:47 +00:00
delete (::com::sun::star::util::Date*)m_aValue.m_pValue;
TRACE_FREE( Date )
2001-03-28 10:26:47 +00:00
m_aValue.m_pValue = NULL;
break;
2001-07-24 12:18:40 +00:00
case DataType::TIME:
2001-03-28 10:26:47 +00:00
delete (::com::sun::star::util::Time*)m_aValue.m_pValue;
TRACE_FREE( Time )
2001-03-28 10:26:47 +00:00
m_aValue.m_pValue = NULL;
break;
2001-07-24 12:18:40 +00:00
case DataType::TIMESTAMP:
2001-03-28 10:26:47 +00:00
delete (::com::sun::star::util::DateTime*)m_aValue.m_pValue;
TRACE_FREE( DateTime )
2001-03-28 10:26:47 +00:00
m_aValue.m_pValue = NULL;
break;
2001-07-24 12:18:40 +00:00
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
delete (Sequence<sal_Int8>*)m_aValue.m_pValue;
TRACE_FREE( Sequence_sal_Int8 )
2001-03-28 10:26:47 +00:00
m_aValue.m_pValue = NULL;
break;
case DataType::BLOB:
case DataType::CLOB:
2001-07-24 12:18:40 +00:00
case DataType::OBJECT:
delete (Any*)m_aValue.m_pValue;
TRACE_FREE( Any )
2001-03-28 10:26:47 +00:00
m_aValue.m_pValue = NULL;
break;
case DataType::BIT:
case DataType::TINYINT:
case DataType::SMALLINT:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
case DataType::INTEGER:
case DataType::BIGINT:
case DataType::BOOLEAN:
case DataType::FLOAT:
case DataType::DOUBLE:
case DataType::REAL:
break;
default:
if ( m_aValue.m_pValue )
{
delete (Any*)m_aValue.m_pValue;
TRACE_FREE( Any )
m_aValue.m_pValue = NULL;
}
break;
2001-03-28 10:26:47 +00:00
}
m_bNull = sal_True;
}
}
// -----------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const ORowSetValue& _rRH)
{
if(&_rRH == this)
return *this;
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
if ( m_eTypeKind != _rRH.m_eTypeKind || (_rRH.m_bNull && !m_bNull) || m_bSigned != _rRH.m_bSigned)
2001-05-07 11:23:42 +00:00
free();
2001-03-28 10:26:47 +00:00
m_bBound = _rRH.m_bBound;
m_eTypeKind = _rRH.m_eTypeKind;
m_bSigned = _rRH.m_bSigned;
2001-05-07 11:23:42 +00:00
if(m_bNull && !_rRH.m_bNull)
2001-03-28 10:26:47 +00:00
{
switch(_rRH.m_eTypeKind)
{
2001-05-07 11:23:42 +00:00
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
2001-10-08 03:53:53 +00:00
rtl_uString_acquire(_rRH.m_aValue.m_pString);
2001-05-07 11:23:42 +00:00
m_aValue.m_pString = _rRH.m_aValue.m_pString;
break;
case DataType::DATE:
m_aValue.m_pValue = new Date(*(Date*)_rRH.m_aValue.m_pValue);
TRACE_ALLOC( Date )
2001-05-07 11:23:42 +00:00
break;
case DataType::TIME:
m_aValue.m_pValue = new Time(*(Time*)_rRH.m_aValue.m_pValue);
TRACE_ALLOC( Time )
2001-05-07 11:23:42 +00:00
break;
case DataType::TIMESTAMP:
m_aValue.m_pValue = new DateTime(*(DateTime*)_rRH.m_aValue.m_pValue);
TRACE_ALLOC( DateTime )
2001-05-07 11:23:42 +00:00
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
m_aValue.m_pValue = new Sequence<sal_Int8>(*(Sequence<sal_Int8>*)_rRH.m_aValue.m_pValue);
TRACE_ALLOC( Sequence_sal_Int8 )
2001-05-07 11:23:42 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-05-07 11:23:42 +00:00
m_aValue.m_bBool = _rRH.m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( _rRH.m_bSigned )
m_aValue.m_nInt8 = _rRH.m_aValue.m_nInt8;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_aValue.m_uInt8 = _rRH.m_aValue.m_uInt8;
2001-10-08 03:53:53 +00:00
break;
2001-05-07 11:23:42 +00:00
case DataType::SMALLINT:
if ( _rRH.m_bSigned )
m_aValue.m_nInt16 = _rRH.m_aValue.m_nInt16;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_aValue.m_uInt16 = _rRH.m_aValue.m_uInt16;
2001-10-08 03:53:53 +00:00
break;
2001-05-07 11:23:42 +00:00
case DataType::INTEGER:
if ( _rRH.m_bSigned )
m_aValue.m_nInt32 = _rRH.m_aValue.m_nInt32;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_aValue.m_uInt32 = _rRH.m_aValue.m_uInt32;
break;
case DataType::BIGINT:
if ( _rRH.m_bSigned )
m_aValue.m_nInt64 = _rRH.m_aValue.m_nInt64;
else
m_aValue.m_uInt64 = _rRH.m_aValue.m_uInt64;
2001-05-07 11:23:42 +00:00
break;
case DataType::FLOAT:
m_aValue.m_nFloat = _rRH.m_aValue.m_nFloat;
break;
case DataType::DOUBLE:
case DataType::REAL:
m_aValue.m_nDouble = _rRH.m_aValue.m_nDouble;
break;
2001-05-07 11:23:42 +00:00
default:
2001-07-24 12:18:40 +00:00
m_aValue.m_pValue = new Any(*(Any*)_rRH.m_aValue.m_pValue);
TRACE_ALLOC( Any )
2001-03-28 10:26:47 +00:00
}
}
2001-05-07 11:23:42 +00:00
else if(!_rRH.m_bNull)
{
switch(_rRH.m_eTypeKind)
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
(*this) = OUString(_rRH.m_aValue.m_pString);
2001-05-07 11:23:42 +00:00
break;
case DataType::DATE:
(*this) = *(Date*)_rRH.m_aValue.m_pValue;
2001-08-06 05:21:03 +00:00
break;
2001-05-07 11:23:42 +00:00
case DataType::TIME:
(*this) = *(Time*)_rRH.m_aValue.m_pValue;
break;
case DataType::TIMESTAMP:
(*this) = *(DateTime*)_rRH.m_aValue.m_pValue;
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
(*this) = *(Sequence<sal_Int8>*)_rRH.m_aValue.m_pValue;
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-05-07 11:23:42 +00:00
m_aValue.m_bBool = _rRH.m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( _rRH.m_bSigned )
m_aValue.m_nInt8 = _rRH.m_aValue.m_nInt8;
else
m_aValue.m_uInt8 = _rRH.m_aValue.m_uInt8;
break;
2001-05-07 11:23:42 +00:00
case DataType::SMALLINT:
if ( _rRH.m_bSigned )
m_aValue.m_nInt16 = _rRH.m_aValue.m_nInt16;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_aValue.m_uInt16 = _rRH.m_aValue.m_uInt16;
break;
2001-05-07 11:23:42 +00:00
case DataType::INTEGER:
if ( _rRH.m_bSigned )
m_aValue.m_nInt32 = _rRH.m_aValue.m_nInt32;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_aValue.m_uInt32 = _rRH.m_aValue.m_uInt32;
break;
case DataType::BIGINT:
if ( _rRH.m_bSigned )
m_aValue.m_nInt64 = _rRH.m_aValue.m_nInt64;
else
m_aValue.m_uInt64 = _rRH.m_aValue.m_uInt64;
2001-05-07 11:23:42 +00:00
break;
case DataType::FLOAT:
m_aValue.m_nFloat = _rRH.m_aValue.m_nFloat;
break;
case DataType::DOUBLE:
case DataType::REAL:
m_aValue.m_nDouble = _rRH.m_aValue.m_nDouble;
break;
2001-05-07 11:23:42 +00:00
default:
2001-07-24 12:18:40 +00:00
(*(Any*)m_aValue.m_pValue) = (*(Any*)_rRH.m_aValue.m_pValue);
2001-05-07 11:23:42 +00:00
}
}
m_bNull = _rRH.m_bNull;
// OJ: BUGID: 96277
m_eTypeKind = _rRH.m_eTypeKind;
2001-03-28 10:26:47 +00:00
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const Date& _rRH)
{
2001-05-07 11:23:42 +00:00
if(m_eTypeKind != DataType::DATE)
free();
2001-03-28 10:26:47 +00:00
if(m_bNull)
{
m_aValue.m_pValue = new Date(_rRH);
TRACE_ALLOC( Date )
2001-03-28 10:26:47 +00:00
m_eTypeKind = DataType::DATE;
m_bNull = sal_False;
2001-03-28 10:26:47 +00:00
}
else
*(Date*)m_aValue.m_pValue = _rRH;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const Time& _rRH)
{
2001-05-07 11:23:42 +00:00
if(m_eTypeKind != DataType::TIME)
free();
2001-03-28 10:26:47 +00:00
if(m_bNull)
{
m_aValue.m_pValue = new Time(_rRH);
TRACE_ALLOC( Time )
2001-03-28 10:26:47 +00:00
m_eTypeKind = DataType::TIME;
m_bNull = sal_False;
2001-03-28 10:26:47 +00:00
}
else
*(Time*)m_aValue.m_pValue = _rRH;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const DateTime& _rRH)
{
2001-05-07 11:23:42 +00:00
if(m_eTypeKind != DataType::TIMESTAMP)
free();
2001-03-28 10:26:47 +00:00
if(m_bNull)
{
m_aValue.m_pValue = new DateTime(_rRH);
TRACE_ALLOC( DateTime )
2001-03-28 10:26:47 +00:00
m_eTypeKind = DataType::TIMESTAMP;
m_bNull = sal_False;
2001-03-28 10:26:47 +00:00
}
else
*(DateTime*)m_aValue.m_pValue = _rRH;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const OUString& _rRH)
2001-03-28 10:26:47 +00:00
{
2001-05-07 11:23:42 +00:00
if(m_eTypeKind != DataType::VARCHAR || m_aValue.m_pString != _rRH.pData)
{
free();
m_bNull = sal_False;
2001-03-28 10:26:47 +00:00
2001-05-07 11:23:42 +00:00
m_aValue.m_pString = _rRH.pData;
rtl_uString_acquire(m_aValue.m_pString);
m_eTypeKind = DataType::VARCHAR;
}
2001-03-28 10:26:47 +00:00
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const double& _rRH)
{
if(m_eTypeKind != DataType::DOUBLE)
2001-05-07 11:23:42 +00:00
free();
m_aValue.m_nDouble = _rRH;
m_eTypeKind = DataType::DOUBLE;
m_bNull = sal_False;
2001-03-28 10:26:47 +00:00
return *this;
}
2001-07-24 12:18:40 +00:00
// -----------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const float& _rRH)
{
if(m_eTypeKind != DataType::FLOAT)
free();
m_aValue.m_nFloat = _rRH;
m_eTypeKind = DataType::FLOAT;
m_bNull = sal_False;
2001-07-24 12:18:40 +00:00
return *this;
}
2001-03-28 10:26:47 +00:00
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const sal_Int8& _rRH)
{
if(m_eTypeKind != DataType::TINYINT )
2001-05-07 11:23:42 +00:00
free();
2001-03-28 10:26:47 +00:00
m_aValue.m_nInt8 = _rRH;
m_eTypeKind = DataType::TINYINT;
m_bNull = sal_False;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_bSigned = sal_True;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const sal_uInt8& _rRH)
{
if(m_eTypeKind != DataType::TINYINT )
free();
m_aValue.m_uInt8 = _rRH;
m_eTypeKind = DataType::TINYINT;
m_bNull = sal_False;
m_bSigned = sal_False;
2001-03-28 10:26:47 +00:00
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const sal_Int16& _rRH)
{
if(m_eTypeKind != DataType::SMALLINT )
2001-05-07 11:23:42 +00:00
free();
2001-03-28 10:26:47 +00:00
m_aValue.m_nInt16 = _rRH;
m_eTypeKind = DataType::SMALLINT;
m_bNull = sal_False;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_bSigned = sal_True;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const sal_uInt16& _rRH)
{
if(m_eTypeKind != DataType::SMALLINT )
free();
m_aValue.m_uInt16 = _rRH;
m_eTypeKind = DataType::SMALLINT;
m_bNull = sal_False;
m_bSigned = sal_False;
2001-03-28 10:26:47 +00:00
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const sal_Int32& _rRH)
{
if(m_eTypeKind != DataType::INTEGER )
2001-05-07 11:23:42 +00:00
free();
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_aValue.m_nInt32 = _rRH;
m_eTypeKind = DataType::INTEGER;
m_bNull = sal_False;
m_bSigned = sal_True;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const sal_uInt32& _rRH)
{
if(m_eTypeKind != DataType::INTEGER )
free();
m_aValue.m_uInt32 = _rRH;
2001-03-28 10:26:47 +00:00
m_eTypeKind = DataType::INTEGER;
m_bNull = sal_False;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_bSigned = sal_False;
2001-03-28 10:26:47 +00:00
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
ORowSetValue& ORowSetValue::operator=(const bool _rRH)
2001-03-28 10:26:47 +00:00
{
if(m_eTypeKind != DataType::BIT && DataType::BOOLEAN != m_eTypeKind )
2001-05-07 11:23:42 +00:00
free();
2001-03-28 10:26:47 +00:00
m_aValue.m_bBool = _rRH;
m_eTypeKind = DataType::BOOLEAN;
m_bNull = sal_False;
2001-03-28 10:26:47 +00:00
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const sal_Int64& _rRH)
{
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
if ( DataType::BIGINT != m_eTypeKind)
free();
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_aValue.m_nInt64 = _rRH;
m_eTypeKind = DataType::BIGINT;
m_bNull = sal_False;
m_bSigned = sal_True;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const sal_uInt64& _rRH)
{
if ( DataType::BIGINT != m_eTypeKind)
free();
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_aValue.m_uInt64 = _rRH;
2001-09-20 11:51:56 +00:00
m_eTypeKind = DataType::BIGINT;
2001-03-28 10:26:47 +00:00
m_bNull = sal_False;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
m_bSigned = sal_False;
2001-03-28 10:26:47 +00:00
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const Sequence<sal_Int8>& _rRH)
{
if (!isStorageCompatible(DataType::LONGVARBINARY,m_eTypeKind))
free();
if (m_bNull)
{
2001-03-28 10:26:47 +00:00
m_aValue.m_pValue = new Sequence<sal_Int8>(_rRH);
TRACE_ALLOC( Sequence_sal_Int8 )
}
2001-03-28 10:26:47 +00:00
else
*static_cast< Sequence< sal_Int8 >* >(m_aValue.m_pValue) = _rRH;
2001-03-28 10:26:47 +00:00
m_eTypeKind = DataType::LONGVARBINARY;
2001-03-28 10:26:47 +00:00
m_bNull = sal_False;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue& ORowSetValue::operator=(const Any& _rAny)
{
if (!isStorageCompatible(DataType::OBJECT,m_eTypeKind))
free();
if ( m_bNull )
{
2001-03-28 10:26:47 +00:00
m_aValue.m_pValue = new Any(_rAny);
TRACE_ALLOC( Any )
}
2001-03-28 10:26:47 +00:00
else
*static_cast<Any*>(m_aValue.m_pValue) = _rAny;
2001-03-28 10:26:47 +00:00
m_eTypeKind = DataType::OBJECT;
2001-03-28 10:26:47 +00:00
m_bNull = sal_False;
return *this;
}
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bool operator==(const Date& _rLH,const Date& _rRH)
2001-03-28 10:26:47 +00:00
{
return _rLH.Day == _rRH.Day && _rLH.Month == _rRH.Month && _rLH.Year == _rRH.Year;
}
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bool operator==(const Time& _rLH,const Time& _rRH)
2001-03-28 10:26:47 +00:00
{
return _rLH.Minutes == _rRH.Minutes && _rLH.Hours == _rRH.Hours && _rLH.Seconds == _rRH.Seconds && _rLH.NanoSeconds == _rRH.NanoSeconds;
2001-03-28 10:26:47 +00:00
}
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bool operator==(const DateTime& _rLH,const DateTime& _rRH)
2001-03-28 10:26:47 +00:00
{
return _rLH.Day == _rRH.Day && _rLH.Month == _rRH.Month && _rLH.Year == _rRH.Year &&
_rLH.Minutes == _rRH.Minutes && _rLH.Hours == _rRH.Hours && _rLH.Seconds == _rRH.Seconds && _rLH.NanoSeconds == _rRH.NanoSeconds;
2001-03-28 10:26:47 +00:00
}
// -------------------------------------------------------------------------
bool ORowSetValue::operator==(const ORowSetValue& _rRH) const
2001-03-28 10:26:47 +00:00
{
if ( m_bNull != _rRH.isNull() )
return false;
if(m_bNull && _rRH.isNull())
return true;
if ( m_eTypeKind != _rRH.m_eTypeKind )
{
switch(m_eTypeKind)
{
case DataType::FLOAT:
case DataType::DOUBLE:
case DataType::REAL:
return getDouble() == _rRH.getDouble();
default:
switch(_rRH.m_eTypeKind)
{
case DataType::FLOAT:
case DataType::DOUBLE:
case DataType::REAL:
return getDouble() == _rRH.getDouble();
default:
break;
}
break;
}
return false;
}
2001-03-28 10:26:47 +00:00
bool bRet = false;
2001-10-08 03:53:53 +00:00
OSL_ENSURE(!m_bNull,"SHould not be null!");
switch(m_eTypeKind)
2001-03-28 10:26:47 +00:00
{
2001-10-08 03:53:53 +00:00
case DataType::VARCHAR:
case DataType::CHAR:
case DataType::LONGVARCHAR:
2001-03-28 10:26:47 +00:00
{
OUString aVal1(m_aValue.m_pString);
OUString aVal2(_rRH.m_aValue.m_pString);
return aVal1 == aVal2;
2001-03-28 10:26:47 +00:00
}
default:
if ( m_bSigned != _rRH.m_bSigned )
return false;
break;
}
switch(m_eTypeKind)
{
case DataType::DECIMAL:
case DataType::NUMERIC:
{
OUString aVal1(m_aValue.m_pString);
OUString aVal2(_rRH.m_aValue.m_pString);
bRet = aVal1 == aVal2;
}
break;
2001-10-08 03:53:53 +00:00
case DataType::FLOAT:
bRet = m_aValue.m_nFloat == _rRH.m_aValue.m_nFloat;
2001-10-08 03:53:53 +00:00
break;
case DataType::DOUBLE:
case DataType::REAL:
bRet = m_aValue.m_nDouble == _rRH.m_aValue.m_nDouble;
2001-10-08 03:53:53 +00:00
break;
case DataType::TINYINT:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bRet = m_bSigned ? ( m_aValue.m_nInt8 == _rRH.m_aValue.m_nInt8 ) : (m_aValue.m_uInt8 == _rRH.m_aValue.m_uInt8);
2001-10-08 03:53:53 +00:00
break;
case DataType::SMALLINT:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bRet = m_bSigned ? ( m_aValue.m_nInt16 == _rRH.m_aValue.m_nInt16 ) : (m_aValue.m_uInt16 == _rRH.m_aValue.m_uInt16);
2001-10-08 03:53:53 +00:00
break;
case DataType::INTEGER:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bRet = m_bSigned ? ( m_aValue.m_nInt32 == _rRH.m_aValue.m_nInt32 ) : (m_aValue.m_uInt32 == _rRH.m_aValue.m_uInt32);
break;
case DataType::BIGINT:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bRet = m_bSigned ? ( m_aValue.m_nInt64 == _rRH.m_aValue.m_nInt64 ) : (m_aValue.m_uInt64 == _rRH.m_aValue.m_uInt64);
2001-10-08 03:53:53 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-10-08 03:53:53 +00:00
bRet = m_aValue.m_bBool == _rRH.m_aValue.m_bBool;
break;
case DataType::DATE:
bRet = *(Date*)m_aValue.m_pValue == *(Date*)_rRH.m_aValue.m_pValue;
break;
case DataType::TIME:
bRet = *(Time*)m_aValue.m_pValue == *(Time*)_rRH.m_aValue.m_pValue;
break;
case DataType::TIMESTAMP:
bRet = *(DateTime*)m_aValue.m_pValue == *(DateTime*)_rRH.m_aValue.m_pValue;
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
bRet = false;
2001-10-08 03:53:53 +00:00
break;
case DataType::BLOB:
case DataType::CLOB:
case DataType::OBJECT:
2009-11-11 13:30:02 +01:00
case DataType::OTHER:
bRet = false;
break;
2001-10-08 03:53:53 +00:00
default:
bRet = false;
SAL_WARN( "connectivity.commontools","ORowSetValue::operator==(): UNSUPPORTED TYPE!");
break;
2001-03-28 10:26:47 +00:00
}
return bRet;
}
// -------------------------------------------------------------------------
Any ORowSetValue::makeAny() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::makeAny" );
2001-03-28 10:26:47 +00:00
Any rValue;
if(isBound() && !isNull())
{
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
2001-09-20 11:51:56 +00:00
OSL_ENSURE(m_aValue.m_pString,"Value is null!");
rValue <<= (OUString)m_aValue.m_pString;
2001-03-28 10:26:47 +00:00
break;
case DataType::FLOAT:
rValue <<= m_aValue.m_nFloat;
2001-07-24 12:18:40 +00:00
break;
case DataType::DOUBLE:
2001-03-28 10:26:47 +00:00
case DataType::REAL:
rValue <<= m_aValue.m_nDouble;
2001-03-28 10:26:47 +00:00
break;
case DataType::DATE:
2001-07-30 08:11:53 +00:00
OSL_ENSURE(m_aValue.m_pValue,"Value is null!");
2001-03-28 10:26:47 +00:00
rValue <<= *(Date*)m_aValue.m_pValue;
break;
case DataType::TIME:
2001-07-30 08:11:53 +00:00
OSL_ENSURE(m_aValue.m_pValue,"Value is null!");
2001-03-28 10:26:47 +00:00
rValue <<= *(Time*)m_aValue.m_pValue;
break;
case DataType::TIMESTAMP:
2001-07-30 08:11:53 +00:00
OSL_ENSURE(m_aValue.m_pValue,"Value is null!");
2001-03-28 10:26:47 +00:00
rValue <<= *(DateTime*)m_aValue.m_pValue;
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
2001-07-30 08:11:53 +00:00
OSL_ENSURE(m_aValue.m_pValue,"Value is null!");
2001-03-28 10:26:47 +00:00
rValue <<= *(Sequence<sal_Int8>*)m_aValue.m_pValue;
break;
case DataType::BLOB:
case DataType::CLOB:
2001-03-28 10:26:47 +00:00
case DataType::OBJECT:
2009-11-11 13:30:02 +01:00
case DataType::OTHER:
2001-03-28 10:26:47 +00:00
rValue = getAny();
break;
case DataType::BIT:
case DataType::BOOLEAN:
rValue <<= m_aValue.m_bBool;
2001-03-28 10:26:47 +00:00
break;
case DataType::TINYINT:
if ( m_bSigned )
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
// TypeClass_BYTE
rValue <<= m_aValue.m_nInt8;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
// There is no TypeClass_UNSIGNED_BYTE,
// so silently promote it to a 16-bit integer,
// that is TypeClass_UNSIGNED_SHORT
rValue <<= static_cast<sal_uInt16>(m_aValue.m_uInt8);
2001-03-28 10:26:47 +00:00
break;
case DataType::SMALLINT:
if ( m_bSigned )
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
// TypeClass_SHORT
rValue <<= m_aValue.m_nInt16;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
// TypeClass_UNSIGNED_SHORT
rValue <<= m_aValue.m_uInt16;
2001-03-28 10:26:47 +00:00
break;
case DataType::INTEGER:
if ( m_bSigned )
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
// TypeClass_LONG
rValue <<= m_aValue.m_nInt32;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
// TypeClass_UNSIGNED_LONG
rValue <<= m_aValue.m_uInt32;
break;
case DataType::BIGINT:
if ( m_bSigned )
// TypeClass_HYPER
rValue <<= m_aValue.m_nInt64;
else
// TypeClass_UNSIGNED_HYPER
rValue <<= m_aValue.m_uInt64;
2001-03-28 10:26:47 +00:00
break;
default:
SAL_WARN( "connectivity.commontools","ORowSetValue::makeAny(): UNSPUPPORTED TYPE!");
rValue = getAny();
break;
2001-03-28 10:26:47 +00:00
}
}
return rValue;
}
// -------------------------------------------------------------------------
OUString ORowSetValue::getString( ) const
2001-03-28 10:26:47 +00:00
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getString" );
OUString aRet;
2001-05-14 10:42:44 +00:00
if(!m_bNull)
2001-03-28 10:26:47 +00:00
{
2001-05-14 10:42:44 +00:00
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
2001-05-14 10:42:44 +00:00
aRet = m_aValue.m_pString;
break;
case DataType::FLOAT:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
aRet = OUString::number(static_cast<float>(*this));
2001-07-24 12:18:40 +00:00
break;
case DataType::DOUBLE:
2001-05-14 10:42:44 +00:00
case DataType::REAL:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
aRet = OUString::number(static_cast<double>(*this));
2001-05-14 10:42:44 +00:00
break;
case DataType::DATE:
aRet = DBTypeConversion::toDateString(*this);
2001-05-14 10:42:44 +00:00
break;
case DataType::TIME:
aRet = DBTypeConversion::toTimeString(*this);
2001-05-14 10:42:44 +00:00
break;
case DataType::TIMESTAMP:
aRet = DBTypeConversion::toDateTimeString(*this);
2001-05-14 10:42:44 +00:00
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
{
OUStringBuffer sVal(OUString("0x"));
2001-05-14 10:42:44 +00:00
Sequence<sal_Int8> aSeq(getSequence());
const sal_Int8* pBegin = aSeq.getConstArray();
const sal_Int8* pEnd = pBegin + aSeq.getLength();
for(;pBegin != pEnd;++pBegin)
CWS-TOOLING: integrate CWS dbaperf2 2009-06-22 11:04:36 +0200 msc r273202 : #100000# fix to run this test 2009-06-22 09:41:04 +0200 msc r273200 : #100000# correct case 2009-06-22 09:02:42 +0200 msc r273199 : new SRB 2009-06-22 08:53:26 +0200 msc r273198 : #100000 2009-06-15 12:36:32 +0200 oj r272976 : test 2009-06-12 14:23:23 +0200 msc r272918 : #101587# unused feature was removed 2009-06-12 06:45:17 +0200 oj r272903 : use GUI instead of OS 2009-06-11 09:12:30 +0200 oj r272845 : remove unused dep 2009-06-11 09:11:09 +0200 oj r272844 : ColumnAliasInOrderBy not supported 2009-06-11 08:27:00 +0200 oj r272843 : CWS-TOOLING: rebase CWS dbaperf2 to trunk@272827 (milestone: DEV300:m50) 2009-06-10 15:30:17 +0200 msc r272822 : #100000 2009-06-10 15:27:40 +0200 msc r272820 : #102515# new function for the database type list 2009-06-10 15:26:09 +0200 msc r272818 : move file to dbaccess/tools/dbcreatetools.inc 2009-06-10 15:20:10 +0200 msc r272816 : #102515# new function for the database type list 2009-06-10 12:34:46 +0200 oj r272803 : #i101587# check extension 2009-06-10 11:53:45 +0200 oj r272800 : remove unused code 2009-06-09 09:57:25 +0200 oj r272752 : #i101587# EscapeDateTime moved into features section 2009-06-09 09:56:26 +0200 oj r272751 : enable finish 2009-06-08 11:50:37 +0200 oj r272726 : #i102588# move convert tzo saveDataSource 2009-06-05 11:54:20 +0200 jsk r272676 : #i102515# - NEW: Functions to retrieve Database names from API 2009-06-05 10:52:22 +0200 jsk r272671 : NEW: Functions to retrieve Database names from API 2009-06-05 09:36:54 +0200 oj r272666 : handle win and unx differently 2009-06-05 09:36:28 +0200 oj r272665 : handle win and unx differently 2009-06-04 13:41:56 +0200 oj r272625 : #i101268# make use of stringbuffer instead of oustring 2009-06-04 10:18:14 +0200 oj r272612 : add SAL_DLLPUBLIC_EXPORT 2009-05-28 10:56:33 +0200 oj r272381 : #i101587# create langpacks 2009-05-28 10:41:59 +0200 oj r272380 : #i101587# create langpacks 2009-05-28 10:29:30 +0200 oj r272379 : #i101587# create langpacks 2009-05-07 10:11:46 +0200 oj r271635 : fix compile warnings 2009-05-07 07:01:31 +0200 oj r271619 : change type from int32 to PathID 2009-05-07 06:53:19 +0200 oj r271618 : add all 2009-05-06 14:57:02 +0200 oj r271595 : add missing lib 2009-05-06 14:56:54 +0200 oj r271594 : no used anymore 2009-05-06 14:56:08 +0200 oj r271593 : no used anymore 2009-05-06 14:54:25 +0200 oj r271592 : no used anymore 2009-05-06 14:45:59 +0200 oj r271591 : add missing lib 2009-05-06 14:41:54 +0200 oj r271590 : change wnt 2009-05-06 14:22:21 +0200 oj r271589 : #i101587# use config for the drivers 2009-05-06 14:21:39 +0200 oj r271588 : #i101587# add new set for metadata and config data for driver 2009-05-06 14:19:36 +0200 oj r271587 : #i101587# handle metadata and props of driver 2009-05-06 10:11:11 +0200 oj r271570 : #i101587# correct wrong use of reference 2009-05-06 09:24:33 +0200 oj r271567 : #i101587# add new set for metadata 2009-05-06 08:52:06 +0200 oj r271563 : #i101587# add driver configuration 2009-05-06 08:51:04 +0200 oj r271562 : #i101587# add driver configuration 2009-05-04 10:24:41 +0200 oj r271442 : CWS-TOOLING: rebase CWS dbaperf2 to trunk@271427 (milestone: DEV300:m47) 2009-04-28 07:06:22 +0200 oj r271303 : compile error 2009-04-27 14:12:48 +0200 oj r271270 : #i101268# make use of stringbuffer instead of oustring 2009-04-27 14:12:27 +0200 oj r271269 : #i101268# make use of stringbuffer instead of oustring 2009-04-27 09:30:10 +0200 oj r271255 : use string buffer when possible 2009-04-24 11:56:01 +0200 oj r271207 : #i101268# remove end() call from loop 2009-04-23 14:24:03 +0200 oj r271171 : add dep 2009-04-23 13:25:25 +0200 oj r271164 : #i101268# add config entries for database drivers 2009-04-23 13:07:13 +0200 oj r271161 : #i101268# add config entries for database drivers 2009-04-23 13:06:33 +0200 oj r271160 : #i101268# add parent url pattern entry for driver node 2009-04-23 13:05:39 +0200 oj r271159 : #i101268# add config entry for driver node 2009-04-23 13:04:09 +0200 oj r271157 : #i101268# add config entry for driver node 2009-04-23 06:41:32 +0200 oj r271142 : #i101268# add new configuration file Drivers 2009-04-22 13:49:19 +0200 oj r271103 : convert EOL 2009-04-22 13:29:17 +0200 oj r271101 : #i101268# use of driver configuration entries for features and properties 2009-04-22 13:28:08 +0200 oj r271100 : #i101268# use of driver configuration entries for features and properties 2009-04-22 13:15:26 +0200 oj r271098 : #i101268# add new configuration file Drivers 2009-04-22 13:14:43 +0200 oj r271097 : #i101268# use of driver configuration entries for features and properties
2009-07-03 12:24:35 +00:00
sVal.append((sal_Int32)*pBegin,16);
aRet = sVal.makeStringAndClear();
2001-05-14 10:42:44 +00:00
}
break;
case DataType::BIT:
aRet = OUString::number(static_cast<bool>(*this));
break;
case DataType::BOOLEAN:
aRet = OUString::boolean(static_cast<bool>(*this));
2001-05-14 10:42:44 +00:00
break;
case DataType::TINYINT:
case DataType::SMALLINT:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
case DataType::INTEGER:
if ( m_bSigned )
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
aRet = OUString::number(static_cast<sal_Int32>(*this));
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
aRet = OUString::number(static_cast<sal_uInt32>(*this));
2001-05-14 10:42:44 +00:00
break;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
case DataType::BIGINT:
if ( m_bSigned )
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
aRet = OUString::number(static_cast<sal_Int64>(*this));
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
aRet = OUString::number(static_cast<sal_uInt64>(*this));
2001-05-14 10:42:44 +00:00
break;
2009-11-11 13:30:02 +01:00
case DataType::CLOB:
{
Any aValue( getAny() );
Reference< XClob > xClob;
if ( aValue >>= xClob )
{
if ( xClob.is() )
{
aRet = xClob->getSubString(1,(sal_Int32)xClob->length() );
}
}
}
break;
default:
{
Any aValue = getAny();
aValue >>= aRet;
break;
}
2001-05-14 10:42:44 +00:00
}
2001-03-28 10:26:47 +00:00
}
return aRet;
}
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bool ORowSetValue::getBool() const
2001-03-28 10:26:47 +00:00
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getBool" );
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bool bRet = sal_False;
2001-05-14 10:42:44 +00:00
if(!m_bNull)
2001-03-28 10:26:47 +00:00
{
2001-05-14 10:42:44 +00:00
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::LONGVARCHAR:
{
const OUString sValue(m_aValue.m_pString);
const static OUString s_sTrue("true");
const static OUString s_sFalse("false");
if ( sValue.equalsIgnoreAsciiCase(s_sTrue) || (sValue == "1") )
{
bRet = true;
break;
}
else if ( sValue.equalsIgnoreAsciiCase(s_sFalse) || (sValue == "0") )
{
bRet = false;
break;
}
}
// run through
2001-05-14 10:42:44 +00:00
case DataType::DECIMAL:
case DataType::NUMERIC:
bRet = OUString(m_aValue.m_pString).toInt32() != 0;
2001-05-14 10:42:44 +00:00
break;
case DataType::FLOAT:
bRet = m_aValue.m_nFloat != 0.0;
2001-07-24 12:18:40 +00:00
break;
case DataType::DOUBLE:
2001-05-14 10:42:44 +00:00
case DataType::REAL:
bRet = m_aValue.m_nDouble != 0.0;
2001-05-14 10:42:44 +00:00
break;
case DataType::DATE:
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
2001-10-08 03:53:53 +00:00
OSL_ASSERT(!"getBool() for this type is not allowed!");
2001-05-14 10:42:44 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-05-14 10:42:44 +00:00
bRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bRet = m_bSigned ? (m_aValue.m_nInt8 != 0) : (m_aValue.m_uInt8 != 0);
2001-05-14 10:42:44 +00:00
break;
case DataType::SMALLINT:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bRet = m_bSigned ? (m_aValue.m_nInt16 != 0) : (m_aValue.m_uInt16 != 0);
2001-05-14 10:42:44 +00:00
break;
case DataType::INTEGER:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bRet = m_bSigned ? (m_aValue.m_nInt32 != 0) : (m_aValue.m_uInt32 != 0);
break;
case DataType::BIGINT:
bRet = m_bSigned ? (m_aValue.m_nInt64 != 0) : (m_aValue.m_uInt64 != 0);
2001-05-14 10:42:44 +00:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aValue = getAny();
aValue >>= bRet;
break;
}
2001-05-14 10:42:44 +00:00
}
2001-03-28 10:26:47 +00:00
}
return bRet;
}
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
2001-03-28 10:26:47 +00:00
sal_Int8 ORowSetValue::getInt8() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getInt8" );
2001-07-24 12:18:40 +00:00
2001-03-28 10:26:47 +00:00
sal_Int8 nRet = 0;
2001-05-14 10:42:44 +00:00
if(!m_bNull)
2001-03-28 10:26:47 +00:00
{
2001-05-14 10:42:44 +00:00
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = sal_Int8(OUString(m_aValue.m_pString).toInt32());
2001-05-14 10:42:44 +00:00
break;
case DataType::FLOAT:
nRet = sal_Int8(m_aValue.m_nFloat);
2001-07-24 12:18:40 +00:00
break;
case DataType::DOUBLE:
2001-05-14 10:42:44 +00:00
case DataType::REAL:
nRet = sal_Int8(m_aValue.m_nDouble);
2001-05-14 10:42:44 +00:00
break;
case DataType::DATE:
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
2009-11-11 13:30:02 +01:00
case DataType::BLOB:
case DataType::CLOB:
2001-10-08 03:53:53 +00:00
OSL_ASSERT(!"getInt8() for this type is not allowed!");
2001-05-14 10:42:44 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-05-14 10:42:44 +00:00
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = static_cast<sal_Int8>(m_aValue.m_uInt8);
2001-05-14 10:42:44 +00:00
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = static_cast<sal_Int8>(m_aValue.m_nInt16);
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = static_cast<sal_Int8>(m_aValue.m_uInt16);
2001-05-14 10:42:44 +00:00
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = static_cast<sal_Int8>(m_aValue.m_nInt32);
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = static_cast<sal_Int8>(m_aValue.m_uInt32);
break;
case DataType::BIGINT:
if ( m_bSigned )
nRet = static_cast<sal_Int8>(m_aValue.m_nInt64);
else
nRet = static_cast<sal_Int8>(m_aValue.m_uInt64);
break;
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
}
}
return nRet;
}
// -------------------------------------------------------------------------
sal_uInt8 ORowSetValue::getUInt8() const
{
sal_uInt8 nRet = 0;
if(!m_bNull)
{
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = sal_uInt8(OUString(m_aValue.m_pString).toInt32());
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::FLOAT:
nRet = sal_uInt8(m_aValue.m_nFloat);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::DOUBLE:
case DataType::REAL:
nRet = sal_uInt8(m_aValue.m_nDouble);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::DATE:
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
case DataType::BLOB:
case DataType::CLOB:
OSL_ASSERT(!"getuInt8() for this type is not allowed!");
break;
case DataType::BIT:
case DataType::BOOLEAN:
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
nRet = m_aValue.m_uInt8;
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = static_cast<sal_uInt8>(m_aValue.m_nInt16);
else
nRet = static_cast<sal_uInt8>(m_aValue.m_uInt16);
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = static_cast<sal_uInt8>(m_aValue.m_nInt32);
else
nRet = static_cast<sal_uInt8>(m_aValue.m_uInt32);
break;
case DataType::BIGINT:
if ( m_bSigned )
nRet = static_cast<sal_uInt8>(m_aValue.m_nInt64);
else
nRet = static_cast<sal_uInt8>(m_aValue.m_uInt64);
2001-05-14 10:42:44 +00:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
2001-05-14 10:42:44 +00:00
}
2001-03-28 10:26:47 +00:00
}
return nRet;
}
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
2001-03-28 10:26:47 +00:00
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
2001-03-28 10:26:47 +00:00
sal_Int16 ORowSetValue::getInt16() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getInt16" );
2001-07-24 12:18:40 +00:00
2001-03-28 10:26:47 +00:00
sal_Int16 nRet = 0;
2001-05-14 10:42:44 +00:00
if(!m_bNull)
2001-03-28 10:26:47 +00:00
{
2001-05-14 10:42:44 +00:00
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = sal_Int16(OUString(m_aValue.m_pString).toInt32());
2001-05-14 10:42:44 +00:00
break;
case DataType::FLOAT:
nRet = sal_Int16(m_aValue.m_nFloat);
2001-07-24 12:18:40 +00:00
break;
case DataType::DOUBLE:
2001-05-14 10:42:44 +00:00
case DataType::REAL:
nRet = sal_Int16(m_aValue.m_nDouble);
2001-05-14 10:42:44 +00:00
break;
case DataType::DATE:
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
2009-11-11 13:30:02 +01:00
case DataType::BLOB:
case DataType::CLOB:
2001-10-08 03:53:53 +00:00
OSL_ASSERT(!"getInt16() for this type is not allowed!");
2001-05-14 10:42:44 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-05-14 10:42:44 +00:00
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt8;
2001-05-14 10:42:44 +00:00
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt16;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = static_cast<sal_Int16>(m_aValue.m_uInt16);
2001-05-14 10:42:44 +00:00
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = static_cast<sal_Int16>(m_aValue.m_nInt32);
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = static_cast<sal_Int16>(m_aValue.m_uInt32);
break;
case DataType::BIGINT:
if ( m_bSigned )
nRet = static_cast<sal_Int16>(m_aValue.m_nInt64);
else
nRet = static_cast<sal_Int16>(m_aValue.m_uInt64);
2001-05-14 10:42:44 +00:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
2001-05-14 10:42:44 +00:00
}
2001-03-28 10:26:47 +00:00
}
return nRet;
}
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
sal_uInt16 ORowSetValue::getUInt16() const
2001-03-28 10:26:47 +00:00
{
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
sal_uInt16 nRet = 0;
2001-05-14 10:42:44 +00:00
if(!m_bNull)
2001-03-28 10:26:47 +00:00
{
2001-05-14 10:42:44 +00:00
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = sal_uInt16(OUString(m_aValue.m_pString).toInt32());
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::FLOAT:
nRet = sal_uInt16(m_aValue.m_nFloat);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::DOUBLE:
case DataType::REAL:
nRet = sal_uInt16(m_aValue.m_nDouble);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::DATE:
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
case DataType::BLOB:
case DataType::CLOB:
OSL_ASSERT(!"getuInt16() for this type is not allowed!");
break;
case DataType::BIT:
case DataType::BOOLEAN:
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
nRet = m_aValue.m_uInt8;
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt16;
else
nRet = m_aValue.m_uInt16;
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = static_cast<sal_uInt16>(m_aValue.m_nInt32);
else
nRet = static_cast<sal_uInt16>(m_aValue.m_uInt32);
2001-05-14 10:42:44 +00:00
break;
2001-09-20 11:51:56 +00:00
case DataType::BIGINT:
if ( m_bSigned )
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = static_cast<sal_uInt16>(m_aValue.m_nInt64);
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = static_cast<sal_uInt16>(m_aValue.m_uInt64);
break;
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
}
}
return nRet;
}
// -------------------------------------------------------------------------
sal_Int32 ORowSetValue::getInt32() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getInt32" );
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
sal_Int32 nRet = 0;
if(!m_bNull)
{
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = OUString(m_aValue.m_pString).toInt32();
2001-09-20 11:51:56 +00:00
break;
2001-05-14 10:42:44 +00:00
case DataType::FLOAT:
nRet = sal_Int32(m_aValue.m_nFloat);
2001-07-24 12:18:40 +00:00
break;
case DataType::DOUBLE:
2001-05-14 10:42:44 +00:00
case DataType::REAL:
nRet = sal_Int32(m_aValue.m_nDouble);
2001-05-14 10:42:44 +00:00
break;
case DataType::DATE:
nRet = dbtools::DBTypeConversion::toDays(*(::com::sun::star::util::Date*)m_aValue.m_pValue);
break;
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
2009-11-11 13:30:02 +01:00
case DataType::BLOB:
case DataType::CLOB:
2001-10-08 03:53:53 +00:00
OSL_ASSERT(!"getInt32() for this type is not allowed!");
2001-05-14 10:42:44 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-05-14 10:42:44 +00:00
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt8;
2001-05-14 10:42:44 +00:00
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt16;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt16;
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = m_aValue.m_nInt32;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
else
nRet = static_cast<sal_Int32>(m_aValue.m_uInt32);
break;
case DataType::BIGINT:
if ( m_bSigned )
nRet = static_cast<sal_Int32>(m_aValue.m_nInt64);
else
nRet = static_cast<sal_Int32>(m_aValue.m_uInt64);
break;
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
}
}
return nRet;
}
// -------------------------------------------------------------------------
sal_uInt32 ORowSetValue::getUInt32() const
{
sal_uInt32 nRet = 0;
if(!m_bNull)
{
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = OUString(m_aValue.m_pString).toUInt32();
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::FLOAT:
nRet = sal_uInt32(m_aValue.m_nFloat);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::DOUBLE:
case DataType::REAL:
nRet = sal_uInt32(m_aValue.m_nDouble);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::DATE:
nRet = dbtools::DBTypeConversion::toDays(*(::com::sun::star::util::Date*)m_aValue.m_pValue);
break;
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
case DataType::BLOB:
case DataType::CLOB:
OSL_ASSERT(!"getuInt32() for this type is not allowed!");
break;
case DataType::BIT:
case DataType::BOOLEAN:
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
nRet = m_aValue.m_uInt8;
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt16;
else
nRet = m_aValue.m_uInt16;
2001-05-14 10:42:44 +00:00
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = m_aValue.m_nInt32;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt32;
break;
case DataType::BIGINT:
if ( m_bSigned )
nRet = static_cast<sal_uInt32>(m_aValue.m_nInt64);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
else
nRet = static_cast<sal_uInt32>(m_aValue.m_uInt64);
2001-05-14 10:42:44 +00:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
2001-05-14 10:42:44 +00:00
}
2001-03-28 10:26:47 +00:00
}
return nRet;
}
// -------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
2001-09-20 11:51:56 +00:00
sal_Int64 ORowSetValue::getLong() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getLong" );
2001-09-20 11:51:56 +00:00
sal_Int64 nRet = 0;
if(!m_bNull)
{
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = OUString(m_aValue.m_pString).toInt64();
2001-09-20 11:51:56 +00:00
break;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
case DataType::FLOAT:
nRet = sal_Int64(m_aValue.m_nFloat);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::DOUBLE:
case DataType::REAL:
nRet = sal_Int64(m_aValue.m_nDouble);
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
case DataType::DATE:
nRet = dbtools::DBTypeConversion::toDays(*(::com::sun::star::util::Date*)m_aValue.m_pValue);
break;
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
case DataType::BLOB:
case DataType::CLOB:
OSL_ASSERT(!"getLong() for this type is not allowed!");
break;
case DataType::BIT:
case DataType::BOOLEAN:
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
nRet = m_aValue.m_uInt8;
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt16;
else
nRet = m_aValue.m_uInt16;
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = m_aValue.m_nInt32;
else
nRet = m_aValue.m_uInt32;
break;
2001-09-20 11:51:56 +00:00
case DataType::BIGINT:
if ( m_bSigned )
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_nInt64;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = static_cast<sal_Int64>(m_aValue.m_uInt64);
break;
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
}
}
return nRet;
}
// -------------------------------------------------------------------------
sal_uInt64 ORowSetValue::getULong() const
{
sal_uInt64 nRet = 0;
if(!m_bNull)
{
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = static_cast<sal_uInt64>(OUString(m_aValue.m_pString).toUInt64());
2001-09-20 11:51:56 +00:00
break;
case DataType::FLOAT:
nRet = sal_uInt64(m_aValue.m_nFloat);
2001-09-20 11:51:56 +00:00
break;
case DataType::DOUBLE:
case DataType::REAL:
nRet = sal_uInt64(m_aValue.m_nDouble);
2001-09-20 11:51:56 +00:00
break;
case DataType::DATE:
nRet = dbtools::DBTypeConversion::toDays(*(::com::sun::star::util::Date*)m_aValue.m_pValue);
break;
case DataType::TIME:
case DataType::TIMESTAMP:
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
2009-11-11 13:30:02 +01:00
case DataType::BLOB:
case DataType::CLOB:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
OSL_ASSERT(!"getULong() for this type is not allowed!");
2001-09-20 11:51:56 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-09-20 11:51:56 +00:00
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
nRet = m_aValue.m_uInt8;
2001-09-20 11:51:56 +00:00
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt16;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt16;
2001-09-20 11:51:56 +00:00
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = m_aValue.m_nInt32;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt32;
break;
case DataType::BIGINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt64;
else
nRet = m_aValue.m_uInt64;
2001-09-20 11:51:56 +00:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
2001-09-20 11:51:56 +00:00
}
}
return nRet;
}
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
2001-09-20 11:51:56 +00:00
// -------------------------------------------------------------------------
2001-07-24 12:18:40 +00:00
float ORowSetValue::getFloat() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getFloat" );
2001-07-24 12:18:40 +00:00
float nRet = 0;
if(!m_bNull)
{
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = OUString(m_aValue.m_pString).toFloat();
2001-07-24 12:18:40 +00:00
break;
case DataType::FLOAT:
nRet = m_aValue.m_nFloat;
2001-07-24 12:18:40 +00:00
break;
case DataType::DOUBLE:
case DataType::REAL:
nRet = (float)m_aValue.m_nDouble;
2001-07-24 12:18:40 +00:00
break;
case DataType::DATE:
nRet = (float)dbtools::DBTypeConversion::toDouble(*(::com::sun::star::util::Date*)m_aValue.m_pValue);
break;
case DataType::TIME:
nRet = (float)dbtools::DBTypeConversion::toDouble(*(::com::sun::star::util::Time*)m_aValue.m_pValue);
break;
case DataType::TIMESTAMP:
nRet = (float)dbtools::DBTypeConversion::toDouble(*(::com::sun::star::util::DateTime*)m_aValue.m_pValue);
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
2009-11-11 13:30:02 +01:00
case DataType::BLOB:
case DataType::CLOB:
2001-10-08 03:53:53 +00:00
OSL_ASSERT(!"getDouble() for this type is not allowed!");
2001-07-24 12:18:40 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-07-24 12:18:40 +00:00
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt8;
2001-07-24 12:18:40 +00:00
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt16;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = (float)m_aValue.m_uInt16;
2001-07-24 12:18:40 +00:00
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = (float)m_aValue.m_nInt32;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = (float)m_aValue.m_uInt32;
break;
case DataType::BIGINT:
if ( m_bSigned )
nRet = (float)m_aValue.m_nInt64;
else
nRet = (float)m_aValue.m_uInt64;
2001-07-24 12:18:40 +00:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
2001-07-24 12:18:40 +00:00
}
}
return nRet;
}
// -------------------------------------------------------------------------
2001-03-28 10:26:47 +00:00
double ORowSetValue::getDouble() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getDouble" );
2001-07-24 12:18:40 +00:00
2001-03-28 10:26:47 +00:00
double nRet = 0;
2001-05-14 10:42:44 +00:00
if(!m_bNull)
2001-03-28 10:26:47 +00:00
{
2001-05-14 10:42:44 +00:00
switch(getTypeKind())
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
nRet = OUString(m_aValue.m_pString).toDouble();
2001-05-14 10:42:44 +00:00
break;
case DataType::FLOAT:
nRet = m_aValue.m_nFloat;
2001-07-24 12:18:40 +00:00
break;
case DataType::DOUBLE:
2001-05-14 10:42:44 +00:00
case DataType::REAL:
nRet = m_aValue.m_nDouble;
2001-05-14 10:42:44 +00:00
break;
case DataType::DATE:
nRet = dbtools::DBTypeConversion::toDouble(*(::com::sun::star::util::Date*)m_aValue.m_pValue);
break;
case DataType::TIME:
nRet = dbtools::DBTypeConversion::toDouble(*(::com::sun::star::util::Time*)m_aValue.m_pValue);
break;
case DataType::TIMESTAMP:
nRet = dbtools::DBTypeConversion::toDouble(*(::com::sun::star::util::DateTime*)m_aValue.m_pValue);
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
2009-11-11 13:30:02 +01:00
case DataType::BLOB:
case DataType::CLOB:
2001-10-08 03:53:53 +00:00
OSL_ASSERT(!"getDouble() for this type is not allowed!");
2001-05-14 10:42:44 +00:00
break;
case DataType::BIT:
case DataType::BOOLEAN:
2001-05-14 10:42:44 +00:00
nRet = m_aValue.m_bBool;
break;
case DataType::TINYINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt8;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt8;
2001-05-14 10:42:44 +00:00
break;
case DataType::SMALLINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt16;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt16;
2001-05-14 10:42:44 +00:00
break;
case DataType::INTEGER:
if ( m_bSigned )
nRet = m_aValue.m_nInt32;
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
nRet = m_aValue.m_uInt32;
break;
case DataType::BIGINT:
if ( m_bSigned )
nRet = m_aValue.m_nInt64;
else
nRet = m_aValue.m_uInt64;
2001-05-14 10:42:44 +00:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aValue = getAny();
aValue >>= nRet;
break;
}
2001-05-14 10:42:44 +00:00
}
2001-03-28 10:26:47 +00:00
}
return nRet;
}
2001-07-24 12:18:40 +00:00
// -----------------------------------------------------------------------------
Sequence<sal_Int8> ORowSetValue::getSequence() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getSequence" );
2001-07-24 12:18:40 +00:00
Sequence<sal_Int8> aSeq;
if (!m_bNull)
{
switch(m_eTypeKind)
{
case DataType::OBJECT:
case DataType::CLOB:
case DataType::BLOB:
{
Reference<XInputStream> xStream;
const Any aValue = makeAny();
2001-07-24 12:18:40 +00:00
if(aValue.hasValue())
{
Reference<XBlob> xBlob(aValue,UNO_QUERY);
if ( xBlob.is() )
xStream = xBlob->getBinaryStream();
else
{
Reference<XClob> xClob(aValue,UNO_QUERY);
if ( xClob.is() )
xStream = xClob->getCharacterStream();
}
2001-07-24 12:18:40 +00:00
if(xStream.is())
{
const sal_uInt32 nBytesToRead = 65535;
sal_uInt32 nRead;
do
{
::com::sun::star::uno::Sequence< sal_Int8 > aReadSeq;
nRead = xStream->readSomeBytes( aReadSeq, nBytesToRead );
if( nRead )
{
const sal_uInt32 nOldLength = aSeq.getLength();
aSeq.realloc( nOldLength + nRead );
memcpy( aSeq.getArray() + nOldLength, aReadSeq.getConstArray(), aReadSeq.getLength() );
}
}
while( nBytesToRead == nRead );
xStream->closeInput();
}
2001-07-24 12:18:40 +00:00
}
}
break;
case DataType::VARCHAR:
case DataType::LONGVARCHAR:
2001-07-24 12:18:40 +00:00
{
OUString sVal(m_aValue.m_pString);
2001-07-24 12:18:40 +00:00
aSeq = Sequence<sal_Int8>(reinterpret_cast<const sal_Int8*>(sVal.getStr()),sizeof(sal_Unicode)*sVal.getLength());
}
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
aSeq = *static_cast< Sequence<sal_Int8>*>(m_aValue.m_pValue);
2001-07-24 12:18:40 +00:00
break;
default:
{
Any aValue = getAny();
aValue >>= aSeq;
break;
}
2001-07-24 12:18:40 +00:00
}
}
return aSeq;
}
// -----------------------------------------------------------------------------
2010-09-27 11:37:17 +02:00
::com::sun::star::util::Date ORowSetValue::getDate() const
2001-07-24 12:18:40 +00:00
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getDate" );
2001-10-01 10:24:29 +00:00
::com::sun::star::util::Date aValue;
if(!m_bNull)
2001-07-24 12:18:40 +00:00
{
2001-10-01 10:24:29 +00:00
switch(m_eTypeKind)
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::LONGVARCHAR:
2001-10-01 10:24:29 +00:00
aValue = DBTypeConversion::toDate(getString());
break;
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::FLOAT:
case DataType::DOUBLE:
case DataType::REAL:
aValue = DBTypeConversion::toDate((double)*this);
break;
case DataType::DATE:
aValue = *static_cast< ::com::sun::star::util::Date*>(m_aValue.m_pValue);
break;
case DataType::TIMESTAMP:
{
::com::sun::star::util::DateTime* pDateTime = static_cast< ::com::sun::star::util::DateTime*>(m_aValue.m_pValue);
aValue.Day = pDateTime->Day;
aValue.Month = pDateTime->Month;
aValue.Year = pDateTime->Year;
}
break;
2010-09-27 11:37:17 +02:00
case DataType::BIT:
case DataType::BOOLEAN:
case DataType::TINYINT:
case DataType::SMALLINT:
case DataType::INTEGER:
case DataType::BIGINT:
aValue = DBTypeConversion::toDate( double( sal_Int64( *this ) ) );
break;
case DataType::BLOB:
case DataType::CLOB:
case DataType::OBJECT:
2009-11-11 13:30:02 +01:00
default:
2010-09-27 11:37:17 +02:00
OSL_ENSURE( false, "ORowSetValue::getDate: cannot retrieve the data!" );
// NO break!
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
case DataType::TIME:
aValue = DBTypeConversion::toDate( (double)0 );
break;
2001-10-01 10:24:29 +00:00
}
2001-07-24 12:18:40 +00:00
}
2001-10-01 10:24:29 +00:00
return aValue;
2001-07-24 12:18:40 +00:00
}
// -----------------------------------------------------------------------------
::com::sun::star::util::Time ORowSetValue::getTime() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getTime" );
2001-10-01 10:24:29 +00:00
::com::sun::star::util::Time aValue;
if(!m_bNull)
2001-07-24 12:18:40 +00:00
{
2001-10-01 10:24:29 +00:00
switch(m_eTypeKind)
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::LONGVARCHAR:
2001-10-01 10:24:29 +00:00
aValue = DBTypeConversion::toTime(getString());
break;
case DataType::DECIMAL:
case DataType::NUMERIC:
aValue = DBTypeConversion::toTime((double)*this);
break;
case DataType::FLOAT:
case DataType::DOUBLE:
case DataType::REAL:
aValue = DBTypeConversion::toTime((double)*this);
break;
case DataType::TIMESTAMP:
{
::com::sun::star::util::DateTime* pDateTime = static_cast< ::com::sun::star::util::DateTime*>(m_aValue.m_pValue);
aValue.NanoSeconds = pDateTime->NanoSeconds;
aValue.Seconds = pDateTime->Seconds;
aValue.Minutes = pDateTime->Minutes;
aValue.Hours = pDateTime->Hours;
}
break;
2001-10-01 10:24:29 +00:00
case DataType::TIME:
aValue = *static_cast< ::com::sun::star::util::Time*>(m_aValue.m_pValue);
2009-11-12 20:56:06 +01:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aAnyValue = getAny();
aAnyValue >>= aValue;
break;
}
2001-10-01 10:24:29 +00:00
}
2001-07-24 12:18:40 +00:00
}
2001-10-01 10:24:29 +00:00
return aValue;
2001-07-24 12:18:40 +00:00
}
// -----------------------------------------------------------------------------
::com::sun::star::util::DateTime ORowSetValue::getDateTime() const
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::getDateTime" );
2001-10-01 10:24:29 +00:00
::com::sun::star::util::DateTime aValue;
if(!m_bNull)
2001-07-24 12:18:40 +00:00
{
2001-10-01 10:24:29 +00:00
switch(m_eTypeKind)
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::LONGVARCHAR:
2001-10-01 10:24:29 +00:00
aValue = DBTypeConversion::toDateTime(getString());
break;
case DataType::DECIMAL:
case DataType::NUMERIC:
aValue = DBTypeConversion::toDateTime((double)*this);
break;
case DataType::FLOAT:
case DataType::DOUBLE:
case DataType::REAL:
aValue = DBTypeConversion::toDateTime((double)*this);
break;
case DataType::DATE:
{
::com::sun::star::util::Date* pDate = static_cast< ::com::sun::star::util::Date*>(m_aValue.m_pValue);
aValue.Day = pDate->Day;
aValue.Month = pDate->Month;
aValue.Year = pDate->Year;
}
break;
case DataType::TIME:
{
::com::sun::star::util::Time* pTime = static_cast< ::com::sun::star::util::Time*>(m_aValue.m_pValue);
aValue.NanoSeconds = pTime->NanoSeconds;
aValue.Seconds = pTime->Seconds;
aValue.Minutes = pTime->Minutes;
aValue.Hours = pTime->Hours;
}
break;
2001-10-01 10:24:29 +00:00
case DataType::TIMESTAMP:
aValue = *static_cast< ::com::sun::star::util::DateTime*>(m_aValue.m_pValue);
2001-10-01 10:24:29 +00:00
break;
2009-11-11 13:30:02 +01:00
default:
{
Any aAnyValue = getAny();
aAnyValue >>= aValue;
break;
}
2001-10-01 10:24:29 +00:00
}
2001-07-24 12:18:40 +00:00
}
2001-10-01 10:24:29 +00:00
return aValue;
2001-07-24 12:18:40 +00:00
}
// -----------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
void ORowSetValue::setSigned(bool _bMod)
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::setSigned" );
if ( m_bSigned != _bMod )
{
m_bSigned = _bMod;
if ( !m_bNull )
{
sal_Int32 nType = m_eTypeKind;
switch(m_eTypeKind)
{
case DataType::TINYINT:
if ( m_bSigned )
(*this) = getInt8();
else
{
m_bSigned = !m_bSigned;
(*this) = getInt16();
m_bSigned = !m_bSigned;
}
break;
case DataType::SMALLINT:
if ( m_bSigned )
(*this) = getInt16();
else
{
m_bSigned = !m_bSigned;
(*this) = getInt32();
m_bSigned = !m_bSigned;
}
break;
case DataType::INTEGER:
if ( m_bSigned )
(*this) = getInt32();
else
{
m_bSigned = !m_bSigned;
(*this) = getLong();
m_bSigned = !m_bSigned;
}
break;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
case DataType::BIGINT:
if ( m_bSigned )
m_aValue.m_nInt64 = static_cast<sal_Int64>(m_aValue.m_uInt64);
else
m_aValue.m_uInt64 = static_cast<sal_uInt64>(m_aValue.m_nInt64);
break;
}
m_eTypeKind = nType;
}
}
}
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
// -----------------------------------------------------------------------------
namespace detail
{
class SAL_NO_VTABLE IValueSource
{
public:
virtual OUString getString() const = 0;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
virtual bool getBoolean() const = 0;
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
virtual sal_Int8 getByte() const = 0;
virtual sal_Int16 getShort() const = 0;
virtual sal_Int32 getInt() const = 0;
virtual sal_Int64 getLong() const = 0;
virtual float getFloat() const = 0;
virtual double getDouble() const = 0;
virtual Date getDate() const = 0;
virtual Time getTime() const = 0;
virtual DateTime getTimestamp() const = 0;
virtual Sequence< sal_Int8 > getBytes() const = 0;
virtual Reference< XInputStream > getBinaryStream() const = 0;
virtual Reference< XInputStream > getCharacterStream() const = 0;
virtual Reference< XBlob > getBlob() const = 0;
virtual Reference< XClob > getClob() const = 0;
2009-12-10 08:08:38 +01:00
virtual Any getObject() const = 0;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
virtual bool wasNull() const = 0;
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
virtual ~IValueSource() { }
};
class RowValue : public IValueSource
{
public:
RowValue( const Reference< XRow >& _xRow, const sal_Int32 _nPos )
:m_xRow( _xRow )
,m_nPos( _nPos )
{
}
// IValueSource
virtual OUString getString() const { return m_xRow->getString( m_nPos ); };
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
virtual bool getBoolean() const { return m_xRow->getBoolean( m_nPos ); };
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
virtual sal_Int8 getByte() const { return m_xRow->getByte( m_nPos ); };
virtual sal_Int16 getShort() const { return m_xRow->getShort( m_nPos ); }
virtual sal_Int32 getInt() const { return m_xRow->getInt( m_nPos ); }
virtual sal_Int64 getLong() const { return m_xRow->getLong( m_nPos ); }
virtual float getFloat() const { return m_xRow->getFloat( m_nPos ); };
virtual double getDouble() const { return m_xRow->getDouble( m_nPos ); };
virtual Date getDate() const { return m_xRow->getDate( m_nPos ); };
virtual Time getTime() const { return m_xRow->getTime( m_nPos ); };
virtual DateTime getTimestamp() const { return m_xRow->getTimestamp( m_nPos ); };
virtual Sequence< sal_Int8 > getBytes() const { return m_xRow->getBytes( m_nPos ); };
virtual Reference< XInputStream > getBinaryStream() const { return m_xRow->getBinaryStream( m_nPos ); };
virtual Reference< XInputStream > getCharacterStream() const { return m_xRow->getCharacterStream( m_nPos ); };
virtual Reference< XBlob > getBlob() const { return m_xRow->getBlob( m_nPos ); };
virtual Reference< XClob > getClob() const { return m_xRow->getClob( m_nPos ); };
2009-12-10 11:45:59 +01:00
virtual Any getObject() const { return m_xRow->getObject( m_nPos ,NULL); };
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
virtual bool wasNull() const { return m_xRow->wasNull( ); };
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
private:
const Reference< XRow > m_xRow;
const sal_Int32 m_nPos;
};
class ColumnValue : public IValueSource
{
public:
ColumnValue( const Reference< XColumn >& _rxColumn )
:m_xColumn( _rxColumn )
{
}
// IValueSource
virtual OUString getString() const { return m_xColumn->getString(); };
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
virtual bool getBoolean() const { return m_xColumn->getBoolean(); };
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
virtual sal_Int8 getByte() const { return m_xColumn->getByte(); };
virtual sal_Int16 getShort() const { return m_xColumn->getShort(); }
virtual sal_Int32 getInt() const { return m_xColumn->getInt(); }
virtual sal_Int64 getLong() const { return m_xColumn->getLong(); }
virtual float getFloat() const { return m_xColumn->getFloat(); };
virtual double getDouble() const { return m_xColumn->getDouble(); };
virtual Date getDate() const { return m_xColumn->getDate(); };
virtual Time getTime() const { return m_xColumn->getTime(); };
virtual DateTime getTimestamp() const { return m_xColumn->getTimestamp(); };
virtual Sequence< sal_Int8 > getBytes() const { return m_xColumn->getBytes(); };
virtual Reference< XInputStream > getBinaryStream() const { return m_xColumn->getBinaryStream(); };
virtual Reference< XInputStream > getCharacterStream() const { return m_xColumn->getCharacterStream(); };
virtual Reference< XBlob > getBlob() const { return m_xColumn->getBlob(); };
virtual Reference< XClob > getClob() const { return m_xColumn->getClob(); };
2009-12-10 08:08:38 +01:00
virtual Any getObject() const { return m_xColumn->getObject( NULL ); };
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
virtual bool wasNull() const { return m_xColumn->wasNull( ); };
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
private:
const Reference< XColumn > m_xColumn;
};
}
// -----------------------------------------------------------------------------
void ORowSetValue::fill( const sal_Int32 _nType, const Reference< XColumn >& _rxColumn )
{
detail::ColumnValue aColumnValue( _rxColumn );
impl_fill( _nType, sal_True, aColumnValue );
}
// -----------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
void ORowSetValue::fill( sal_Int32 _nPos, sal_Int32 _nType, bool _bNullable, const Reference< XRow>& _xRow )
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
{
detail::RowValue aRowValue( _xRow, _nPos );
impl_fill( _nType, _bNullable, aRowValue );
}
// -----------------------------------------------------------------------------
void ORowSetValue::fill(sal_Int32 _nPos,
sal_Int32 _nType,
const ::com::sun::star::uno::Reference< ::com::sun::star::sdbc::XRow>& _xRow)
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::fill (1)" );
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
fill(_nPos,_nType,sal_True,_xRow);
}
CWS-TOOLING: integrate CWS dba32b 2009-06-03 14:58:08 +0200 fs r272581 : #i102439# 2009-05-29 13:56:18 +0200 fs r272456 : remove the sub form when the 'add subform' setting changes from <true/> to <false/>, not only its controls 2009-05-29 13:19:27 +0200 fs r272454 : display '(Default)' instead of an empty string when a control has the default font 2009-05-28 20:49:18 +0200 fs r272428 : #i98162# getFirstSelectedValue: do not return reference to a temporary 2009-05-27 15:30:22 +0200 msc r272353 : #102303# 2009-05-26 13:03:06 +0200 fs r272295 : spelling 2009-05-26 12:59:54 +0200 fs r272294 : merge m48 version to get latest cygwin related fixes 2009-05-25 14:02:06 +0200 fs r272239 : remove references to local files, needed for debugging sessions only 2009-05-25 14:01:16 +0200 fs r272238 : #i102021# ensure members such as bNumberFormat are initialized before actually returning them 2009-05-25 13:10:20 +0200 fs r272236 : #i10000# reset ENABLE_EVOAB2 2009-05-22 06:44:45 +0200 oj r272167 : #i99104# add import handler for calc-sett 2009-05-22 06:42:27 +0200 oj r272166 : #i99104# impl NullDate as member 2009-05-22 06:36:22 +0200 oj r272165 : #i99104# handle nulldate from parent model 2009-05-22 06:33:13 +0200 oj r272164 : #i99104# export and import calculation-settings and nulldate 2009-05-22 06:27:50 +0200 oj r272163 : #i99104# export and import calculation-settings and nulldate 2009-05-20 13:29:58 +0200 oj r272118 : #i99911# handle invalid name 2009-05-20 13:28:49 +0200 oj r272117 : #i101261# handle different rowsets 2009-05-20 11:29:55 +0200 msc r272111 : #i100000# 2009-05-20 11:28:27 +0200 msc r272110 : merge in change from dba32a 2009-05-20 11:27:38 +0200 msc r272109 : #102082# CTRL + C does not work 2009-05-20 09:43:36 +0200 oj r272106 : clean up includes 2009-05-20 09:32:15 +0200 oj r272105 : #i99060# merge error resovled now VERTICAL_ALIGN is in both stmt 2009-05-20 08:37:21 +0200 msc r272104 : add workaround for issue #102010# 2009-05-20 08:10:12 +0200 oj r272103 : #i99104# use numberformatkey 2009-05-20 08:07:02 +0200 oj r272102 : #i99104# use column info from rowset 2009-05-20 08:04:43 +0200 oj r272101 : #i102032# use a special column type where prec and scale are the values currently set at the column 2009-05-20 08:03:04 +0200 oj r272100 : #i102032# correct type info, we have to use SQL defined type names 2009-05-19 10:27:02 +0200 oj r272061 : #i99104# export null-date 2009-05-19 08:26:53 +0200 oj r272056 : #i99104# export null-date 2009-05-18 13:15:10 +0200 msc r272014 : add issue #102019# 2009-05-18 11:33:07 +0200 msc r272005 : add issue #102019# 2009-05-18 08:59:45 +0200 msc r271996 : add workaroud for issue #102010# 2009-05-15 10:21:24 +0200 msc r271929 : #101944# 2009-05-11 21:18:30 +0200 fs r271792 : #i99914# 2009-05-08 13:52:06 +0200 oj r271715 : #i96423# remember column span 2009-05-08 11:26:19 +0200 oj r271708 : #i98605# impl new scale mode 2009-05-08 10:33:35 +0200 fs r271706 : SendUserCall: only call into the shape notification routine for UserCall types where this is necessary (performance issue) 2009-05-07 20:52:44 +0200 fs r271698 : outsource ShapeProperty from shapepropertynotifier.hxx 2009-05-07 20:43:33 +0200 fs r271697 : #i99056# use notifyShapePropertyChange, instead of getShapePropertyChangeNotifier - the latter throws if no shape exists, yet 2009-05-07 20:33:58 +0200 fs r271696 : #i99056# +notifyShapePropertyChange: allow notifying chages without checking whether there actually already exists an SvxShape 2009-05-07 16:22:15 +0200 fs r271679 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:21:37 +0200 fs r271678 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:01:11 +0200 oj r271677 : #i99914# notify parent handler 2009-05-07 15:54:54 +0200 fs r271676 : #i10000# cygwin needs some quoting 2009-05-07 14:49:48 +0200 oj r271672 : #i99277# quote alias name 2009-05-07 14:48:12 +0200 oj r271671 : #i92538# add ~ in front of type 2009-05-07 14:37:13 +0200 oj r271667 : #i99118# change type from char to varchar 2009-05-07 14:36:23 +0200 oj r271666 : #i99118# clear dest columns when changing to create new table 2009-05-07 13:35:32 +0200 oj r271657 : #i94467# handle type 0 as double as well 2009-05-07 13:20:49 +0200 oj r271655 : i99743# setNull when varchar is no text 2009-05-07 12:58:06 +0200 fs r271651 : initialize the SdrObject's property change notifier after the ctor, if necessary 2009-05-07 11:47:18 +0200 fs r271647 : #i10000# 2009-05-07 10:57:16 +0200 fs r271639 : OPropertyBrowserController::propertyChange: care for the current property state, too, and properly forward it to the UI 2009-05-07 10:18:14 +0200 fs r271636 : onNewComponent: do not ask the map for grid columns, it will throw 2009-05-07 10:09:55 +0200 fs r271634 : #i101623# 2009-05-07 09:53:44 +0200 fs r271631 : #i101622# 2009-05-06 21:55:53 +0200 fs r271615 : #i10000# 2009-05-06 21:10:42 +0200 fs r271611 : #i10000# 2009-05-06 13:11:48 +0200 fs r271583 : #i10000# 2009-05-05 22:29:31 +0200 fs r271559 : proper assertion message 2009-05-05 22:29:03 +0200 fs r271558 : diagnostics 2009-05-05 22:16:16 +0200 fs r271557 : #i10000# 2009-05-05 13:50:32 +0200 fs r271513 : #i10000# 2009-05-05 10:21:50 +0200 fs r271503 : #i10000# 2009-05-05 09:30:26 +0200 fs r271501 : why did those survive the rebase -C step? 2009-05-05 09:18:12 +0200 fs r271500 : #i10000# 2009-05-04 17:08:17 +0200 fs r271475 : CWS-TOOLING: rebase CWS dba32b to trunk@271427 (milestone: DEV300:m47) 2009-05-04 14:51:26 +0200 fs r271456 : line ends 2009-04-30 15:55:27 +0200 fs r271418 : NewURL -> PublicConnectionURL 2009-04-22 21:18:34 +0200 fs r271141 : #i100944# 2009-04-22 09:12:26 +0200 oj r271071 : #i101261# little code change 2009-04-22 09:11:43 +0200 oj r271070 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:11:25 +0200 oj r271069 : #i101261# new grabage container for nodes 2009-04-22 09:11:02 +0200 oj r271068 : #i101261# new grabage container for nodes 2009-04-22 09:10:44 +0200 oj r271067 : #i101261# new grabage container for nodes 2009-04-22 09:10:21 +0200 oj r271066 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:08:24 +0200 oj r271065 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:07:25 +0200 oj r271064 : #i101261# only ask for parameters which aren't set before 2009-04-22 08:49:07 +0200 oj r271062 : #i77501# preview only when needed 2009-04-22 08:45:44 +0200 oj r271061 : #i101261# new prop max rows 2009-04-22 08:44:18 +0200 oj r271060 : #i101261# create dataprovider earlier to avoid the wrong legend in chart 2009-04-22 08:42:48 +0200 oj r271059 : #i101261# handle parameter 2009-04-17 21:00:23 +0200 fs r270954 : #i98350# 2009-04-17 13:54:19 +0200 fs r270942 : #i99565# 2009-04-17 13:51:34 +0200 fs r270940 : #i101153# only localize the (potentially) localizable properties when there really is support at the control model 2009-04-17 11:43:14 +0200 fs r270932 : removed superfluous include 2009-04-17 10:10:15 +0200 fs r270926 : #i10000# 2009-04-17 10:02:36 +0200 fs r270925 : #i10000# 2009-04-17 09:15:13 +0200 fs r270918 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:14:56 +0200 fs r270917 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:13:25 +0200 fs r270915 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 08:30:34 +0200 fs r270914 : removed unotools/servicehelper.hxx in favour of the (duplicated) comphelper/servicehelper.hxx 2009-04-16 21:05:25 +0200 fs r270903 : #i10000# 2009-04-16 20:43:43 +0200 fs r270902 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-16 15:57:36 +0200 fs r270891 : #i99056# make SetUnoControlModel virtual 2009-04-16 15:44:02 +0200 fs r270890 : #i99056# let the ScShapeObj notify changes in its Anchor property 2009-04-16 15:36:47 +0200 fs r270889 : #i99056# enable the sheet anchor type property, too 2009-04-16 15:33:45 +0200 fs r270887 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-15 14:53:13 +0200 fs r270844 : #i10000# 2009-04-15 13:08:29 +0200 fs r270836 : #i10000# 2009-04-15 12:28:14 +0200 fs r270832 : #i10000# 2009-04-15 10:59:14 +0200 fs r270827 : #i10000# 2009-04-15 09:41:08 +0200 oj r270823 : fix issues found with findbugs and pmd 2009-04-14 21:08:04 +0200 fs r270808 : #i99056# implement SheetAnchorType - now the only thing missing to enable it is the proper notification when it is modified 2009-04-14 17:09:00 +0200 fs r270799 : #i99056# implement XServiceInfo for the ScShapeObj 2009-04-14 17:07:55 +0200 fs r270798 : #i99056# implement TextAnchorType, partially implement SheetAnchorType 2009-04-14 15:54:05 +0200 fs r270786 : #i99056# SwXShape: notify changes of the AnchorType property 2009-04-14 15:47:32 +0200 fs r270785 : #i99056# deliver shapepropertynotifier.hxx 2009-04-14 15:46:54 +0200 fs r270784 : diagnostics 2009-04-14 15:08:28 +0200 fs r270781 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:41:09 +0200 fs r270773 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:37:23 +0200 fs r270772 : in dtor, remove the properties from the temporary component context 2009-04-14 14:36:34 +0200 fs r270771 : getWeakUnoShape made const 2009-04-14 12:23:08 +0200 oj r270757 : #i101064# add missing braces 2009-04-14 12:21:25 +0200 oj r270756 : #i101065# add braces for gcc 4.3.2 2009-04-14 12:17:45 +0200 oj r270755 : #i101059# add dep for manifest 2009-04-09 12:06:58 +0200 oj r270686 : #i93100# use OptimalSize from control to get height 2009-04-08 09:56:55 +0200 oj r270619 : #i92537# handle calculations in the select columns as well 2009-04-08 09:30:39 +0200 oj r270615 : #i96657# throw error message when the key doesn't have any columns 2009-04-07 12:08:26 +0200 oj r270592 : #i77501# impl preview of the executed report 2009-04-07 12:01:56 +0200 oj r270591 : #i77501# impl preview of the executed report 2009-04-07 11:41:03 +0200 oj r270590 : #i77501# impl preview of the executed report 2009-04-07 11:39:32 +0200 oj r270589 : #i77501# impl preview of the executed report 2009-04-07 11:29:25 +0200 oj r270588 : #i77501# convert dos to unix lineends 2009-04-07 11:28:23 +0200 oj r270587 : #i77501# impl preview of the executed report 2009-04-07 11:16:50 +0200 oj r270586 : #i77501# impl preview of the executed report 2009-04-07 11:16:00 +0200 oj r270585 : #i77501# impl preview of the executed report 2009-04-07 11:15:44 +0200 oj r270584 : #i77501# impl preview of the executed report 2009-04-07 11:15:28 +0200 oj r270583 : #i77501# impl preview of the executed report 2009-04-07 11:15:04 +0200 oj r270582 : #i77501# impl preview of the executed report 2009-04-06 15:38:54 +0200 fs r270559 : merge changes from CWS dba32a herein 2009-04-03 15:56:16 +0200 fs r270494 : ImpSvMEdit::Resize: do multiple iterations, if necessary 2009-04-03 14:35:49 +0200 fs r270487 : #i10000# 2009-04-03 13:17:16 +0200 fs r270476 : #i99056# display geometry information for controls, too 2009-04-03 13:16:37 +0200 fs r270475 : #i99056# better ordering of the geometry properties 2009-04-03 13:16:07 +0200 fs r270473 : #i99056# now that SvxShape supports property change listeners, forward add/remove requests to it 2009-04-03 13:13:18 +0200 fs r270472 : #i99056# at SvxShape, allow for PropertyChangeListeners for Size/Position 2009-04-03 09:29:27 +0200 oj r270456 : #i94571# use correct prop name 2009-04-03 09:14:54 +0200 fs r270451 : merge changes from CWS dba32a herein 2009-04-02 17:00:51 +0200 fs r270424 : better diagnostics 2009-04-02 16:35:19 +0200 fs r270421 : diagnostics 2009-04-02 16:34:50 +0200 fs r270420 : #i99056# mxUnoShape not accessible anymore, use impl_setUnoShape instead 2009-04-02 16:32:48 +0200 fs r270419 : #i99056# make getUnoShape cheaper: keep the pointer to the SvxShape all the time, so there's no need to ask for it in getUnoShape. As a consequence, we will later be able to use the pointer in scenarious where performance (potentially) matters 2009-04-02 16:31:13 +0200 fs r270417 : merge changes from CWS dba32a herein 2009-04-02 16:23:16 +0200 fs r270414 : merge changes from CWS dba32a herein 2009-04-02 14:10:35 +0200 fs r270405 : #i10000# 2009-04-02 14:06:26 +0200 fs r270404 : merge changes from CWS dba32a herein 2009-04-02 14:03:03 +0200 fs r270401 : #i10000# 2009-04-02 13:58:13 +0200 fs r270400 : #i10000# 2009-04-02 12:59:44 +0200 fs r270397 : merge changes from CWS dba32a herein 2009-04-02 12:46:30 +0200 fs r270396 : #i99056# let the form page maintain a mapping between control models and control shapes 2009-04-02 12:44:07 +0200 fs r270395 : merge changes from CWS dba32a herein 2009-04-02 12:42:06 +0200 fs r270394 : merge changes from CWS dba32a herein 2009-04-02 12:35:20 +0200 fs r270393 : #i10000# precompiled header 2009-04-02 12:05:31 +0200 fs r270392 : merge changes from CWS dba32a herein 2009-04-02 12:00:42 +0200 fs r270391 : merge changes from CWS dba32a herein 2009-04-02 11:47:26 +0200 fs r270390 : merge changes from CWS dba32a herein 2009-04-02 11:39:15 +0200 oj r270389 : #i94467# foxpro impl several new types 2009-04-02 11:35:58 +0200 fs r270387 : merge changes from CWS dba32a herein 2009-04-01 14:10:51 +0200 fs r270329 : merge changes from CWS dba32a herein 2009-03-31 17:29:50 +0200 fs r270290 : merge changes from CWS dba32a herein 2009-03-30 14:53:56 +0200 fs r270233 : #i100417# don't set grid column widths to 0, but to <void/> 2009-03-30 12:31:03 +0200 oj r270213 : #i100552# wrong orb used 2009-03-30 12:19:20 +0200 oj r270212 : #i98303# convertlike corrected to sal_Unicode 2009-03-30 11:58:25 +0200 fs r270210 : merge changes from CWS dba32a herein 2009-03-30 11:38:16 +0200 oj r270205 : remove duplicate code from merge 2009-03-30 11:02:27 +0200 fs r270202 : merge changes from CWS dba32a herein 2009-03-30 11:02:19 +0200 fs r270201 : merge changes from CWS dba32a herein 2009-03-30 10:31:26 +0200 oj r270200 : #i100665# only throw exception and do not drop table 2009-03-30 09:36:24 +0200 fs r270195 : assertion text 2009-03-28 20:21:58 +0100 fs r270187 : #ii10000# 2009-03-28 20:19:54 +0100 fs r270186 : removed unused help ids 2009-03-28 20:19:40 +0100 fs r270185 : removed unused help ids 2009-03-28 20:19:10 +0100 fs r270184 : #i100237# +DefaultState/XReset 2009-03-28 00:29:29 +0100 fs r270177 : CWS-TOOLING: rebase CWS dba32b to trunk@270033 (milestone: DEV300:m45) 2009-03-27 22:56:46 +0100 fs r270173 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:55:52 +0100 fs r270172 : #i100237# DefaultState property for buttons, enabled only when Toggle=Yes 2009-03-27 22:54:15 +0100 fs r270171 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:53:54 +0100 fs r270170 : #i100237# introduce a DefaultState property for buttons, which implies buttongs supporting XReset, which needed some refactoring 2009-03-27 13:31:41 +0100 fs r270152 : ignore output paths 2009-03-27 11:23:44 +0100 fs r270139 : tuned behavior with respect to invalid keys/values 2009-03-27 09:57:14 +0100 fs r270136 : don't allow Double.NaN 2009-03-27 09:56:16 +0100 fs r270135 : talk about Double.NaN 2009-03-26 12:14:30 +0100 fs r270067 : removed unused parameter 2009-03-26 12:14:02 +0100 fs r270066 : removed widening conversion when checking keys 2009-03-26 09:17:34 +0100 fs r270053 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:17:11 +0100 fs r270052 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:16:49 +0100 fs r270051 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-25 21:55:20 +0100 fs r270044 : #i100541# properly calculat bNeedScrollBox 2009-03-25 12:56:17 +0100 fs r270019 : renamed the previously introduced error condition 2009-03-25 12:11:48 +0100 fs r270015 : #i100095# when the error messages contain non-trivial details (SQLState/ErrorCode), then always display the 'Details' button 2009-03-25 12:10:05 +0100 fs r270012 : renamed the previously introduced error condition 2009-03-25 12:01:04 +0100 fs r270011 : #i100095# when no address book (or respective profile) can be found, then use a dedicated ErrorCondition 2009-03-25 10:29:07 +0100 fs r270003 : add missing localization 2009-03-25 10:23:12 +0100 fs r270001 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-25 10:21:55 +0100 fs r270000 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-20 23:05:38 +0100 fs r269829 : XMap::getSize removed 2009-03-20 23:05:19 +0100 fs r269828 : changes as suggested by sb 2009-03-20 23:04:56 +0100 fs r269827 : enhanced documentation 2009-03-20 15:31:40 +0100 fs r269815 : changes as suggested by sb 2009-03-20 13:23:52 +0100 oj r269804 : #i92538# correct the zorder that fixedtext is infront of control 2009-03-20 12:59:38 +0100 oj r269801 : #i94571# paraadjust now supports BLOCK 2009-03-20 12:58:11 +0100 oj r269800 : #i94571# export style with data style 2009-03-20 12:57:05 +0100 oj r269799 : #i94571# text align is now paraadjust 2009-03-20 12:37:02 +0100 fs r269797 : enum keys only accepted if they have the exact type, not if they have *any* enum type 2009-03-20 12:28:31 +0100 fs r269794 : some changes requested by sb 2009-03-20 08:52:47 +0100 fs r269780 : doc 2009-03-20 07:37:31 +0100 oj r269779 : #i99913# only notifiy when values are different 2009-03-20 07:36:58 +0100 oj r269778 : #i99913# add undoenv as listener at the dataprovider 2009-03-19 22:52:52 +0100 fs r269771 : added comment 2009-03-19 22:40:06 +0100 fs r269770 : +testEnumeration 2009-03-19 22:39:41 +0100 fs r269769 : implemented enumeration, getKeySet, and getValues. Should be finished now. 2009-03-19 14:01:01 +0100 oj r269743 : #i99913# reset the modified state when selecting an object 2009-03-19 12:19:54 +0100 lla r269739 : #i72390# cleanups 2009-03-19 09:25:27 +0100 fs r269727 : #i10000# 2009-03-18 23:37:02 +0100 fs r269708 : extended checks for value type acceptance 2009-03-18 23:36:41 +0100 fs r269707 : fixed value type checks 2009-03-18 14:59:56 +0100 fs r269678 : initial complex test case for the new css.container.Map implementation 2009-03-18 14:59:24 +0100 fs r269677 : verifyExpectedException moved to base class (in complexlib), and renamed to assureException for consistency 2009-03-18 14:58:35 +0100 fs r269676 : removed unused imports 2009-03-18 14:58:03 +0100 fs r269675 : first implementation of the new css.container.Map service (not completed, yet) 2009-03-18 14:57:17 +0100 fs r269674 : base class for UNO components, freeing you from some repeating work 2009-03-18 14:55:53 +0100 fs r269672 : +assureException: call a given method with given parameters on a given object, ensure that a given exception is thrown by the method implementation 2009-03-18 14:54:58 +0100 fs r269671 : +getComponentContext 2009-03-18 14:54:00 +0100 fs r269670 : isEmpty returns a boolean, not a long 2009-03-18 14:14:43 +0100 oj r269663 : #i99743# now text also supports null 2009-03-18 13:54:14 +0100 oj r269660 : #i99223# remove check for 2 params 2009-03-18 13:33:35 +0100 oj r269659 : #i99060# replace text::ParagraphVertAlign with style::VerticalAlignment 2009-03-18 13:32:18 +0100 oj r269658 : #i99060# don't set void property when void isn't allowed 2009-03-18 13:31:11 +0100 oj r269657 : #i99060# handle vertical alignment 2009-03-18 13:28:28 +0100 oj r269656 : #i99060# remove unused elements from sytle 2009-03-18 09:35:42 +0100 lla r269639 : #i72390# cleanups 2009-03-18 09:31:20 +0100 lla r269638 : #i72390# add ButtonList 2009-03-18 09:30:46 +0100 lla r269637 : #i72390# renamed interface 2009-03-18 09:30:15 +0100 lla r269636 : #i72390# use ButtonList instead of ImageList 2009-03-18 09:29:05 +0100 lla r269635 : #i72390# new ButtonList, cleanups 2009-03-18 09:26:34 +0100 lla r269634 : #i72390# cleanups 2009-03-17 12:21:20 +0100 oj r269590 : #i99222# remove assertion 2009-03-17 12:17:22 +0100 oj r269589 : #i98605# impl scale mode 2009-03-17 12:10:42 +0100 oj r269588 : #i98605# impl scale mode 2009-03-17 11:40:15 +0100 oj r269584 : #i96944# doesn't create equation for shapes 2009-03-17 11:33:16 +0100 oj r269583 : #i96423# switch calc from float to long 2009-03-16 15:19:18 +0100 fs r269550 : #i41930# enable zoom for embedded/outplace documents 2009-03-16 14:25:54 +0100 oj r269542 : #i93734# remove ContextSensitive 2009-03-16 14:21:58 +0100 oj r269541 : #i99274# page header before group header 2009-03-16 14:18:23 +0100 oj r269539 : #i99110# fix value type 2009-03-16 14:14:16 +0100 fs r269537 : line ends 2009-03-16 14:11:06 +0100 fs r269535 : line ends 2009-03-16 14:08:34 +0100 fs r269534 : #i100087# (provided my np): allow for polymorphic types with more than one parameter 2009-03-16 12:30:31 +0100 oj r269521 : compile error 2009-03-16 12:19:12 +0100 oj r269519 : compile error 2009-03-16 10:39:28 +0100 oj r269511 : compile error under linux with swap 2009-03-13 10:33:04 +0100 oj r269462 : CWS-TOOLING: rebase CWS dba32b to trunk@269297 (milestone: DEV300:m43) 2009-03-12 14:37:25 +0100 fs r269416 : interface SequenceOutputStreamTest is unneeded, and pollutes the namespace here :) 2009-03-12 14:35:07 +0100 fs r269414 : not needed 2009-03-12 14:34:15 +0100 fs r269413 : preparation for multiple tests in this module 2009-03-12 14:33:02 +0100 fs r269412 : ShowTargets was moved from module integration.forms to module complexlib 2009-03-12 14:32:48 +0100 fs r269411 : helper class for projects containing multiple complex test cases (and following a certain structure) 2009-03-12 14:00:14 +0100 fs r269407 : proper module after the move 2009-03-12 13:59:10 +0100 fs r269406 : superseded by ../makefile.mk 2009-03-12 13:47:38 +0100 fs r269403 : not needed anymore 2009-03-12 13:45:46 +0100 fs r269402 : moved, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:45:07 +0100 fs r269401 : moved from ../ 2009-03-12 13:43:59 +0100 fs r269400 : moved to ./comphelper, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:29:47 +0100 oj r269396 : #i99914# set parent on dataprovider 2009-03-12 13:10:35 +0100 oj r269393 : #i99832# check thrown exception and show error 2009-03-12 13:08:10 +0100 fs r269392 : reorganizing tests 2009-03-12 12:52:55 +0100 oj r269390 : #i99118# convert formatkey in numberformat 2009-03-12 12:34:53 +0100 fs r269388 : new API tests 2009-03-12 12:29:05 +0100 fs r269386 : Map not yet committed 2009-03-12 12:28:36 +0100 fs r269385 : oops, forgot the SequenceInputStream during the previous refactoring 2009-03-12 12:12:39 +0100 oj r269384 : #i99104# set HasCategories prop 2009-03-12 12:12:08 +0100 oj r269383 : #i99104# check HasCategories even for internal dataprovider 2009-03-12 12:10:40 +0100 oj r269382 : #i99104# set HasCategories prop 2009-03-12 10:51:49 +0100 fs r269373 : #i10000# exception specifications 2009-03-12 10:49:18 +0100 fs r269372 : #i10000# exception specifications 2009-03-12 10:44:02 +0100 fs r269371 : #i10000# exception specifications 2009-03-12 10:30:55 +0100 fs r269368 : refactored the UNO service registration in this module, using the helper classes provided by comphelper itself, so you have less effort when extending the list of to-be-registered components 2009-03-12 10:30:37 +0100 fs r269367 : module-local includes 2009-03-12 07:05:54 +0100 oj r269357 : #i99104# database dataprovider doesn't need dataranges and diagramdata 2009-03-11 10:58:28 +0100 oj r269306 : #i99911# check if name of the report is a valid file name 2009-03-11 10:03:23 +0100 oj r269299 : #i99666# the report is new when the HierarchicalDocumentName is empty 2009-03-10 11:32:45 +0100 oj r269258 : #i99221# use fallback for language 2009-03-10 10:48:40 +0100 oj r269255 : #i99433# now use OStringBuffer 2009-03-10 10:36:21 +0100 fs r269252 : initial version of (X)Map 2009-03-10 09:52:23 +0100 oj r269246 : #i99433# now use OStringBuffer 2009-03-10 08:56:13 +0100 oj r269240 : #i99655# patch applied 2009-03-09 07:35:33 +0100 lla r269058 : #i10000# wrong variable assignment fixed 2009-03-06 17:20:40 +0100 fs r269030 : some explicit defaults 2009-03-06 17:20:30 +0100 fs r269029 : #i98600# 2009-03-06 14:40:34 +0100 fs r269009 : #i87692# during reload, prevent the document being modified just because of some control content changes ... 2009-03-06 12:52:20 +0100 lla r268997 : #i10000# ambigous problem with FontWeight fixed 2009-03-06 11:39:32 +0100 fs r268989 : #i10000# (approved by pl): use --without-t1-library configure option 2009-03-06 10:55:43 +0100 fs r268986 : #i99953# depends on xmlscript module now 2009-03-06 10:54:04 +0100 fs r268985 : #i99953# also adjust the event names found in dialogs embedded in the forms 2009-03-06 09:53:41 +0100 fs r268977 : #i10000# 2009-03-06 09:30:41 +0100 lla r268973 : #i10000# merge problems 2009-03-05 17:52:34 +0100 fs r268932 : #i98593# for sub components which are actually controlled by a DocumentDefinition (aka XComponentSupplier aka XCommandProcessor), close them by executing the 'close' command, not by suspending/closing the controller (which cannot be intercepted) 2009-03-05 11:41:56 +0100 fs r268889 : default the drop down line count for list/combo boxes to 20 2009-03-05 11:39:10 +0100 fs r268887 : do not display empty error messages 2009-03-02 10:13:57 +0100 lla r268639 : #i91541# CWS rebase m41 to m42 2009-03-02 09:06:27 +0100 lla r268635 : #i10000# add ';' to strings 2009-02-26 11:18:00 +0100 fs r268492 : reportdesign depends on REPORTBUILDER, not REPORTDESIGN 2009-02-26 10:11:38 +0100 lla r268489 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-26 09:04:11 +0100 lla r268488 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-24 12:09:13 +0100 lla r268392 : #i91541# #i91542# cleanups 2009-02-24 08:08:06 +0100 lla r268382 : merge all dba32a changes into dba32b 2009-02-24 07:14:55 +0100 lla r268381 : merge all dba32a changes into dba32b 2009-02-23 21:44:28 +0100 fs r268377 : oops ... don't tamper with m_aListSourceValues at the end of loadData 2009-02-23 20:57:05 +0100 fs r268376 : #i98162# don't hold the values as strings, but as ORowSetValue, this way preserving their type, and being agnostic to different result/rowset implementations doing different to-string-conversations 2009-02-23 20:55:44 +0100 fs r268375 : getObject: throwFunctionNotSupportedException, instead of silently returning NULL 2009-02-23 20:55:20 +0100 fs r268374 : #i98162# some more supported types 2009-02-23 20:54:43 +0100 fs r268373 : #i98162# +operator != 2009-02-20 09:35:39 +0100 fs r268306 : #i99422# for a font, display the font name, the style, and the size 2009-02-20 09:33:45 +0100 fs r268305 : #i99422# in the property browser, FONT supersedes CHARFONTNAME: the aggregated FormComponentHandler displays them more nicely now 2009-02-19 16:12:06 +0100 fs r268293 : #i99372# recognize DataType::FLOAT as numeric 2009-02-19 15:43:12 +0100 fs r268291 : #i99415# 2009-02-19 15:40:15 +0100 fs r268290 : #i99242# lcl_firstFocussableControl: take disabled controls into account 2009-02-19 15:34:36 +0100 fs r268289 : #i99396# properly decode the base name of the URL when using it as title 2009-02-19 15:19:05 +0100 fs r268287 : #i98247#
2009-06-05 09:47:55 +00:00
// -----------------------------------------------------------------------------
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
void ORowSetValue::impl_fill( const sal_Int32 _nType, bool _bNullable, const detail::IValueSource& _rValueSource )
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
CWS-TOOLING: integrate CWS dbaperf1 2009-04-02 09:12:25 +0200 oj r270372 : CWS-TOOLING: rebase CWS dbaperf1 to trunk@270033 (milestone: DEV300:m45) 2009-03-17 10:20:34 +0100 oj r269577 : add parameter 2009-03-17 10:20:05 +0100 oj r269576 : add parameter 2009-03-12 12:32:24 +0100 oj r269387 : add missing ++ 2009-03-11 12:53:35 +0100 oj r269315 : compile error 2009-03-06 08:52:11 +0100 oj r268968 : #i99264# remove duplicate code 2009-03-06 08:20:08 +0100 oj r268966 : add missingheader 2009-03-06 08:17:41 +0100 oj r268965 : add header 2009-03-06 08:15:07 +0100 oj r268964 : #i99708# remove duplicate code 2009-03-06 07:24:11 +0100 oj r268963 : #i99708# remove duplicate code 2009-03-06 07:22:24 +0100 oj r268962 : #i99708# use tenary operator 2009-03-06 07:19:21 +0100 oj r268961 : remove unused rtl_logs 2009-03-06 07:15:55 +0100 oj r268960 : #i99708# extract getLength from for loop 2009-03-06 07:15:35 +0100 oj r268959 : #i99708# extract getLength from for loop 2009-03-06 07:14:57 +0100 oj r268958 : remove unused rtl_logs 2009-03-06 07:13:01 +0100 oj r268957 : insert DBG_ helper 2009-03-06 07:12:17 +0100 oj r268956 : remove unused rtl_logs 2009-03-06 07:10:26 +0100 oj r268955 : remove unused rtl_logs 2009-03-06 07:04:51 +0100 oj r268954 : #i99708# use tenary operator 2009-03-05 12:04:46 +0100 oj r268895 : comment RTL_LOG out 2009-03-05 09:05:07 +0100 oj r268874 : add rtl logfile 2009-03-04 14:21:18 +0100 oj r268828 : #i99708# make static inplace and some method calls in for loop removed 2009-03-04 14:20:34 +0100 oj r268827 : #i99708# impl double check pattern for getInfohelper 2009-03-02 09:31:42 +0100 oj r268636 : add rtl logfile 2009-03-02 08:18:37 +0100 oj r268633 : add rtl logfile 2009-02-27 11:22:16 +0100 oj r268570 : #i99709# change algorithm for marking objects 2009-02-27 11:17:04 +0100 oj r268568 : #i99708# some improvements to load forms, controls faster and replacement of size() call with !empty() which is much faster 2009-02-24 10:09:35 +0100 sb r268383 : #i99290# no longer care to set a sensible context class loader for native threads attached to the VM 2009-02-23 13:44:04 +0100 oj r268354 : #i76606# seekrow changes 2009-02-23 12:27:30 +0100 oj r268349 : deleted 2009-02-20 15:14:23 +0100 oj r268325 : #i76606# some code changes 2009-02-19 14:15:25 +0100 oj r268284 : #i76606# use of simple prep stmt instead of full blown rowset, when source and dest connection are the same use insert into ... ( select ... ) 2009-02-19 11:27:55 +0100 oj r268265 : #i76606# insert some RTL_LOG and setObject impl 2009-02-18 14:45:37 +0100 oj r268222 : #i99363# insert RTL_LOG 2009-02-18 14:45:18 +0100 oj r268221 : #i99363# insert RTL_LOG 2009-02-18 11:14:54 +0100 oj r268207 : #i99363# call some impl_ methods to avoid duplicate cechCache calls 2009-02-18 11:10:47 +0100 oj r268206 : #i99363# make isCount inline 2009-02-18 10:33:22 +0100 oj r268203 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-18 10:32:57 +0100 oj r268202 : #i99363# use bookmarkable if available and inserted some RTL_LOG 2009-02-17 07:29:05 +0100 oj r267843 : #i96897# remove some dll public 2009-02-16 15:01:04 +0100 oj r267816 : #i96897# remove some dll public 2009-02-16 14:25:53 +0100 oj r267810 : #i99264# remove duplicate code 2009-02-16 14:25:33 +0100 oj r267809 : #i99264# remove duplicate code 2009-02-16 14:24:59 +0100 oj r267808 : #i99264# remove duplicate code 2009-02-13 10:56:17 +0100 oj r267703 : #i99191# comment the contextclassloader 2009-02-13 10:32:40 +0100 oj r267700 : reduce call to resultset meta data 2009-02-13 10:27:31 +0100 oj r267699 : reduce call to resultset meta data 2009-02-13 10:27:08 +0100 oj r267698 : reduce call to resultset meta data
2009-04-23 10:42:05 +00:00
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::fill (2)" );
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bool bReadData = sal_True;
switch(_nType)
{
case DataType::CHAR:
case DataType::VARCHAR:
case DataType::DECIMAL:
case DataType::NUMERIC:
case DataType::LONGVARCHAR:
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getString();
break;
case DataType::BIGINT:
if ( isSigned() )
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getLong();
else
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
// TODO: this is rather horrible performance-wise
// but fixing it needs extending the ::com::sun::star::sdbc::XRow API
// to have a getULong(), and needs updating all drivers :-|
// When doing that, add getUByte, getUShort, getUInt for symmetry/completeness
(*this) = _rValueSource.getString().toUInt64();
break;
case DataType::FLOAT:
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getFloat();
break;
case DataType::DOUBLE:
case DataType::REAL:
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getDouble();
break;
case DataType::DATE:
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getDate();
break;
case DataType::TIME:
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getTime();
break;
case DataType::TIMESTAMP:
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getTimestamp();
break;
case DataType::BINARY:
case DataType::VARBINARY:
case DataType::LONGVARBINARY:
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getBytes();
break;
case DataType::BIT:
case DataType::BOOLEAN:
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getBoolean();
break;
case DataType::TINYINT:
if ( isSigned() )
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getByte();
else
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getShort();
break;
case DataType::SMALLINT:
if ( isSigned() )
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getShort();
else
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getInt();
break;
case DataType::INTEGER:
if ( isSigned() )
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getInt();
else
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
(*this) = _rValueSource.getLong();
break;
case DataType::CLOB:
(*this) = ::com::sun::star::uno::makeAny(_rValueSource.getClob());
setTypeKind(DataType::CLOB);
break;
case DataType::BLOB:
(*this) = ::com::sun::star::uno::makeAny(_rValueSource.getBlob());
setTypeKind(DataType::BLOB);
break;
2009-11-11 13:30:02 +01:00
case DataType::OTHER:
2009-12-10 08:08:38 +01:00
(*this) = _rValueSource.getObject();
2009-11-11 13:30:02 +01:00
setTypeKind(DataType::OTHER);
break;
default:
SAL_WARN( "connectivity.commontools", "ORowSetValue::fill: unsupported type!" );
(*this) = _rValueSource.getObject();
break;
}
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
if ( bReadData && _bNullable && _rValueSource.wasNull() )
setNull();
setTypeKind(_nType);
}
// -----------------------------------------------------------------------------
void ORowSetValue::fill(const Any& _rValue)
{
SAL_INFO( "connectivity.commontools", "dbtools Ocke.Janssen@sun.com ORowSetValue::fill (3)" );
switch (_rValue.getValueType().getTypeClass())
{
case TypeClass_VOID:
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
setNull(); break;
case TypeClass_BOOLEAN:
{
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
bool bValue( sal_False );
_rValue >>= bValue;
(*this) = bValue;
break;
}
case TypeClass_CHAR:
{
sal_Unicode aDummy(0);
_rValue >>= aDummy;
(*this) = OUString(aDummy);
break;
}
case TypeClass_STRING:
{
OUString sDummy;
_rValue >>= sDummy;
(*this) = sDummy;
break;
}
case TypeClass_FLOAT:
{
float aDummy(0.0);
_rValue >>= aDummy;
(*this) = aDummy;
break;
}
case TypeClass_DOUBLE:
{
double aDummy(0.0);
_rValue >>= aDummy;
(*this) = aDummy;
break;
}
case TypeClass_BYTE:
{
sal_Int8 aDummy(0);
_rValue >>= aDummy;
(*this) = aDummy;
break;
}
case TypeClass_SHORT:
{
sal_Int16 aDummy(0);
_rValue >>= aDummy;
(*this) = aDummy;
break;
}
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
case TypeClass_UNSIGNED_SHORT:
{
sal_uInt16 nValue(0);
_rValue >>= nValue;
(*this) = nValue;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
break;
}
case TypeClass_LONG:
{
sal_Int32 aDummy(0);
_rValue >>= aDummy;
(*this) = aDummy;
break;
}
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
case TypeClass_UNSIGNED_LONG:
{
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
sal_uInt32 nValue(0);
_rValue >>= nValue;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
(*this) = static_cast<sal_Int64>(nValue);
setSigned(sal_False);
break;
}
CWS-TOOLING: integrate CWS dbaperf2 2009-06-22 11:04:36 +0200 msc r273202 : #100000# fix to run this test 2009-06-22 09:41:04 +0200 msc r273200 : #100000# correct case 2009-06-22 09:02:42 +0200 msc r273199 : new SRB 2009-06-22 08:53:26 +0200 msc r273198 : #100000 2009-06-15 12:36:32 +0200 oj r272976 : test 2009-06-12 14:23:23 +0200 msc r272918 : #101587# unused feature was removed 2009-06-12 06:45:17 +0200 oj r272903 : use GUI instead of OS 2009-06-11 09:12:30 +0200 oj r272845 : remove unused dep 2009-06-11 09:11:09 +0200 oj r272844 : ColumnAliasInOrderBy not supported 2009-06-11 08:27:00 +0200 oj r272843 : CWS-TOOLING: rebase CWS dbaperf2 to trunk@272827 (milestone: DEV300:m50) 2009-06-10 15:30:17 +0200 msc r272822 : #100000 2009-06-10 15:27:40 +0200 msc r272820 : #102515# new function for the database type list 2009-06-10 15:26:09 +0200 msc r272818 : move file to dbaccess/tools/dbcreatetools.inc 2009-06-10 15:20:10 +0200 msc r272816 : #102515# new function for the database type list 2009-06-10 12:34:46 +0200 oj r272803 : #i101587# check extension 2009-06-10 11:53:45 +0200 oj r272800 : remove unused code 2009-06-09 09:57:25 +0200 oj r272752 : #i101587# EscapeDateTime moved into features section 2009-06-09 09:56:26 +0200 oj r272751 : enable finish 2009-06-08 11:50:37 +0200 oj r272726 : #i102588# move convert tzo saveDataSource 2009-06-05 11:54:20 +0200 jsk r272676 : #i102515# - NEW: Functions to retrieve Database names from API 2009-06-05 10:52:22 +0200 jsk r272671 : NEW: Functions to retrieve Database names from API 2009-06-05 09:36:54 +0200 oj r272666 : handle win and unx differently 2009-06-05 09:36:28 +0200 oj r272665 : handle win and unx differently 2009-06-04 13:41:56 +0200 oj r272625 : #i101268# make use of stringbuffer instead of oustring 2009-06-04 10:18:14 +0200 oj r272612 : add SAL_DLLPUBLIC_EXPORT 2009-05-28 10:56:33 +0200 oj r272381 : #i101587# create langpacks 2009-05-28 10:41:59 +0200 oj r272380 : #i101587# create langpacks 2009-05-28 10:29:30 +0200 oj r272379 : #i101587# create langpacks 2009-05-07 10:11:46 +0200 oj r271635 : fix compile warnings 2009-05-07 07:01:31 +0200 oj r271619 : change type from int32 to PathID 2009-05-07 06:53:19 +0200 oj r271618 : add all 2009-05-06 14:57:02 +0200 oj r271595 : add missing lib 2009-05-06 14:56:54 +0200 oj r271594 : no used anymore 2009-05-06 14:56:08 +0200 oj r271593 : no used anymore 2009-05-06 14:54:25 +0200 oj r271592 : no used anymore 2009-05-06 14:45:59 +0200 oj r271591 : add missing lib 2009-05-06 14:41:54 +0200 oj r271590 : change wnt 2009-05-06 14:22:21 +0200 oj r271589 : #i101587# use config for the drivers 2009-05-06 14:21:39 +0200 oj r271588 : #i101587# add new set for metadata and config data for driver 2009-05-06 14:19:36 +0200 oj r271587 : #i101587# handle metadata and props of driver 2009-05-06 10:11:11 +0200 oj r271570 : #i101587# correct wrong use of reference 2009-05-06 09:24:33 +0200 oj r271567 : #i101587# add new set for metadata 2009-05-06 08:52:06 +0200 oj r271563 : #i101587# add driver configuration 2009-05-06 08:51:04 +0200 oj r271562 : #i101587# add driver configuration 2009-05-04 10:24:41 +0200 oj r271442 : CWS-TOOLING: rebase CWS dbaperf2 to trunk@271427 (milestone: DEV300:m47) 2009-04-28 07:06:22 +0200 oj r271303 : compile error 2009-04-27 14:12:48 +0200 oj r271270 : #i101268# make use of stringbuffer instead of oustring 2009-04-27 14:12:27 +0200 oj r271269 : #i101268# make use of stringbuffer instead of oustring 2009-04-27 09:30:10 +0200 oj r271255 : use string buffer when possible 2009-04-24 11:56:01 +0200 oj r271207 : #i101268# remove end() call from loop 2009-04-23 14:24:03 +0200 oj r271171 : add dep 2009-04-23 13:25:25 +0200 oj r271164 : #i101268# add config entries for database drivers 2009-04-23 13:07:13 +0200 oj r271161 : #i101268# add config entries for database drivers 2009-04-23 13:06:33 +0200 oj r271160 : #i101268# add parent url pattern entry for driver node 2009-04-23 13:05:39 +0200 oj r271159 : #i101268# add config entry for driver node 2009-04-23 13:04:09 +0200 oj r271157 : #i101268# add config entry for driver node 2009-04-23 06:41:32 +0200 oj r271142 : #i101268# add new configuration file Drivers 2009-04-22 13:49:19 +0200 oj r271103 : convert EOL 2009-04-22 13:29:17 +0200 oj r271101 : #i101268# use of driver configuration entries for features and properties 2009-04-22 13:28:08 +0200 oj r271100 : #i101268# use of driver configuration entries for features and properties 2009-04-22 13:15:26 +0200 oj r271098 : #i101268# add new configuration file Drivers 2009-04-22 13:14:43 +0200 oj r271097 : #i101268# use of driver configuration entries for features and properties
2009-07-03 12:24:35 +00:00
case TypeClass_HYPER:
{
sal_Int64 nValue(0);
_rValue >>= nValue;
(*this) = nValue;
break;
}
case TypeClass_UNSIGNED_HYPER:
{
sal_uInt64 nValue(0);
_rValue >>= nValue;
ORowSetValue: clean up sign/unsigned union member Also switch BOOLEAN constructor from sal_Bool to bool. old/new signed/unsigned storage situation: ------------------------------------------------------- SQL type | signed | unsigned old | unsigned new ------------------------------------------------------- TINYINT | sal_Int8 | sal_Int16 | sal_uInt8 SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16 INTEGER | sal_Int32 | sal_Int64 | sal_uInt32 BIGINT | sal_Int64 | pValue (String*) | sal_uInt64 ------------------------------------------------------- When sticking an UNSIGNED TINYINT into an Any, silently promote it to UNSIGNED SMALLINT (that is sal_uInt16), else Any would take it as a sal_Bool and normalise to sal_True (1) or sal_False (0). When constructing an ORowSetValue from a sal_Bool, silently keep it as an unsigned 8 bit integer (that is understand it as a sal_uInt8). This will work in most cases, since when asked back for a bool or sal_Bool, we'll give back the right value. Only code looking at the type tag could possibly make a "wrong" decision. The main (hopefully only?) path through which this would happen is through an implementation of XParameters::setBoolean XRowUpdate::updateBoolean that would use its sal_Bool argument to construct an ORowSetValue. So make sure each implementation constructs a proper BOOLEAN so as not to get confused. For authorship/copyright purposes, this patch is a cooperation between Lionel Elie Mamane <lionel@mamane.lu> and David Ostrovsky <david@ostrovsky.org> Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
2013-02-04 11:12:57 +01:00
(*this) = nValue;
setSigned(sal_False);
break;
}
CWS-TOOLING: integrate CWS dba32b 2009-06-03 14:58:08 +0200 fs r272581 : #i102439# 2009-05-29 13:56:18 +0200 fs r272456 : remove the sub form when the 'add subform' setting changes from <true/> to <false/>, not only its controls 2009-05-29 13:19:27 +0200 fs r272454 : display '(Default)' instead of an empty string when a control has the default font 2009-05-28 20:49:18 +0200 fs r272428 : #i98162# getFirstSelectedValue: do not return reference to a temporary 2009-05-27 15:30:22 +0200 msc r272353 : #102303# 2009-05-26 13:03:06 +0200 fs r272295 : spelling 2009-05-26 12:59:54 +0200 fs r272294 : merge m48 version to get latest cygwin related fixes 2009-05-25 14:02:06 +0200 fs r272239 : remove references to local files, needed for debugging sessions only 2009-05-25 14:01:16 +0200 fs r272238 : #i102021# ensure members such as bNumberFormat are initialized before actually returning them 2009-05-25 13:10:20 +0200 fs r272236 : #i10000# reset ENABLE_EVOAB2 2009-05-22 06:44:45 +0200 oj r272167 : #i99104# add import handler for calc-sett 2009-05-22 06:42:27 +0200 oj r272166 : #i99104# impl NullDate as member 2009-05-22 06:36:22 +0200 oj r272165 : #i99104# handle nulldate from parent model 2009-05-22 06:33:13 +0200 oj r272164 : #i99104# export and import calculation-settings and nulldate 2009-05-22 06:27:50 +0200 oj r272163 : #i99104# export and import calculation-settings and nulldate 2009-05-20 13:29:58 +0200 oj r272118 : #i99911# handle invalid name 2009-05-20 13:28:49 +0200 oj r272117 : #i101261# handle different rowsets 2009-05-20 11:29:55 +0200 msc r272111 : #i100000# 2009-05-20 11:28:27 +0200 msc r272110 : merge in change from dba32a 2009-05-20 11:27:38 +0200 msc r272109 : #102082# CTRL + C does not work 2009-05-20 09:43:36 +0200 oj r272106 : clean up includes 2009-05-20 09:32:15 +0200 oj r272105 : #i99060# merge error resovled now VERTICAL_ALIGN is in both stmt 2009-05-20 08:37:21 +0200 msc r272104 : add workaround for issue #102010# 2009-05-20 08:10:12 +0200 oj r272103 : #i99104# use numberformatkey 2009-05-20 08:07:02 +0200 oj r272102 : #i99104# use column info from rowset 2009-05-20 08:04:43 +0200 oj r272101 : #i102032# use a special column type where prec and scale are the values currently set at the column 2009-05-20 08:03:04 +0200 oj r272100 : #i102032# correct type info, we have to use SQL defined type names 2009-05-19 10:27:02 +0200 oj r272061 : #i99104# export null-date 2009-05-19 08:26:53 +0200 oj r272056 : #i99104# export null-date 2009-05-18 13:15:10 +0200 msc r272014 : add issue #102019# 2009-05-18 11:33:07 +0200 msc r272005 : add issue #102019# 2009-05-18 08:59:45 +0200 msc r271996 : add workaroud for issue #102010# 2009-05-15 10:21:24 +0200 msc r271929 : #101944# 2009-05-11 21:18:30 +0200 fs r271792 : #i99914# 2009-05-08 13:52:06 +0200 oj r271715 : #i96423# remember column span 2009-05-08 11:26:19 +0200 oj r271708 : #i98605# impl new scale mode 2009-05-08 10:33:35 +0200 fs r271706 : SendUserCall: only call into the shape notification routine for UserCall types where this is necessary (performance issue) 2009-05-07 20:52:44 +0200 fs r271698 : outsource ShapeProperty from shapepropertynotifier.hxx 2009-05-07 20:43:33 +0200 fs r271697 : #i99056# use notifyShapePropertyChange, instead of getShapePropertyChangeNotifier - the latter throws if no shape exists, yet 2009-05-07 20:33:58 +0200 fs r271696 : #i99056# +notifyShapePropertyChange: allow notifying chages without checking whether there actually already exists an SvxShape 2009-05-07 16:22:15 +0200 fs r271679 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:21:37 +0200 fs r271678 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:01:11 +0200 oj r271677 : #i99914# notify parent handler 2009-05-07 15:54:54 +0200 fs r271676 : #i10000# cygwin needs some quoting 2009-05-07 14:49:48 +0200 oj r271672 : #i99277# quote alias name 2009-05-07 14:48:12 +0200 oj r271671 : #i92538# add ~ in front of type 2009-05-07 14:37:13 +0200 oj r271667 : #i99118# change type from char to varchar 2009-05-07 14:36:23 +0200 oj r271666 : #i99118# clear dest columns when changing to create new table 2009-05-07 13:35:32 +0200 oj r271657 : #i94467# handle type 0 as double as well 2009-05-07 13:20:49 +0200 oj r271655 : i99743# setNull when varchar is no text 2009-05-07 12:58:06 +0200 fs r271651 : initialize the SdrObject's property change notifier after the ctor, if necessary 2009-05-07 11:47:18 +0200 fs r271647 : #i10000# 2009-05-07 10:57:16 +0200 fs r271639 : OPropertyBrowserController::propertyChange: care for the current property state, too, and properly forward it to the UI 2009-05-07 10:18:14 +0200 fs r271636 : onNewComponent: do not ask the map for grid columns, it will throw 2009-05-07 10:09:55 +0200 fs r271634 : #i101623# 2009-05-07 09:53:44 +0200 fs r271631 : #i101622# 2009-05-06 21:55:53 +0200 fs r271615 : #i10000# 2009-05-06 21:10:42 +0200 fs r271611 : #i10000# 2009-05-06 13:11:48 +0200 fs r271583 : #i10000# 2009-05-05 22:29:31 +0200 fs r271559 : proper assertion message 2009-05-05 22:29:03 +0200 fs r271558 : diagnostics 2009-05-05 22:16:16 +0200 fs r271557 : #i10000# 2009-05-05 13:50:32 +0200 fs r271513 : #i10000# 2009-05-05 10:21:50 +0200 fs r271503 : #i10000# 2009-05-05 09:30:26 +0200 fs r271501 : why did those survive the rebase -C step? 2009-05-05 09:18:12 +0200 fs r271500 : #i10000# 2009-05-04 17:08:17 +0200 fs r271475 : CWS-TOOLING: rebase CWS dba32b to trunk@271427 (milestone: DEV300:m47) 2009-05-04 14:51:26 +0200 fs r271456 : line ends 2009-04-30 15:55:27 +0200 fs r271418 : NewURL -> PublicConnectionURL 2009-04-22 21:18:34 +0200 fs r271141 : #i100944# 2009-04-22 09:12:26 +0200 oj r271071 : #i101261# little code change 2009-04-22 09:11:43 +0200 oj r271070 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:11:25 +0200 oj r271069 : #i101261# new grabage container for nodes 2009-04-22 09:11:02 +0200 oj r271068 : #i101261# new grabage container for nodes 2009-04-22 09:10:44 +0200 oj r271067 : #i101261# new grabage container for nodes 2009-04-22 09:10:21 +0200 oj r271066 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:08:24 +0200 oj r271065 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:07:25 +0200 oj r271064 : #i101261# only ask for parameters which aren't set before 2009-04-22 08:49:07 +0200 oj r271062 : #i77501# preview only when needed 2009-04-22 08:45:44 +0200 oj r271061 : #i101261# new prop max rows 2009-04-22 08:44:18 +0200 oj r271060 : #i101261# create dataprovider earlier to avoid the wrong legend in chart 2009-04-22 08:42:48 +0200 oj r271059 : #i101261# handle parameter 2009-04-17 21:00:23 +0200 fs r270954 : #i98350# 2009-04-17 13:54:19 +0200 fs r270942 : #i99565# 2009-04-17 13:51:34 +0200 fs r270940 : #i101153# only localize the (potentially) localizable properties when there really is support at the control model 2009-04-17 11:43:14 +0200 fs r270932 : removed superfluous include 2009-04-17 10:10:15 +0200 fs r270926 : #i10000# 2009-04-17 10:02:36 +0200 fs r270925 : #i10000# 2009-04-17 09:15:13 +0200 fs r270918 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:14:56 +0200 fs r270917 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:13:25 +0200 fs r270915 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 08:30:34 +0200 fs r270914 : removed unotools/servicehelper.hxx in favour of the (duplicated) comphelper/servicehelper.hxx 2009-04-16 21:05:25 +0200 fs r270903 : #i10000# 2009-04-16 20:43:43 +0200 fs r270902 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-16 15:57:36 +0200 fs r270891 : #i99056# make SetUnoControlModel virtual 2009-04-16 15:44:02 +0200 fs r270890 : #i99056# let the ScShapeObj notify changes in its Anchor property 2009-04-16 15:36:47 +0200 fs r270889 : #i99056# enable the sheet anchor type property, too 2009-04-16 15:33:45 +0200 fs r270887 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-15 14:53:13 +0200 fs r270844 : #i10000# 2009-04-15 13:08:29 +0200 fs r270836 : #i10000# 2009-04-15 12:28:14 +0200 fs r270832 : #i10000# 2009-04-15 10:59:14 +0200 fs r270827 : #i10000# 2009-04-15 09:41:08 +0200 oj r270823 : fix issues found with findbugs and pmd 2009-04-14 21:08:04 +0200 fs r270808 : #i99056# implement SheetAnchorType - now the only thing missing to enable it is the proper notification when it is modified 2009-04-14 17:09:00 +0200 fs r270799 : #i99056# implement XServiceInfo for the ScShapeObj 2009-04-14 17:07:55 +0200 fs r270798 : #i99056# implement TextAnchorType, partially implement SheetAnchorType 2009-04-14 15:54:05 +0200 fs r270786 : #i99056# SwXShape: notify changes of the AnchorType property 2009-04-14 15:47:32 +0200 fs r270785 : #i99056# deliver shapepropertynotifier.hxx 2009-04-14 15:46:54 +0200 fs r270784 : diagnostics 2009-04-14 15:08:28 +0200 fs r270781 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:41:09 +0200 fs r270773 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:37:23 +0200 fs r270772 : in dtor, remove the properties from the temporary component context 2009-04-14 14:36:34 +0200 fs r270771 : getWeakUnoShape made const 2009-04-14 12:23:08 +0200 oj r270757 : #i101064# add missing braces 2009-04-14 12:21:25 +0200 oj r270756 : #i101065# add braces for gcc 4.3.2 2009-04-14 12:17:45 +0200 oj r270755 : #i101059# add dep for manifest 2009-04-09 12:06:58 +0200 oj r270686 : #i93100# use OptimalSize from control to get height 2009-04-08 09:56:55 +0200 oj r270619 : #i92537# handle calculations in the select columns as well 2009-04-08 09:30:39 +0200 oj r270615 : #i96657# throw error message when the key doesn't have any columns 2009-04-07 12:08:26 +0200 oj r270592 : #i77501# impl preview of the executed report 2009-04-07 12:01:56 +0200 oj r270591 : #i77501# impl preview of the executed report 2009-04-07 11:41:03 +0200 oj r270590 : #i77501# impl preview of the executed report 2009-04-07 11:39:32 +0200 oj r270589 : #i77501# impl preview of the executed report 2009-04-07 11:29:25 +0200 oj r270588 : #i77501# convert dos to unix lineends 2009-04-07 11:28:23 +0200 oj r270587 : #i77501# impl preview of the executed report 2009-04-07 11:16:50 +0200 oj r270586 : #i77501# impl preview of the executed report 2009-04-07 11:16:00 +0200 oj r270585 : #i77501# impl preview of the executed report 2009-04-07 11:15:44 +0200 oj r270584 : #i77501# impl preview of the executed report 2009-04-07 11:15:28 +0200 oj r270583 : #i77501# impl preview of the executed report 2009-04-07 11:15:04 +0200 oj r270582 : #i77501# impl preview of the executed report 2009-04-06 15:38:54 +0200 fs r270559 : merge changes from CWS dba32a herein 2009-04-03 15:56:16 +0200 fs r270494 : ImpSvMEdit::Resize: do multiple iterations, if necessary 2009-04-03 14:35:49 +0200 fs r270487 : #i10000# 2009-04-03 13:17:16 +0200 fs r270476 : #i99056# display geometry information for controls, too 2009-04-03 13:16:37 +0200 fs r270475 : #i99056# better ordering of the geometry properties 2009-04-03 13:16:07 +0200 fs r270473 : #i99056# now that SvxShape supports property change listeners, forward add/remove requests to it 2009-04-03 13:13:18 +0200 fs r270472 : #i99056# at SvxShape, allow for PropertyChangeListeners for Size/Position 2009-04-03 09:29:27 +0200 oj r270456 : #i94571# use correct prop name 2009-04-03 09:14:54 +0200 fs r270451 : merge changes from CWS dba32a herein 2009-04-02 17:00:51 +0200 fs r270424 : better diagnostics 2009-04-02 16:35:19 +0200 fs r270421 : diagnostics 2009-04-02 16:34:50 +0200 fs r270420 : #i99056# mxUnoShape not accessible anymore, use impl_setUnoShape instead 2009-04-02 16:32:48 +0200 fs r270419 : #i99056# make getUnoShape cheaper: keep the pointer to the SvxShape all the time, so there's no need to ask for it in getUnoShape. As a consequence, we will later be able to use the pointer in scenarious where performance (potentially) matters 2009-04-02 16:31:13 +0200 fs r270417 : merge changes from CWS dba32a herein 2009-04-02 16:23:16 +0200 fs r270414 : merge changes from CWS dba32a herein 2009-04-02 14:10:35 +0200 fs r270405 : #i10000# 2009-04-02 14:06:26 +0200 fs r270404 : merge changes from CWS dba32a herein 2009-04-02 14:03:03 +0200 fs r270401 : #i10000# 2009-04-02 13:58:13 +0200 fs r270400 : #i10000# 2009-04-02 12:59:44 +0200 fs r270397 : merge changes from CWS dba32a herein 2009-04-02 12:46:30 +0200 fs r270396 : #i99056# let the form page maintain a mapping between control models and control shapes 2009-04-02 12:44:07 +0200 fs r270395 : merge changes from CWS dba32a herein 2009-04-02 12:42:06 +0200 fs r270394 : merge changes from CWS dba32a herein 2009-04-02 12:35:20 +0200 fs r270393 : #i10000# precompiled header 2009-04-02 12:05:31 +0200 fs r270392 : merge changes from CWS dba32a herein 2009-04-02 12:00:42 +0200 fs r270391 : merge changes from CWS dba32a herein 2009-04-02 11:47:26 +0200 fs r270390 : merge changes from CWS dba32a herein 2009-04-02 11:39:15 +0200 oj r270389 : #i94467# foxpro impl several new types 2009-04-02 11:35:58 +0200 fs r270387 : merge changes from CWS dba32a herein 2009-04-01 14:10:51 +0200 fs r270329 : merge changes from CWS dba32a herein 2009-03-31 17:29:50 +0200 fs r270290 : merge changes from CWS dba32a herein 2009-03-30 14:53:56 +0200 fs r270233 : #i100417# don't set grid column widths to 0, but to <void/> 2009-03-30 12:31:03 +0200 oj r270213 : #i100552# wrong orb used 2009-03-30 12:19:20 +0200 oj r270212 : #i98303# convertlike corrected to sal_Unicode 2009-03-30 11:58:25 +0200 fs r270210 : merge changes from CWS dba32a herein 2009-03-30 11:38:16 +0200 oj r270205 : remove duplicate code from merge 2009-03-30 11:02:27 +0200 fs r270202 : merge changes from CWS dba32a herein 2009-03-30 11:02:19 +0200 fs r270201 : merge changes from CWS dba32a herein 2009-03-30 10:31:26 +0200 oj r270200 : #i100665# only throw exception and do not drop table 2009-03-30 09:36:24 +0200 fs r270195 : assertion text 2009-03-28 20:21:58 +0100 fs r270187 : #ii10000# 2009-03-28 20:19:54 +0100 fs r270186 : removed unused help ids 2009-03-28 20:19:40 +0100 fs r270185 : removed unused help ids 2009-03-28 20:19:10 +0100 fs r270184 : #i100237# +DefaultState/XReset 2009-03-28 00:29:29 +0100 fs r270177 : CWS-TOOLING: rebase CWS dba32b to trunk@270033 (milestone: DEV300:m45) 2009-03-27 22:56:46 +0100 fs r270173 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:55:52 +0100 fs r270172 : #i100237# DefaultState property for buttons, enabled only when Toggle=Yes 2009-03-27 22:54:15 +0100 fs r270171 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:53:54 +0100 fs r270170 : #i100237# introduce a DefaultState property for buttons, which implies buttongs supporting XReset, which needed some refactoring 2009-03-27 13:31:41 +0100 fs r270152 : ignore output paths 2009-03-27 11:23:44 +0100 fs r270139 : tuned behavior with respect to invalid keys/values 2009-03-27 09:57:14 +0100 fs r270136 : don't allow Double.NaN 2009-03-27 09:56:16 +0100 fs r270135 : talk about Double.NaN 2009-03-26 12:14:30 +0100 fs r270067 : removed unused parameter 2009-03-26 12:14:02 +0100 fs r270066 : removed widening conversion when checking keys 2009-03-26 09:17:34 +0100 fs r270053 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:17:11 +0100 fs r270052 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:16:49 +0100 fs r270051 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-25 21:55:20 +0100 fs r270044 : #i100541# properly calculat bNeedScrollBox 2009-03-25 12:56:17 +0100 fs r270019 : renamed the previously introduced error condition 2009-03-25 12:11:48 +0100 fs r270015 : #i100095# when the error messages contain non-trivial details (SQLState/ErrorCode), then always display the 'Details' button 2009-03-25 12:10:05 +0100 fs r270012 : renamed the previously introduced error condition 2009-03-25 12:01:04 +0100 fs r270011 : #i100095# when no address book (or respective profile) can be found, then use a dedicated ErrorCondition 2009-03-25 10:29:07 +0100 fs r270003 : add missing localization 2009-03-25 10:23:12 +0100 fs r270001 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-25 10:21:55 +0100 fs r270000 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-20 23:05:38 +0100 fs r269829 : XMap::getSize removed 2009-03-20 23:05:19 +0100 fs r269828 : changes as suggested by sb 2009-03-20 23:04:56 +0100 fs r269827 : enhanced documentation 2009-03-20 15:31:40 +0100 fs r269815 : changes as suggested by sb 2009-03-20 13:23:52 +0100 oj r269804 : #i92538# correct the zorder that fixedtext is infront of control 2009-03-20 12:59:38 +0100 oj r269801 : #i94571# paraadjust now supports BLOCK 2009-03-20 12:58:11 +0100 oj r269800 : #i94571# export style with data style 2009-03-20 12:57:05 +0100 oj r269799 : #i94571# text align is now paraadjust 2009-03-20 12:37:02 +0100 fs r269797 : enum keys only accepted if they have the exact type, not if they have *any* enum type 2009-03-20 12:28:31 +0100 fs r269794 : some changes requested by sb 2009-03-20 08:52:47 +0100 fs r269780 : doc 2009-03-20 07:37:31 +0100 oj r269779 : #i99913# only notifiy when values are different 2009-03-20 07:36:58 +0100 oj r269778 : #i99913# add undoenv as listener at the dataprovider 2009-03-19 22:52:52 +0100 fs r269771 : added comment 2009-03-19 22:40:06 +0100 fs r269770 : +testEnumeration 2009-03-19 22:39:41 +0100 fs r269769 : implemented enumeration, getKeySet, and getValues. Should be finished now. 2009-03-19 14:01:01 +0100 oj r269743 : #i99913# reset the modified state when selecting an object 2009-03-19 12:19:54 +0100 lla r269739 : #i72390# cleanups 2009-03-19 09:25:27 +0100 fs r269727 : #i10000# 2009-03-18 23:37:02 +0100 fs r269708 : extended checks for value type acceptance 2009-03-18 23:36:41 +0100 fs r269707 : fixed value type checks 2009-03-18 14:59:56 +0100 fs r269678 : initial complex test case for the new css.container.Map implementation 2009-03-18 14:59:24 +0100 fs r269677 : verifyExpectedException moved to base class (in complexlib), and renamed to assureException for consistency 2009-03-18 14:58:35 +0100 fs r269676 : removed unused imports 2009-03-18 14:58:03 +0100 fs r269675 : first implementation of the new css.container.Map service (not completed, yet) 2009-03-18 14:57:17 +0100 fs r269674 : base class for UNO components, freeing you from some repeating work 2009-03-18 14:55:53 +0100 fs r269672 : +assureException: call a given method with given parameters on a given object, ensure that a given exception is thrown by the method implementation 2009-03-18 14:54:58 +0100 fs r269671 : +getComponentContext 2009-03-18 14:54:00 +0100 fs r269670 : isEmpty returns a boolean, not a long 2009-03-18 14:14:43 +0100 oj r269663 : #i99743# now text also supports null 2009-03-18 13:54:14 +0100 oj r269660 : #i99223# remove check for 2 params 2009-03-18 13:33:35 +0100 oj r269659 : #i99060# replace text::ParagraphVertAlign with style::VerticalAlignment 2009-03-18 13:32:18 +0100 oj r269658 : #i99060# don't set void property when void isn't allowed 2009-03-18 13:31:11 +0100 oj r269657 : #i99060# handle vertical alignment 2009-03-18 13:28:28 +0100 oj r269656 : #i99060# remove unused elements from sytle 2009-03-18 09:35:42 +0100 lla r269639 : #i72390# cleanups 2009-03-18 09:31:20 +0100 lla r269638 : #i72390# add ButtonList 2009-03-18 09:30:46 +0100 lla r269637 : #i72390# renamed interface 2009-03-18 09:30:15 +0100 lla r269636 : #i72390# use ButtonList instead of ImageList 2009-03-18 09:29:05 +0100 lla r269635 : #i72390# new ButtonList, cleanups 2009-03-18 09:26:34 +0100 lla r269634 : #i72390# cleanups 2009-03-17 12:21:20 +0100 oj r269590 : #i99222# remove assertion 2009-03-17 12:17:22 +0100 oj r269589 : #i98605# impl scale mode 2009-03-17 12:10:42 +0100 oj r269588 : #i98605# impl scale mode 2009-03-17 11:40:15 +0100 oj r269584 : #i96944# doesn't create equation for shapes 2009-03-17 11:33:16 +0100 oj r269583 : #i96423# switch calc from float to long 2009-03-16 15:19:18 +0100 fs r269550 : #i41930# enable zoom for embedded/outplace documents 2009-03-16 14:25:54 +0100 oj r269542 : #i93734# remove ContextSensitive 2009-03-16 14:21:58 +0100 oj r269541 : #i99274# page header before group header 2009-03-16 14:18:23 +0100 oj r269539 : #i99110# fix value type 2009-03-16 14:14:16 +0100 fs r269537 : line ends 2009-03-16 14:11:06 +0100 fs r269535 : line ends 2009-03-16 14:08:34 +0100 fs r269534 : #i100087# (provided my np): allow for polymorphic types with more than one parameter 2009-03-16 12:30:31 +0100 oj r269521 : compile error 2009-03-16 12:19:12 +0100 oj r269519 : compile error 2009-03-16 10:39:28 +0100 oj r269511 : compile error under linux with swap 2009-03-13 10:33:04 +0100 oj r269462 : CWS-TOOLING: rebase CWS dba32b to trunk@269297 (milestone: DEV300:m43) 2009-03-12 14:37:25 +0100 fs r269416 : interface SequenceOutputStreamTest is unneeded, and pollutes the namespace here :) 2009-03-12 14:35:07 +0100 fs r269414 : not needed 2009-03-12 14:34:15 +0100 fs r269413 : preparation for multiple tests in this module 2009-03-12 14:33:02 +0100 fs r269412 : ShowTargets was moved from module integration.forms to module complexlib 2009-03-12 14:32:48 +0100 fs r269411 : helper class for projects containing multiple complex test cases (and following a certain structure) 2009-03-12 14:00:14 +0100 fs r269407 : proper module after the move 2009-03-12 13:59:10 +0100 fs r269406 : superseded by ../makefile.mk 2009-03-12 13:47:38 +0100 fs r269403 : not needed anymore 2009-03-12 13:45:46 +0100 fs r269402 : moved, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:45:07 +0100 fs r269401 : moved from ../ 2009-03-12 13:43:59 +0100 fs r269400 : moved to ./comphelper, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:29:47 +0100 oj r269396 : #i99914# set parent on dataprovider 2009-03-12 13:10:35 +0100 oj r269393 : #i99832# check thrown exception and show error 2009-03-12 13:08:10 +0100 fs r269392 : reorganizing tests 2009-03-12 12:52:55 +0100 oj r269390 : #i99118# convert formatkey in numberformat 2009-03-12 12:34:53 +0100 fs r269388 : new API tests 2009-03-12 12:29:05 +0100 fs r269386 : Map not yet committed 2009-03-12 12:28:36 +0100 fs r269385 : oops, forgot the SequenceInputStream during the previous refactoring 2009-03-12 12:12:39 +0100 oj r269384 : #i99104# set HasCategories prop 2009-03-12 12:12:08 +0100 oj r269383 : #i99104# check HasCategories even for internal dataprovider 2009-03-12 12:10:40 +0100 oj r269382 : #i99104# set HasCategories prop 2009-03-12 10:51:49 +0100 fs r269373 : #i10000# exception specifications 2009-03-12 10:49:18 +0100 fs r269372 : #i10000# exception specifications 2009-03-12 10:44:02 +0100 fs r269371 : #i10000# exception specifications 2009-03-12 10:30:55 +0100 fs r269368 : refactored the UNO service registration in this module, using the helper classes provided by comphelper itself, so you have less effort when extending the list of to-be-registered components 2009-03-12 10:30:37 +0100 fs r269367 : module-local includes 2009-03-12 07:05:54 +0100 oj r269357 : #i99104# database dataprovider doesn't need dataranges and diagramdata 2009-03-11 10:58:28 +0100 oj r269306 : #i99911# check if name of the report is a valid file name 2009-03-11 10:03:23 +0100 oj r269299 : #i99666# the report is new when the HierarchicalDocumentName is empty 2009-03-10 11:32:45 +0100 oj r269258 : #i99221# use fallback for language 2009-03-10 10:48:40 +0100 oj r269255 : #i99433# now use OStringBuffer 2009-03-10 10:36:21 +0100 fs r269252 : initial version of (X)Map 2009-03-10 09:52:23 +0100 oj r269246 : #i99433# now use OStringBuffer 2009-03-10 08:56:13 +0100 oj r269240 : #i99655# patch applied 2009-03-09 07:35:33 +0100 lla r269058 : #i10000# wrong variable assignment fixed 2009-03-06 17:20:40 +0100 fs r269030 : some explicit defaults 2009-03-06 17:20:30 +0100 fs r269029 : #i98600# 2009-03-06 14:40:34 +0100 fs r269009 : #i87692# during reload, prevent the document being modified just because of some control content changes ... 2009-03-06 12:52:20 +0100 lla r268997 : #i10000# ambigous problem with FontWeight fixed 2009-03-06 11:39:32 +0100 fs r268989 : #i10000# (approved by pl): use --without-t1-library configure option 2009-03-06 10:55:43 +0100 fs r268986 : #i99953# depends on xmlscript module now 2009-03-06 10:54:04 +0100 fs r268985 : #i99953# also adjust the event names found in dialogs embedded in the forms 2009-03-06 09:53:41 +0100 fs r268977 : #i10000# 2009-03-06 09:30:41 +0100 lla r268973 : #i10000# merge problems 2009-03-05 17:52:34 +0100 fs r268932 : #i98593# for sub components which are actually controlled by a DocumentDefinition (aka XComponentSupplier aka XCommandProcessor), close them by executing the 'close' command, not by suspending/closing the controller (which cannot be intercepted) 2009-03-05 11:41:56 +0100 fs r268889 : default the drop down line count for list/combo boxes to 20 2009-03-05 11:39:10 +0100 fs r268887 : do not display empty error messages 2009-03-02 10:13:57 +0100 lla r268639 : #i91541# CWS rebase m41 to m42 2009-03-02 09:06:27 +0100 lla r268635 : #i10000# add ';' to strings 2009-02-26 11:18:00 +0100 fs r268492 : reportdesign depends on REPORTBUILDER, not REPORTDESIGN 2009-02-26 10:11:38 +0100 lla r268489 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-26 09:04:11 +0100 lla r268488 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-24 12:09:13 +0100 lla r268392 : #i91541# #i91542# cleanups 2009-02-24 08:08:06 +0100 lla r268382 : merge all dba32a changes into dba32b 2009-02-24 07:14:55 +0100 lla r268381 : merge all dba32a changes into dba32b 2009-02-23 21:44:28 +0100 fs r268377 : oops ... don't tamper with m_aListSourceValues at the end of loadData 2009-02-23 20:57:05 +0100 fs r268376 : #i98162# don't hold the values as strings, but as ORowSetValue, this way preserving their type, and being agnostic to different result/rowset implementations doing different to-string-conversations 2009-02-23 20:55:44 +0100 fs r268375 : getObject: throwFunctionNotSupportedException, instead of silently returning NULL 2009-02-23 20:55:20 +0100 fs r268374 : #i98162# some more supported types 2009-02-23 20:54:43 +0100 fs r268373 : #i98162# +operator != 2009-02-20 09:35:39 +0100 fs r268306 : #i99422# for a font, display the font name, the style, and the size 2009-02-20 09:33:45 +0100 fs r268305 : #i99422# in the property browser, FONT supersedes CHARFONTNAME: the aggregated FormComponentHandler displays them more nicely now 2009-02-19 16:12:06 +0100 fs r268293 : #i99372# recognize DataType::FLOAT as numeric 2009-02-19 15:43:12 +0100 fs r268291 : #i99415# 2009-02-19 15:40:15 +0100 fs r268290 : #i99242# lcl_firstFocussableControl: take disabled controls into account 2009-02-19 15:34:36 +0100 fs r268289 : #i99396# properly decode the base name of the URL when using it as title 2009-02-19 15:19:05 +0100 fs r268287 : #i98247#
2009-06-05 09:47:55 +00:00
case TypeClass_ENUM:
{
sal_Int32 enumValue( 0 );
::cppu::enum2int( enumValue, _rValue );
(*this) = enumValue;
}
break;
case TypeClass_SEQUENCE:
{
Sequence<sal_Int8> aDummy;
if ( _rValue >>= aDummy )
(*this) = aDummy;
CWS-TOOLING: integrate CWS dba32b 2009-06-03 14:58:08 +0200 fs r272581 : #i102439# 2009-05-29 13:56:18 +0200 fs r272456 : remove the sub form when the 'add subform' setting changes from <true/> to <false/>, not only its controls 2009-05-29 13:19:27 +0200 fs r272454 : display '(Default)' instead of an empty string when a control has the default font 2009-05-28 20:49:18 +0200 fs r272428 : #i98162# getFirstSelectedValue: do not return reference to a temporary 2009-05-27 15:30:22 +0200 msc r272353 : #102303# 2009-05-26 13:03:06 +0200 fs r272295 : spelling 2009-05-26 12:59:54 +0200 fs r272294 : merge m48 version to get latest cygwin related fixes 2009-05-25 14:02:06 +0200 fs r272239 : remove references to local files, needed for debugging sessions only 2009-05-25 14:01:16 +0200 fs r272238 : #i102021# ensure members such as bNumberFormat are initialized before actually returning them 2009-05-25 13:10:20 +0200 fs r272236 : #i10000# reset ENABLE_EVOAB2 2009-05-22 06:44:45 +0200 oj r272167 : #i99104# add import handler for calc-sett 2009-05-22 06:42:27 +0200 oj r272166 : #i99104# impl NullDate as member 2009-05-22 06:36:22 +0200 oj r272165 : #i99104# handle nulldate from parent model 2009-05-22 06:33:13 +0200 oj r272164 : #i99104# export and import calculation-settings and nulldate 2009-05-22 06:27:50 +0200 oj r272163 : #i99104# export and import calculation-settings and nulldate 2009-05-20 13:29:58 +0200 oj r272118 : #i99911# handle invalid name 2009-05-20 13:28:49 +0200 oj r272117 : #i101261# handle different rowsets 2009-05-20 11:29:55 +0200 msc r272111 : #i100000# 2009-05-20 11:28:27 +0200 msc r272110 : merge in change from dba32a 2009-05-20 11:27:38 +0200 msc r272109 : #102082# CTRL + C does not work 2009-05-20 09:43:36 +0200 oj r272106 : clean up includes 2009-05-20 09:32:15 +0200 oj r272105 : #i99060# merge error resovled now VERTICAL_ALIGN is in both stmt 2009-05-20 08:37:21 +0200 msc r272104 : add workaround for issue #102010# 2009-05-20 08:10:12 +0200 oj r272103 : #i99104# use numberformatkey 2009-05-20 08:07:02 +0200 oj r272102 : #i99104# use column info from rowset 2009-05-20 08:04:43 +0200 oj r272101 : #i102032# use a special column type where prec and scale are the values currently set at the column 2009-05-20 08:03:04 +0200 oj r272100 : #i102032# correct type info, we have to use SQL defined type names 2009-05-19 10:27:02 +0200 oj r272061 : #i99104# export null-date 2009-05-19 08:26:53 +0200 oj r272056 : #i99104# export null-date 2009-05-18 13:15:10 +0200 msc r272014 : add issue #102019# 2009-05-18 11:33:07 +0200 msc r272005 : add issue #102019# 2009-05-18 08:59:45 +0200 msc r271996 : add workaroud for issue #102010# 2009-05-15 10:21:24 +0200 msc r271929 : #101944# 2009-05-11 21:18:30 +0200 fs r271792 : #i99914# 2009-05-08 13:52:06 +0200 oj r271715 : #i96423# remember column span 2009-05-08 11:26:19 +0200 oj r271708 : #i98605# impl new scale mode 2009-05-08 10:33:35 +0200 fs r271706 : SendUserCall: only call into the shape notification routine for UserCall types where this is necessary (performance issue) 2009-05-07 20:52:44 +0200 fs r271698 : outsource ShapeProperty from shapepropertynotifier.hxx 2009-05-07 20:43:33 +0200 fs r271697 : #i99056# use notifyShapePropertyChange, instead of getShapePropertyChangeNotifier - the latter throws if no shape exists, yet 2009-05-07 20:33:58 +0200 fs r271696 : #i99056# +notifyShapePropertyChange: allow notifying chages without checking whether there actually already exists an SvxShape 2009-05-07 16:22:15 +0200 fs r271679 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:21:37 +0200 fs r271678 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:01:11 +0200 oj r271677 : #i99914# notify parent handler 2009-05-07 15:54:54 +0200 fs r271676 : #i10000# cygwin needs some quoting 2009-05-07 14:49:48 +0200 oj r271672 : #i99277# quote alias name 2009-05-07 14:48:12 +0200 oj r271671 : #i92538# add ~ in front of type 2009-05-07 14:37:13 +0200 oj r271667 : #i99118# change type from char to varchar 2009-05-07 14:36:23 +0200 oj r271666 : #i99118# clear dest columns when changing to create new table 2009-05-07 13:35:32 +0200 oj r271657 : #i94467# handle type 0 as double as well 2009-05-07 13:20:49 +0200 oj r271655 : i99743# setNull when varchar is no text 2009-05-07 12:58:06 +0200 fs r271651 : initialize the SdrObject's property change notifier after the ctor, if necessary 2009-05-07 11:47:18 +0200 fs r271647 : #i10000# 2009-05-07 10:57:16 +0200 fs r271639 : OPropertyBrowserController::propertyChange: care for the current property state, too, and properly forward it to the UI 2009-05-07 10:18:14 +0200 fs r271636 : onNewComponent: do not ask the map for grid columns, it will throw 2009-05-07 10:09:55 +0200 fs r271634 : #i101623# 2009-05-07 09:53:44 +0200 fs r271631 : #i101622# 2009-05-06 21:55:53 +0200 fs r271615 : #i10000# 2009-05-06 21:10:42 +0200 fs r271611 : #i10000# 2009-05-06 13:11:48 +0200 fs r271583 : #i10000# 2009-05-05 22:29:31 +0200 fs r271559 : proper assertion message 2009-05-05 22:29:03 +0200 fs r271558 : diagnostics 2009-05-05 22:16:16 +0200 fs r271557 : #i10000# 2009-05-05 13:50:32 +0200 fs r271513 : #i10000# 2009-05-05 10:21:50 +0200 fs r271503 : #i10000# 2009-05-05 09:30:26 +0200 fs r271501 : why did those survive the rebase -C step? 2009-05-05 09:18:12 +0200 fs r271500 : #i10000# 2009-05-04 17:08:17 +0200 fs r271475 : CWS-TOOLING: rebase CWS dba32b to trunk@271427 (milestone: DEV300:m47) 2009-05-04 14:51:26 +0200 fs r271456 : line ends 2009-04-30 15:55:27 +0200 fs r271418 : NewURL -> PublicConnectionURL 2009-04-22 21:18:34 +0200 fs r271141 : #i100944# 2009-04-22 09:12:26 +0200 oj r271071 : #i101261# little code change 2009-04-22 09:11:43 +0200 oj r271070 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:11:25 +0200 oj r271069 : #i101261# new grabage container for nodes 2009-04-22 09:11:02 +0200 oj r271068 : #i101261# new grabage container for nodes 2009-04-22 09:10:44 +0200 oj r271067 : #i101261# new grabage container for nodes 2009-04-22 09:10:21 +0200 oj r271066 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:08:24 +0200 oj r271065 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:07:25 +0200 oj r271064 : #i101261# only ask for parameters which aren't set before 2009-04-22 08:49:07 +0200 oj r271062 : #i77501# preview only when needed 2009-04-22 08:45:44 +0200 oj r271061 : #i101261# new prop max rows 2009-04-22 08:44:18 +0200 oj r271060 : #i101261# create dataprovider earlier to avoid the wrong legend in chart 2009-04-22 08:42:48 +0200 oj r271059 : #i101261# handle parameter 2009-04-17 21:00:23 +0200 fs r270954 : #i98350# 2009-04-17 13:54:19 +0200 fs r270942 : #i99565# 2009-04-17 13:51:34 +0200 fs r270940 : #i101153# only localize the (potentially) localizable properties when there really is support at the control model 2009-04-17 11:43:14 +0200 fs r270932 : removed superfluous include 2009-04-17 10:10:15 +0200 fs r270926 : #i10000# 2009-04-17 10:02:36 +0200 fs r270925 : #i10000# 2009-04-17 09:15:13 +0200 fs r270918 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:14:56 +0200 fs r270917 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:13:25 +0200 fs r270915 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 08:30:34 +0200 fs r270914 : removed unotools/servicehelper.hxx in favour of the (duplicated) comphelper/servicehelper.hxx 2009-04-16 21:05:25 +0200 fs r270903 : #i10000# 2009-04-16 20:43:43 +0200 fs r270902 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-16 15:57:36 +0200 fs r270891 : #i99056# make SetUnoControlModel virtual 2009-04-16 15:44:02 +0200 fs r270890 : #i99056# let the ScShapeObj notify changes in its Anchor property 2009-04-16 15:36:47 +0200 fs r270889 : #i99056# enable the sheet anchor type property, too 2009-04-16 15:33:45 +0200 fs r270887 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-15 14:53:13 +0200 fs r270844 : #i10000# 2009-04-15 13:08:29 +0200 fs r270836 : #i10000# 2009-04-15 12:28:14 +0200 fs r270832 : #i10000# 2009-04-15 10:59:14 +0200 fs r270827 : #i10000# 2009-04-15 09:41:08 +0200 oj r270823 : fix issues found with findbugs and pmd 2009-04-14 21:08:04 +0200 fs r270808 : #i99056# implement SheetAnchorType - now the only thing missing to enable it is the proper notification when it is modified 2009-04-14 17:09:00 +0200 fs r270799 : #i99056# implement XServiceInfo for the ScShapeObj 2009-04-14 17:07:55 +0200 fs r270798 : #i99056# implement TextAnchorType, partially implement SheetAnchorType 2009-04-14 15:54:05 +0200 fs r270786 : #i99056# SwXShape: notify changes of the AnchorType property 2009-04-14 15:47:32 +0200 fs r270785 : #i99056# deliver shapepropertynotifier.hxx 2009-04-14 15:46:54 +0200 fs r270784 : diagnostics 2009-04-14 15:08:28 +0200 fs r270781 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:41:09 +0200 fs r270773 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:37:23 +0200 fs r270772 : in dtor, remove the properties from the temporary component context 2009-04-14 14:36:34 +0200 fs r270771 : getWeakUnoShape made const 2009-04-14 12:23:08 +0200 oj r270757 : #i101064# add missing braces 2009-04-14 12:21:25 +0200 oj r270756 : #i101065# add braces for gcc 4.3.2 2009-04-14 12:17:45 +0200 oj r270755 : #i101059# add dep for manifest 2009-04-09 12:06:58 +0200 oj r270686 : #i93100# use OptimalSize from control to get height 2009-04-08 09:56:55 +0200 oj r270619 : #i92537# handle calculations in the select columns as well 2009-04-08 09:30:39 +0200 oj r270615 : #i96657# throw error message when the key doesn't have any columns 2009-04-07 12:08:26 +0200 oj r270592 : #i77501# impl preview of the executed report 2009-04-07 12:01:56 +0200 oj r270591 : #i77501# impl preview of the executed report 2009-04-07 11:41:03 +0200 oj r270590 : #i77501# impl preview of the executed report 2009-04-07 11:39:32 +0200 oj r270589 : #i77501# impl preview of the executed report 2009-04-07 11:29:25 +0200 oj r270588 : #i77501# convert dos to unix lineends 2009-04-07 11:28:23 +0200 oj r270587 : #i77501# impl preview of the executed report 2009-04-07 11:16:50 +0200 oj r270586 : #i77501# impl preview of the executed report 2009-04-07 11:16:00 +0200 oj r270585 : #i77501# impl preview of the executed report 2009-04-07 11:15:44 +0200 oj r270584 : #i77501# impl preview of the executed report 2009-04-07 11:15:28 +0200 oj r270583 : #i77501# impl preview of the executed report 2009-04-07 11:15:04 +0200 oj r270582 : #i77501# impl preview of the executed report 2009-04-06 15:38:54 +0200 fs r270559 : merge changes from CWS dba32a herein 2009-04-03 15:56:16 +0200 fs r270494 : ImpSvMEdit::Resize: do multiple iterations, if necessary 2009-04-03 14:35:49 +0200 fs r270487 : #i10000# 2009-04-03 13:17:16 +0200 fs r270476 : #i99056# display geometry information for controls, too 2009-04-03 13:16:37 +0200 fs r270475 : #i99056# better ordering of the geometry properties 2009-04-03 13:16:07 +0200 fs r270473 : #i99056# now that SvxShape supports property change listeners, forward add/remove requests to it 2009-04-03 13:13:18 +0200 fs r270472 : #i99056# at SvxShape, allow for PropertyChangeListeners for Size/Position 2009-04-03 09:29:27 +0200 oj r270456 : #i94571# use correct prop name 2009-04-03 09:14:54 +0200 fs r270451 : merge changes from CWS dba32a herein 2009-04-02 17:00:51 +0200 fs r270424 : better diagnostics 2009-04-02 16:35:19 +0200 fs r270421 : diagnostics 2009-04-02 16:34:50 +0200 fs r270420 : #i99056# mxUnoShape not accessible anymore, use impl_setUnoShape instead 2009-04-02 16:32:48 +0200 fs r270419 : #i99056# make getUnoShape cheaper: keep the pointer to the SvxShape all the time, so there's no need to ask for it in getUnoShape. As a consequence, we will later be able to use the pointer in scenarious where performance (potentially) matters 2009-04-02 16:31:13 +0200 fs r270417 : merge changes from CWS dba32a herein 2009-04-02 16:23:16 +0200 fs r270414 : merge changes from CWS dba32a herein 2009-04-02 14:10:35 +0200 fs r270405 : #i10000# 2009-04-02 14:06:26 +0200 fs r270404 : merge changes from CWS dba32a herein 2009-04-02 14:03:03 +0200 fs r270401 : #i10000# 2009-04-02 13:58:13 +0200 fs r270400 : #i10000# 2009-04-02 12:59:44 +0200 fs r270397 : merge changes from CWS dba32a herein 2009-04-02 12:46:30 +0200 fs r270396 : #i99056# let the form page maintain a mapping between control models and control shapes 2009-04-02 12:44:07 +0200 fs r270395 : merge changes from CWS dba32a herein 2009-04-02 12:42:06 +0200 fs r270394 : merge changes from CWS dba32a herein 2009-04-02 12:35:20 +0200 fs r270393 : #i10000# precompiled header 2009-04-02 12:05:31 +0200 fs r270392 : merge changes from CWS dba32a herein 2009-04-02 12:00:42 +0200 fs r270391 : merge changes from CWS dba32a herein 2009-04-02 11:47:26 +0200 fs r270390 : merge changes from CWS dba32a herein 2009-04-02 11:39:15 +0200 oj r270389 : #i94467# foxpro impl several new types 2009-04-02 11:35:58 +0200 fs r270387 : merge changes from CWS dba32a herein 2009-04-01 14:10:51 +0200 fs r270329 : merge changes from CWS dba32a herein 2009-03-31 17:29:50 +0200 fs r270290 : merge changes from CWS dba32a herein 2009-03-30 14:53:56 +0200 fs r270233 : #i100417# don't set grid column widths to 0, but to <void/> 2009-03-30 12:31:03 +0200 oj r270213 : #i100552# wrong orb used 2009-03-30 12:19:20 +0200 oj r270212 : #i98303# convertlike corrected to sal_Unicode 2009-03-30 11:58:25 +0200 fs r270210 : merge changes from CWS dba32a herein 2009-03-30 11:38:16 +0200 oj r270205 : remove duplicate code from merge 2009-03-30 11:02:27 +0200 fs r270202 : merge changes from CWS dba32a herein 2009-03-30 11:02:19 +0200 fs r270201 : merge changes from CWS dba32a herein 2009-03-30 10:31:26 +0200 oj r270200 : #i100665# only throw exception and do not drop table 2009-03-30 09:36:24 +0200 fs r270195 : assertion text 2009-03-28 20:21:58 +0100 fs r270187 : #ii10000# 2009-03-28 20:19:54 +0100 fs r270186 : removed unused help ids 2009-03-28 20:19:40 +0100 fs r270185 : removed unused help ids 2009-03-28 20:19:10 +0100 fs r270184 : #i100237# +DefaultState/XReset 2009-03-28 00:29:29 +0100 fs r270177 : CWS-TOOLING: rebase CWS dba32b to trunk@270033 (milestone: DEV300:m45) 2009-03-27 22:56:46 +0100 fs r270173 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:55:52 +0100 fs r270172 : #i100237# DefaultState property for buttons, enabled only when Toggle=Yes 2009-03-27 22:54:15 +0100 fs r270171 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:53:54 +0100 fs r270170 : #i100237# introduce a DefaultState property for buttons, which implies buttongs supporting XReset, which needed some refactoring 2009-03-27 13:31:41 +0100 fs r270152 : ignore output paths 2009-03-27 11:23:44 +0100 fs r270139 : tuned behavior with respect to invalid keys/values 2009-03-27 09:57:14 +0100 fs r270136 : don't allow Double.NaN 2009-03-27 09:56:16 +0100 fs r270135 : talk about Double.NaN 2009-03-26 12:14:30 +0100 fs r270067 : removed unused parameter 2009-03-26 12:14:02 +0100 fs r270066 : removed widening conversion when checking keys 2009-03-26 09:17:34 +0100 fs r270053 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:17:11 +0100 fs r270052 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:16:49 +0100 fs r270051 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-25 21:55:20 +0100 fs r270044 : #i100541# properly calculat bNeedScrollBox 2009-03-25 12:56:17 +0100 fs r270019 : renamed the previously introduced error condition 2009-03-25 12:11:48 +0100 fs r270015 : #i100095# when the error messages contain non-trivial details (SQLState/ErrorCode), then always display the 'Details' button 2009-03-25 12:10:05 +0100 fs r270012 : renamed the previously introduced error condition 2009-03-25 12:01:04 +0100 fs r270011 : #i100095# when no address book (or respective profile) can be found, then use a dedicated ErrorCondition 2009-03-25 10:29:07 +0100 fs r270003 : add missing localization 2009-03-25 10:23:12 +0100 fs r270001 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-25 10:21:55 +0100 fs r270000 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-20 23:05:38 +0100 fs r269829 : XMap::getSize removed 2009-03-20 23:05:19 +0100 fs r269828 : changes as suggested by sb 2009-03-20 23:04:56 +0100 fs r269827 : enhanced documentation 2009-03-20 15:31:40 +0100 fs r269815 : changes as suggested by sb 2009-03-20 13:23:52 +0100 oj r269804 : #i92538# correct the zorder that fixedtext is infront of control 2009-03-20 12:59:38 +0100 oj r269801 : #i94571# paraadjust now supports BLOCK 2009-03-20 12:58:11 +0100 oj r269800 : #i94571# export style with data style 2009-03-20 12:57:05 +0100 oj r269799 : #i94571# text align is now paraadjust 2009-03-20 12:37:02 +0100 fs r269797 : enum keys only accepted if they have the exact type, not if they have *any* enum type 2009-03-20 12:28:31 +0100 fs r269794 : some changes requested by sb 2009-03-20 08:52:47 +0100 fs r269780 : doc 2009-03-20 07:37:31 +0100 oj r269779 : #i99913# only notifiy when values are different 2009-03-20 07:36:58 +0100 oj r269778 : #i99913# add undoenv as listener at the dataprovider 2009-03-19 22:52:52 +0100 fs r269771 : added comment 2009-03-19 22:40:06 +0100 fs r269770 : +testEnumeration 2009-03-19 22:39:41 +0100 fs r269769 : implemented enumeration, getKeySet, and getValues. Should be finished now. 2009-03-19 14:01:01 +0100 oj r269743 : #i99913# reset the modified state when selecting an object 2009-03-19 12:19:54 +0100 lla r269739 : #i72390# cleanups 2009-03-19 09:25:27 +0100 fs r269727 : #i10000# 2009-03-18 23:37:02 +0100 fs r269708 : extended checks for value type acceptance 2009-03-18 23:36:41 +0100 fs r269707 : fixed value type checks 2009-03-18 14:59:56 +0100 fs r269678 : initial complex test case for the new css.container.Map implementation 2009-03-18 14:59:24 +0100 fs r269677 : verifyExpectedException moved to base class (in complexlib), and renamed to assureException for consistency 2009-03-18 14:58:35 +0100 fs r269676 : removed unused imports 2009-03-18 14:58:03 +0100 fs r269675 : first implementation of the new css.container.Map service (not completed, yet) 2009-03-18 14:57:17 +0100 fs r269674 : base class for UNO components, freeing you from some repeating work 2009-03-18 14:55:53 +0100 fs r269672 : +assureException: call a given method with given parameters on a given object, ensure that a given exception is thrown by the method implementation 2009-03-18 14:54:58 +0100 fs r269671 : +getComponentContext 2009-03-18 14:54:00 +0100 fs r269670 : isEmpty returns a boolean, not a long 2009-03-18 14:14:43 +0100 oj r269663 : #i99743# now text also supports null 2009-03-18 13:54:14 +0100 oj r269660 : #i99223# remove check for 2 params 2009-03-18 13:33:35 +0100 oj r269659 : #i99060# replace text::ParagraphVertAlign with style::VerticalAlignment 2009-03-18 13:32:18 +0100 oj r269658 : #i99060# don't set void property when void isn't allowed 2009-03-18 13:31:11 +0100 oj r269657 : #i99060# handle vertical alignment 2009-03-18 13:28:28 +0100 oj r269656 : #i99060# remove unused elements from sytle 2009-03-18 09:35:42 +0100 lla r269639 : #i72390# cleanups 2009-03-18 09:31:20 +0100 lla r269638 : #i72390# add ButtonList 2009-03-18 09:30:46 +0100 lla r269637 : #i72390# renamed interface 2009-03-18 09:30:15 +0100 lla r269636 : #i72390# use ButtonList instead of ImageList 2009-03-18 09:29:05 +0100 lla r269635 : #i72390# new ButtonList, cleanups 2009-03-18 09:26:34 +0100 lla r269634 : #i72390# cleanups 2009-03-17 12:21:20 +0100 oj r269590 : #i99222# remove assertion 2009-03-17 12:17:22 +0100 oj r269589 : #i98605# impl scale mode 2009-03-17 12:10:42 +0100 oj r269588 : #i98605# impl scale mode 2009-03-17 11:40:15 +0100 oj r269584 : #i96944# doesn't create equation for shapes 2009-03-17 11:33:16 +0100 oj r269583 : #i96423# switch calc from float to long 2009-03-16 15:19:18 +0100 fs r269550 : #i41930# enable zoom for embedded/outplace documents 2009-03-16 14:25:54 +0100 oj r269542 : #i93734# remove ContextSensitive 2009-03-16 14:21:58 +0100 oj r269541 : #i99274# page header before group header 2009-03-16 14:18:23 +0100 oj r269539 : #i99110# fix value type 2009-03-16 14:14:16 +0100 fs r269537 : line ends 2009-03-16 14:11:06 +0100 fs r269535 : line ends 2009-03-16 14:08:34 +0100 fs r269534 : #i100087# (provided my np): allow for polymorphic types with more than one parameter 2009-03-16 12:30:31 +0100 oj r269521 : compile error 2009-03-16 12:19:12 +0100 oj r269519 : compile error 2009-03-16 10:39:28 +0100 oj r269511 : compile error under linux with swap 2009-03-13 10:33:04 +0100 oj r269462 : CWS-TOOLING: rebase CWS dba32b to trunk@269297 (milestone: DEV300:m43) 2009-03-12 14:37:25 +0100 fs r269416 : interface SequenceOutputStreamTest is unneeded, and pollutes the namespace here :) 2009-03-12 14:35:07 +0100 fs r269414 : not needed 2009-03-12 14:34:15 +0100 fs r269413 : preparation for multiple tests in this module 2009-03-12 14:33:02 +0100 fs r269412 : ShowTargets was moved from module integration.forms to module complexlib 2009-03-12 14:32:48 +0100 fs r269411 : helper class for projects containing multiple complex test cases (and following a certain structure) 2009-03-12 14:00:14 +0100 fs r269407 : proper module after the move 2009-03-12 13:59:10 +0100 fs r269406 : superseded by ../makefile.mk 2009-03-12 13:47:38 +0100 fs r269403 : not needed anymore 2009-03-12 13:45:46 +0100 fs r269402 : moved, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:45:07 +0100 fs r269401 : moved from ../ 2009-03-12 13:43:59 +0100 fs r269400 : moved to ./comphelper, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:29:47 +0100 oj r269396 : #i99914# set parent on dataprovider 2009-03-12 13:10:35 +0100 oj r269393 : #i99832# check thrown exception and show error 2009-03-12 13:08:10 +0100 fs r269392 : reorganizing tests 2009-03-12 12:52:55 +0100 oj r269390 : #i99118# convert formatkey in numberformat 2009-03-12 12:34:53 +0100 fs r269388 : new API tests 2009-03-12 12:29:05 +0100 fs r269386 : Map not yet committed 2009-03-12 12:28:36 +0100 fs r269385 : oops, forgot the SequenceInputStream during the previous refactoring 2009-03-12 12:12:39 +0100 oj r269384 : #i99104# set HasCategories prop 2009-03-12 12:12:08 +0100 oj r269383 : #i99104# check HasCategories even for internal dataprovider 2009-03-12 12:10:40 +0100 oj r269382 : #i99104# set HasCategories prop 2009-03-12 10:51:49 +0100 fs r269373 : #i10000# exception specifications 2009-03-12 10:49:18 +0100 fs r269372 : #i10000# exception specifications 2009-03-12 10:44:02 +0100 fs r269371 : #i10000# exception specifications 2009-03-12 10:30:55 +0100 fs r269368 : refactored the UNO service registration in this module, using the helper classes provided by comphelper itself, so you have less effort when extending the list of to-be-registered components 2009-03-12 10:30:37 +0100 fs r269367 : module-local includes 2009-03-12 07:05:54 +0100 oj r269357 : #i99104# database dataprovider doesn't need dataranges and diagramdata 2009-03-11 10:58:28 +0100 oj r269306 : #i99911# check if name of the report is a valid file name 2009-03-11 10:03:23 +0100 oj r269299 : #i99666# the report is new when the HierarchicalDocumentName is empty 2009-03-10 11:32:45 +0100 oj r269258 : #i99221# use fallback for language 2009-03-10 10:48:40 +0100 oj r269255 : #i99433# now use OStringBuffer 2009-03-10 10:36:21 +0100 fs r269252 : initial version of (X)Map 2009-03-10 09:52:23 +0100 oj r269246 : #i99433# now use OStringBuffer 2009-03-10 08:56:13 +0100 oj r269240 : #i99655# patch applied 2009-03-09 07:35:33 +0100 lla r269058 : #i10000# wrong variable assignment fixed 2009-03-06 17:20:40 +0100 fs r269030 : some explicit defaults 2009-03-06 17:20:30 +0100 fs r269029 : #i98600# 2009-03-06 14:40:34 +0100 fs r269009 : #i87692# during reload, prevent the document being modified just because of some control content changes ... 2009-03-06 12:52:20 +0100 lla r268997 : #i10000# ambigous problem with FontWeight fixed 2009-03-06 11:39:32 +0100 fs r268989 : #i10000# (approved by pl): use --without-t1-library configure option 2009-03-06 10:55:43 +0100 fs r268986 : #i99953# depends on xmlscript module now 2009-03-06 10:54:04 +0100 fs r268985 : #i99953# also adjust the event names found in dialogs embedded in the forms 2009-03-06 09:53:41 +0100 fs r268977 : #i10000# 2009-03-06 09:30:41 +0100 lla r268973 : #i10000# merge problems 2009-03-05 17:52:34 +0100 fs r268932 : #i98593# for sub components which are actually controlled by a DocumentDefinition (aka XComponentSupplier aka XCommandProcessor), close them by executing the 'close' command, not by suspending/closing the controller (which cannot be intercepted) 2009-03-05 11:41:56 +0100 fs r268889 : default the drop down line count for list/combo boxes to 20 2009-03-05 11:39:10 +0100 fs r268887 : do not display empty error messages 2009-03-02 10:13:57 +0100 lla r268639 : #i91541# CWS rebase m41 to m42 2009-03-02 09:06:27 +0100 lla r268635 : #i10000# add ';' to strings 2009-02-26 11:18:00 +0100 fs r268492 : reportdesign depends on REPORTBUILDER, not REPORTDESIGN 2009-02-26 10:11:38 +0100 lla r268489 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-26 09:04:11 +0100 lla r268488 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-24 12:09:13 +0100 lla r268392 : #i91541# #i91542# cleanups 2009-02-24 08:08:06 +0100 lla r268382 : merge all dba32a changes into dba32b 2009-02-24 07:14:55 +0100 lla r268381 : merge all dba32a changes into dba32b 2009-02-23 21:44:28 +0100 fs r268377 : oops ... don't tamper with m_aListSourceValues at the end of loadData 2009-02-23 20:57:05 +0100 fs r268376 : #i98162# don't hold the values as strings, but as ORowSetValue, this way preserving their type, and being agnostic to different result/rowset implementations doing different to-string-conversations 2009-02-23 20:55:44 +0100 fs r268375 : getObject: throwFunctionNotSupportedException, instead of silently returning NULL 2009-02-23 20:55:20 +0100 fs r268374 : #i98162# some more supported types 2009-02-23 20:54:43 +0100 fs r268373 : #i98162# +operator != 2009-02-20 09:35:39 +0100 fs r268306 : #i99422# for a font, display the font name, the style, and the size 2009-02-20 09:33:45 +0100 fs r268305 : #i99422# in the property browser, FONT supersedes CHARFONTNAME: the aggregated FormComponentHandler displays them more nicely now 2009-02-19 16:12:06 +0100 fs r268293 : #i99372# recognize DataType::FLOAT as numeric 2009-02-19 15:43:12 +0100 fs r268291 : #i99415# 2009-02-19 15:40:15 +0100 fs r268290 : #i99242# lcl_firstFocussableControl: take disabled controls into account 2009-02-19 15:34:36 +0100 fs r268289 : #i99396# properly decode the base name of the URL when using it as title 2009-02-19 15:19:05 +0100 fs r268287 : #i98247#
2009-06-05 09:47:55 +00:00
else
SAL_WARN( "connectivity.commontools", "ORowSetValue::fill: unsupported sequence type!" );
CWS-TOOLING: integrate CWS dba32b 2009-06-03 14:58:08 +0200 fs r272581 : #i102439# 2009-05-29 13:56:18 +0200 fs r272456 : remove the sub form when the 'add subform' setting changes from <true/> to <false/>, not only its controls 2009-05-29 13:19:27 +0200 fs r272454 : display '(Default)' instead of an empty string when a control has the default font 2009-05-28 20:49:18 +0200 fs r272428 : #i98162# getFirstSelectedValue: do not return reference to a temporary 2009-05-27 15:30:22 +0200 msc r272353 : #102303# 2009-05-26 13:03:06 +0200 fs r272295 : spelling 2009-05-26 12:59:54 +0200 fs r272294 : merge m48 version to get latest cygwin related fixes 2009-05-25 14:02:06 +0200 fs r272239 : remove references to local files, needed for debugging sessions only 2009-05-25 14:01:16 +0200 fs r272238 : #i102021# ensure members such as bNumberFormat are initialized before actually returning them 2009-05-25 13:10:20 +0200 fs r272236 : #i10000# reset ENABLE_EVOAB2 2009-05-22 06:44:45 +0200 oj r272167 : #i99104# add import handler for calc-sett 2009-05-22 06:42:27 +0200 oj r272166 : #i99104# impl NullDate as member 2009-05-22 06:36:22 +0200 oj r272165 : #i99104# handle nulldate from parent model 2009-05-22 06:33:13 +0200 oj r272164 : #i99104# export and import calculation-settings and nulldate 2009-05-22 06:27:50 +0200 oj r272163 : #i99104# export and import calculation-settings and nulldate 2009-05-20 13:29:58 +0200 oj r272118 : #i99911# handle invalid name 2009-05-20 13:28:49 +0200 oj r272117 : #i101261# handle different rowsets 2009-05-20 11:29:55 +0200 msc r272111 : #i100000# 2009-05-20 11:28:27 +0200 msc r272110 : merge in change from dba32a 2009-05-20 11:27:38 +0200 msc r272109 : #102082# CTRL + C does not work 2009-05-20 09:43:36 +0200 oj r272106 : clean up includes 2009-05-20 09:32:15 +0200 oj r272105 : #i99060# merge error resovled now VERTICAL_ALIGN is in both stmt 2009-05-20 08:37:21 +0200 msc r272104 : add workaround for issue #102010# 2009-05-20 08:10:12 +0200 oj r272103 : #i99104# use numberformatkey 2009-05-20 08:07:02 +0200 oj r272102 : #i99104# use column info from rowset 2009-05-20 08:04:43 +0200 oj r272101 : #i102032# use a special column type where prec and scale are the values currently set at the column 2009-05-20 08:03:04 +0200 oj r272100 : #i102032# correct type info, we have to use SQL defined type names 2009-05-19 10:27:02 +0200 oj r272061 : #i99104# export null-date 2009-05-19 08:26:53 +0200 oj r272056 : #i99104# export null-date 2009-05-18 13:15:10 +0200 msc r272014 : add issue #102019# 2009-05-18 11:33:07 +0200 msc r272005 : add issue #102019# 2009-05-18 08:59:45 +0200 msc r271996 : add workaroud for issue #102010# 2009-05-15 10:21:24 +0200 msc r271929 : #101944# 2009-05-11 21:18:30 +0200 fs r271792 : #i99914# 2009-05-08 13:52:06 +0200 oj r271715 : #i96423# remember column span 2009-05-08 11:26:19 +0200 oj r271708 : #i98605# impl new scale mode 2009-05-08 10:33:35 +0200 fs r271706 : SendUserCall: only call into the shape notification routine for UserCall types where this is necessary (performance issue) 2009-05-07 20:52:44 +0200 fs r271698 : outsource ShapeProperty from shapepropertynotifier.hxx 2009-05-07 20:43:33 +0200 fs r271697 : #i99056# use notifyShapePropertyChange, instead of getShapePropertyChangeNotifier - the latter throws if no shape exists, yet 2009-05-07 20:33:58 +0200 fs r271696 : #i99056# +notifyShapePropertyChange: allow notifying chages without checking whether there actually already exists an SvxShape 2009-05-07 16:22:15 +0200 fs r271679 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:21:37 +0200 fs r271678 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:01:11 +0200 oj r271677 : #i99914# notify parent handler 2009-05-07 15:54:54 +0200 fs r271676 : #i10000# cygwin needs some quoting 2009-05-07 14:49:48 +0200 oj r271672 : #i99277# quote alias name 2009-05-07 14:48:12 +0200 oj r271671 : #i92538# add ~ in front of type 2009-05-07 14:37:13 +0200 oj r271667 : #i99118# change type from char to varchar 2009-05-07 14:36:23 +0200 oj r271666 : #i99118# clear dest columns when changing to create new table 2009-05-07 13:35:32 +0200 oj r271657 : #i94467# handle type 0 as double as well 2009-05-07 13:20:49 +0200 oj r271655 : i99743# setNull when varchar is no text 2009-05-07 12:58:06 +0200 fs r271651 : initialize the SdrObject's property change notifier after the ctor, if necessary 2009-05-07 11:47:18 +0200 fs r271647 : #i10000# 2009-05-07 10:57:16 +0200 fs r271639 : OPropertyBrowserController::propertyChange: care for the current property state, too, and properly forward it to the UI 2009-05-07 10:18:14 +0200 fs r271636 : onNewComponent: do not ask the map for grid columns, it will throw 2009-05-07 10:09:55 +0200 fs r271634 : #i101623# 2009-05-07 09:53:44 +0200 fs r271631 : #i101622# 2009-05-06 21:55:53 +0200 fs r271615 : #i10000# 2009-05-06 21:10:42 +0200 fs r271611 : #i10000# 2009-05-06 13:11:48 +0200 fs r271583 : #i10000# 2009-05-05 22:29:31 +0200 fs r271559 : proper assertion message 2009-05-05 22:29:03 +0200 fs r271558 : diagnostics 2009-05-05 22:16:16 +0200 fs r271557 : #i10000# 2009-05-05 13:50:32 +0200 fs r271513 : #i10000# 2009-05-05 10:21:50 +0200 fs r271503 : #i10000# 2009-05-05 09:30:26 +0200 fs r271501 : why did those survive the rebase -C step? 2009-05-05 09:18:12 +0200 fs r271500 : #i10000# 2009-05-04 17:08:17 +0200 fs r271475 : CWS-TOOLING: rebase CWS dba32b to trunk@271427 (milestone: DEV300:m47) 2009-05-04 14:51:26 +0200 fs r271456 : line ends 2009-04-30 15:55:27 +0200 fs r271418 : NewURL -> PublicConnectionURL 2009-04-22 21:18:34 +0200 fs r271141 : #i100944# 2009-04-22 09:12:26 +0200 oj r271071 : #i101261# little code change 2009-04-22 09:11:43 +0200 oj r271070 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:11:25 +0200 oj r271069 : #i101261# new grabage container for nodes 2009-04-22 09:11:02 +0200 oj r271068 : #i101261# new grabage container for nodes 2009-04-22 09:10:44 +0200 oj r271067 : #i101261# new grabage container for nodes 2009-04-22 09:10:21 +0200 oj r271066 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:08:24 +0200 oj r271065 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:07:25 +0200 oj r271064 : #i101261# only ask for parameters which aren't set before 2009-04-22 08:49:07 +0200 oj r271062 : #i77501# preview only when needed 2009-04-22 08:45:44 +0200 oj r271061 : #i101261# new prop max rows 2009-04-22 08:44:18 +0200 oj r271060 : #i101261# create dataprovider earlier to avoid the wrong legend in chart 2009-04-22 08:42:48 +0200 oj r271059 : #i101261# handle parameter 2009-04-17 21:00:23 +0200 fs r270954 : #i98350# 2009-04-17 13:54:19 +0200 fs r270942 : #i99565# 2009-04-17 13:51:34 +0200 fs r270940 : #i101153# only localize the (potentially) localizable properties when there really is support at the control model 2009-04-17 11:43:14 +0200 fs r270932 : removed superfluous include 2009-04-17 10:10:15 +0200 fs r270926 : #i10000# 2009-04-17 10:02:36 +0200 fs r270925 : #i10000# 2009-04-17 09:15:13 +0200 fs r270918 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:14:56 +0200 fs r270917 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:13:25 +0200 fs r270915 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 08:30:34 +0200 fs r270914 : removed unotools/servicehelper.hxx in favour of the (duplicated) comphelper/servicehelper.hxx 2009-04-16 21:05:25 +0200 fs r270903 : #i10000# 2009-04-16 20:43:43 +0200 fs r270902 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-16 15:57:36 +0200 fs r270891 : #i99056# make SetUnoControlModel virtual 2009-04-16 15:44:02 +0200 fs r270890 : #i99056# let the ScShapeObj notify changes in its Anchor property 2009-04-16 15:36:47 +0200 fs r270889 : #i99056# enable the sheet anchor type property, too 2009-04-16 15:33:45 +0200 fs r270887 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-15 14:53:13 +0200 fs r270844 : #i10000# 2009-04-15 13:08:29 +0200 fs r270836 : #i10000# 2009-04-15 12:28:14 +0200 fs r270832 : #i10000# 2009-04-15 10:59:14 +0200 fs r270827 : #i10000# 2009-04-15 09:41:08 +0200 oj r270823 : fix issues found with findbugs and pmd 2009-04-14 21:08:04 +0200 fs r270808 : #i99056# implement SheetAnchorType - now the only thing missing to enable it is the proper notification when it is modified 2009-04-14 17:09:00 +0200 fs r270799 : #i99056# implement XServiceInfo for the ScShapeObj 2009-04-14 17:07:55 +0200 fs r270798 : #i99056# implement TextAnchorType, partially implement SheetAnchorType 2009-04-14 15:54:05 +0200 fs r270786 : #i99056# SwXShape: notify changes of the AnchorType property 2009-04-14 15:47:32 +0200 fs r270785 : #i99056# deliver shapepropertynotifier.hxx 2009-04-14 15:46:54 +0200 fs r270784 : diagnostics 2009-04-14 15:08:28 +0200 fs r270781 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:41:09 +0200 fs r270773 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:37:23 +0200 fs r270772 : in dtor, remove the properties from the temporary component context 2009-04-14 14:36:34 +0200 fs r270771 : getWeakUnoShape made const 2009-04-14 12:23:08 +0200 oj r270757 : #i101064# add missing braces 2009-04-14 12:21:25 +0200 oj r270756 : #i101065# add braces for gcc 4.3.2 2009-04-14 12:17:45 +0200 oj r270755 : #i101059# add dep for manifest 2009-04-09 12:06:58 +0200 oj r270686 : #i93100# use OptimalSize from control to get height 2009-04-08 09:56:55 +0200 oj r270619 : #i92537# handle calculations in the select columns as well 2009-04-08 09:30:39 +0200 oj r270615 : #i96657# throw error message when the key doesn't have any columns 2009-04-07 12:08:26 +0200 oj r270592 : #i77501# impl preview of the executed report 2009-04-07 12:01:56 +0200 oj r270591 : #i77501# impl preview of the executed report 2009-04-07 11:41:03 +0200 oj r270590 : #i77501# impl preview of the executed report 2009-04-07 11:39:32 +0200 oj r270589 : #i77501# impl preview of the executed report 2009-04-07 11:29:25 +0200 oj r270588 : #i77501# convert dos to unix lineends 2009-04-07 11:28:23 +0200 oj r270587 : #i77501# impl preview of the executed report 2009-04-07 11:16:50 +0200 oj r270586 : #i77501# impl preview of the executed report 2009-04-07 11:16:00 +0200 oj r270585 : #i77501# impl preview of the executed report 2009-04-07 11:15:44 +0200 oj r270584 : #i77501# impl preview of the executed report 2009-04-07 11:15:28 +0200 oj r270583 : #i77501# impl preview of the executed report 2009-04-07 11:15:04 +0200 oj r270582 : #i77501# impl preview of the executed report 2009-04-06 15:38:54 +0200 fs r270559 : merge changes from CWS dba32a herein 2009-04-03 15:56:16 +0200 fs r270494 : ImpSvMEdit::Resize: do multiple iterations, if necessary 2009-04-03 14:35:49 +0200 fs r270487 : #i10000# 2009-04-03 13:17:16 +0200 fs r270476 : #i99056# display geometry information for controls, too 2009-04-03 13:16:37 +0200 fs r270475 : #i99056# better ordering of the geometry properties 2009-04-03 13:16:07 +0200 fs r270473 : #i99056# now that SvxShape supports property change listeners, forward add/remove requests to it 2009-04-03 13:13:18 +0200 fs r270472 : #i99056# at SvxShape, allow for PropertyChangeListeners for Size/Position 2009-04-03 09:29:27 +0200 oj r270456 : #i94571# use correct prop name 2009-04-03 09:14:54 +0200 fs r270451 : merge changes from CWS dba32a herein 2009-04-02 17:00:51 +0200 fs r270424 : better diagnostics 2009-04-02 16:35:19 +0200 fs r270421 : diagnostics 2009-04-02 16:34:50 +0200 fs r270420 : #i99056# mxUnoShape not accessible anymore, use impl_setUnoShape instead 2009-04-02 16:32:48 +0200 fs r270419 : #i99056# make getUnoShape cheaper: keep the pointer to the SvxShape all the time, so there's no need to ask for it in getUnoShape. As a consequence, we will later be able to use the pointer in scenarious where performance (potentially) matters 2009-04-02 16:31:13 +0200 fs r270417 : merge changes from CWS dba32a herein 2009-04-02 16:23:16 +0200 fs r270414 : merge changes from CWS dba32a herein 2009-04-02 14:10:35 +0200 fs r270405 : #i10000# 2009-04-02 14:06:26 +0200 fs r270404 : merge changes from CWS dba32a herein 2009-04-02 14:03:03 +0200 fs r270401 : #i10000# 2009-04-02 13:58:13 +0200 fs r270400 : #i10000# 2009-04-02 12:59:44 +0200 fs r270397 : merge changes from CWS dba32a herein 2009-04-02 12:46:30 +0200 fs r270396 : #i99056# let the form page maintain a mapping between control models and control shapes 2009-04-02 12:44:07 +0200 fs r270395 : merge changes from CWS dba32a herein 2009-04-02 12:42:06 +0200 fs r270394 : merge changes from CWS dba32a herein 2009-04-02 12:35:20 +0200 fs r270393 : #i10000# precompiled header 2009-04-02 12:05:31 +0200 fs r270392 : merge changes from CWS dba32a herein 2009-04-02 12:00:42 +0200 fs r270391 : merge changes from CWS dba32a herein 2009-04-02 11:47:26 +0200 fs r270390 : merge changes from CWS dba32a herein 2009-04-02 11:39:15 +0200 oj r270389 : #i94467# foxpro impl several new types 2009-04-02 11:35:58 +0200 fs r270387 : merge changes from CWS dba32a herein 2009-04-01 14:10:51 +0200 fs r270329 : merge changes from CWS dba32a herein 2009-03-31 17:29:50 +0200 fs r270290 : merge changes from CWS dba32a herein 2009-03-30 14:53:56 +0200 fs r270233 : #i100417# don't set grid column widths to 0, but to <void/> 2009-03-30 12:31:03 +0200 oj r270213 : #i100552# wrong orb used 2009-03-30 12:19:20 +0200 oj r270212 : #i98303# convertlike corrected to sal_Unicode 2009-03-30 11:58:25 +0200 fs r270210 : merge changes from CWS dba32a herein 2009-03-30 11:38:16 +0200 oj r270205 : remove duplicate code from merge 2009-03-30 11:02:27 +0200 fs r270202 : merge changes from CWS dba32a herein 2009-03-30 11:02:19 +0200 fs r270201 : merge changes from CWS dba32a herein 2009-03-30 10:31:26 +0200 oj r270200 : #i100665# only throw exception and do not drop table 2009-03-30 09:36:24 +0200 fs r270195 : assertion text 2009-03-28 20:21:58 +0100 fs r270187 : #ii10000# 2009-03-28 20:19:54 +0100 fs r270186 : removed unused help ids 2009-03-28 20:19:40 +0100 fs r270185 : removed unused help ids 2009-03-28 20:19:10 +0100 fs r270184 : #i100237# +DefaultState/XReset 2009-03-28 00:29:29 +0100 fs r270177 : CWS-TOOLING: rebase CWS dba32b to trunk@270033 (milestone: DEV300:m45) 2009-03-27 22:56:46 +0100 fs r270173 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:55:52 +0100 fs r270172 : #i100237# DefaultState property for buttons, enabled only when Toggle=Yes 2009-03-27 22:54:15 +0100 fs r270171 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:53:54 +0100 fs r270170 : #i100237# introduce a DefaultState property for buttons, which implies buttongs supporting XReset, which needed some refactoring 2009-03-27 13:31:41 +0100 fs r270152 : ignore output paths 2009-03-27 11:23:44 +0100 fs r270139 : tuned behavior with respect to invalid keys/values 2009-03-27 09:57:14 +0100 fs r270136 : don't allow Double.NaN 2009-03-27 09:56:16 +0100 fs r270135 : talk about Double.NaN 2009-03-26 12:14:30 +0100 fs r270067 : removed unused parameter 2009-03-26 12:14:02 +0100 fs r270066 : removed widening conversion when checking keys 2009-03-26 09:17:34 +0100 fs r270053 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:17:11 +0100 fs r270052 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:16:49 +0100 fs r270051 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-25 21:55:20 +0100 fs r270044 : #i100541# properly calculat bNeedScrollBox 2009-03-25 12:56:17 +0100 fs r270019 : renamed the previously introduced error condition 2009-03-25 12:11:48 +0100 fs r270015 : #i100095# when the error messages contain non-trivial details (SQLState/ErrorCode), then always display the 'Details' button 2009-03-25 12:10:05 +0100 fs r270012 : renamed the previously introduced error condition 2009-03-25 12:01:04 +0100 fs r270011 : #i100095# when no address book (or respective profile) can be found, then use a dedicated ErrorCondition 2009-03-25 10:29:07 +0100 fs r270003 : add missing localization 2009-03-25 10:23:12 +0100 fs r270001 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-25 10:21:55 +0100 fs r270000 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-20 23:05:38 +0100 fs r269829 : XMap::getSize removed 2009-03-20 23:05:19 +0100 fs r269828 : changes as suggested by sb 2009-03-20 23:04:56 +0100 fs r269827 : enhanced documentation 2009-03-20 15:31:40 +0100 fs r269815 : changes as suggested by sb 2009-03-20 13:23:52 +0100 oj r269804 : #i92538# correct the zorder that fixedtext is infront of control 2009-03-20 12:59:38 +0100 oj r269801 : #i94571# paraadjust now supports BLOCK 2009-03-20 12:58:11 +0100 oj r269800 : #i94571# export style with data style 2009-03-20 12:57:05 +0100 oj r269799 : #i94571# text align is now paraadjust 2009-03-20 12:37:02 +0100 fs r269797 : enum keys only accepted if they have the exact type, not if they have *any* enum type 2009-03-20 12:28:31 +0100 fs r269794 : some changes requested by sb 2009-03-20 08:52:47 +0100 fs r269780 : doc 2009-03-20 07:37:31 +0100 oj r269779 : #i99913# only notifiy when values are different 2009-03-20 07:36:58 +0100 oj r269778 : #i99913# add undoenv as listener at the dataprovider 2009-03-19 22:52:52 +0100 fs r269771 : added comment 2009-03-19 22:40:06 +0100 fs r269770 : +testEnumeration 2009-03-19 22:39:41 +0100 fs r269769 : implemented enumeration, getKeySet, and getValues. Should be finished now. 2009-03-19 14:01:01 +0100 oj r269743 : #i99913# reset the modified state when selecting an object 2009-03-19 12:19:54 +0100 lla r269739 : #i72390# cleanups 2009-03-19 09:25:27 +0100 fs r269727 : #i10000# 2009-03-18 23:37:02 +0100 fs r269708 : extended checks for value type acceptance 2009-03-18 23:36:41 +0100 fs r269707 : fixed value type checks 2009-03-18 14:59:56 +0100 fs r269678 : initial complex test case for the new css.container.Map implementation 2009-03-18 14:59:24 +0100 fs r269677 : verifyExpectedException moved to base class (in complexlib), and renamed to assureException for consistency 2009-03-18 14:58:35 +0100 fs r269676 : removed unused imports 2009-03-18 14:58:03 +0100 fs r269675 : first implementation of the new css.container.Map service (not completed, yet) 2009-03-18 14:57:17 +0100 fs r269674 : base class for UNO components, freeing you from some repeating work 2009-03-18 14:55:53 +0100 fs r269672 : +assureException: call a given method with given parameters on a given object, ensure that a given exception is thrown by the method implementation 2009-03-18 14:54:58 +0100 fs r269671 : +getComponentContext 2009-03-18 14:54:00 +0100 fs r269670 : isEmpty returns a boolean, not a long 2009-03-18 14:14:43 +0100 oj r269663 : #i99743# now text also supports null 2009-03-18 13:54:14 +0100 oj r269660 : #i99223# remove check for 2 params 2009-03-18 13:33:35 +0100 oj r269659 : #i99060# replace text::ParagraphVertAlign with style::VerticalAlignment 2009-03-18 13:32:18 +0100 oj r269658 : #i99060# don't set void property when void isn't allowed 2009-03-18 13:31:11 +0100 oj r269657 : #i99060# handle vertical alignment 2009-03-18 13:28:28 +0100 oj r269656 : #i99060# remove unused elements from sytle 2009-03-18 09:35:42 +0100 lla r269639 : #i72390# cleanups 2009-03-18 09:31:20 +0100 lla r269638 : #i72390# add ButtonList 2009-03-18 09:30:46 +0100 lla r269637 : #i72390# renamed interface 2009-03-18 09:30:15 +0100 lla r269636 : #i72390# use ButtonList instead of ImageList 2009-03-18 09:29:05 +0100 lla r269635 : #i72390# new ButtonList, cleanups 2009-03-18 09:26:34 +0100 lla r269634 : #i72390# cleanups 2009-03-17 12:21:20 +0100 oj r269590 : #i99222# remove assertion 2009-03-17 12:17:22 +0100 oj r269589 : #i98605# impl scale mode 2009-03-17 12:10:42 +0100 oj r269588 : #i98605# impl scale mode 2009-03-17 11:40:15 +0100 oj r269584 : #i96944# doesn't create equation for shapes 2009-03-17 11:33:16 +0100 oj r269583 : #i96423# switch calc from float to long 2009-03-16 15:19:18 +0100 fs r269550 : #i41930# enable zoom for embedded/outplace documents 2009-03-16 14:25:54 +0100 oj r269542 : #i93734# remove ContextSensitive 2009-03-16 14:21:58 +0100 oj r269541 : #i99274# page header before group header 2009-03-16 14:18:23 +0100 oj r269539 : #i99110# fix value type 2009-03-16 14:14:16 +0100 fs r269537 : line ends 2009-03-16 14:11:06 +0100 fs r269535 : line ends 2009-03-16 14:08:34 +0100 fs r269534 : #i100087# (provided my np): allow for polymorphic types with more than one parameter 2009-03-16 12:30:31 +0100 oj r269521 : compile error 2009-03-16 12:19:12 +0100 oj r269519 : compile error 2009-03-16 10:39:28 +0100 oj r269511 : compile error under linux with swap 2009-03-13 10:33:04 +0100 oj r269462 : CWS-TOOLING: rebase CWS dba32b to trunk@269297 (milestone: DEV300:m43) 2009-03-12 14:37:25 +0100 fs r269416 : interface SequenceOutputStreamTest is unneeded, and pollutes the namespace here :) 2009-03-12 14:35:07 +0100 fs r269414 : not needed 2009-03-12 14:34:15 +0100 fs r269413 : preparation for multiple tests in this module 2009-03-12 14:33:02 +0100 fs r269412 : ShowTargets was moved from module integration.forms to module complexlib 2009-03-12 14:32:48 +0100 fs r269411 : helper class for projects containing multiple complex test cases (and following a certain structure) 2009-03-12 14:00:14 +0100 fs r269407 : proper module after the move 2009-03-12 13:59:10 +0100 fs r269406 : superseded by ../makefile.mk 2009-03-12 13:47:38 +0100 fs r269403 : not needed anymore 2009-03-12 13:45:46 +0100 fs r269402 : moved, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:45:07 +0100 fs r269401 : moved from ../ 2009-03-12 13:43:59 +0100 fs r269400 : moved to ./comphelper, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:29:47 +0100 oj r269396 : #i99914# set parent on dataprovider 2009-03-12 13:10:35 +0100 oj r269393 : #i99832# check thrown exception and show error 2009-03-12 13:08:10 +0100 fs r269392 : reorganizing tests 2009-03-12 12:52:55 +0100 oj r269390 : #i99118# convert formatkey in numberformat 2009-03-12 12:34:53 +0100 fs r269388 : new API tests 2009-03-12 12:29:05 +0100 fs r269386 : Map not yet committed 2009-03-12 12:28:36 +0100 fs r269385 : oops, forgot the SequenceInputStream during the previous refactoring 2009-03-12 12:12:39 +0100 oj r269384 : #i99104# set HasCategories prop 2009-03-12 12:12:08 +0100 oj r269383 : #i99104# check HasCategories even for internal dataprovider 2009-03-12 12:10:40 +0100 oj r269382 : #i99104# set HasCategories prop 2009-03-12 10:51:49 +0100 fs r269373 : #i10000# exception specifications 2009-03-12 10:49:18 +0100 fs r269372 : #i10000# exception specifications 2009-03-12 10:44:02 +0100 fs r269371 : #i10000# exception specifications 2009-03-12 10:30:55 +0100 fs r269368 : refactored the UNO service registration in this module, using the helper classes provided by comphelper itself, so you have less effort when extending the list of to-be-registered components 2009-03-12 10:30:37 +0100 fs r269367 : module-local includes 2009-03-12 07:05:54 +0100 oj r269357 : #i99104# database dataprovider doesn't need dataranges and diagramdata 2009-03-11 10:58:28 +0100 oj r269306 : #i99911# check if name of the report is a valid file name 2009-03-11 10:03:23 +0100 oj r269299 : #i99666# the report is new when the HierarchicalDocumentName is empty 2009-03-10 11:32:45 +0100 oj r269258 : #i99221# use fallback for language 2009-03-10 10:48:40 +0100 oj r269255 : #i99433# now use OStringBuffer 2009-03-10 10:36:21 +0100 fs r269252 : initial version of (X)Map 2009-03-10 09:52:23 +0100 oj r269246 : #i99433# now use OStringBuffer 2009-03-10 08:56:13 +0100 oj r269240 : #i99655# patch applied 2009-03-09 07:35:33 +0100 lla r269058 : #i10000# wrong variable assignment fixed 2009-03-06 17:20:40 +0100 fs r269030 : some explicit defaults 2009-03-06 17:20:30 +0100 fs r269029 : #i98600# 2009-03-06 14:40:34 +0100 fs r269009 : #i87692# during reload, prevent the document being modified just because of some control content changes ... 2009-03-06 12:52:20 +0100 lla r268997 : #i10000# ambigous problem with FontWeight fixed 2009-03-06 11:39:32 +0100 fs r268989 : #i10000# (approved by pl): use --without-t1-library configure option 2009-03-06 10:55:43 +0100 fs r268986 : #i99953# depends on xmlscript module now 2009-03-06 10:54:04 +0100 fs r268985 : #i99953# also adjust the event names found in dialogs embedded in the forms 2009-03-06 09:53:41 +0100 fs r268977 : #i10000# 2009-03-06 09:30:41 +0100 lla r268973 : #i10000# merge problems 2009-03-05 17:52:34 +0100 fs r268932 : #i98593# for sub components which are actually controlled by a DocumentDefinition (aka XComponentSupplier aka XCommandProcessor), close them by executing the 'close' command, not by suspending/closing the controller (which cannot be intercepted) 2009-03-05 11:41:56 +0100 fs r268889 : default the drop down line count for list/combo boxes to 20 2009-03-05 11:39:10 +0100 fs r268887 : do not display empty error messages 2009-03-02 10:13:57 +0100 lla r268639 : #i91541# CWS rebase m41 to m42 2009-03-02 09:06:27 +0100 lla r268635 : #i10000# add ';' to strings 2009-02-26 11:18:00 +0100 fs r268492 : reportdesign depends on REPORTBUILDER, not REPORTDESIGN 2009-02-26 10:11:38 +0100 lla r268489 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-26 09:04:11 +0100 lla r268488 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-24 12:09:13 +0100 lla r268392 : #i91541# #i91542# cleanups 2009-02-24 08:08:06 +0100 lla r268382 : merge all dba32a changes into dba32b 2009-02-24 07:14:55 +0100 lla r268381 : merge all dba32a changes into dba32b 2009-02-23 21:44:28 +0100 fs r268377 : oops ... don't tamper with m_aListSourceValues at the end of loadData 2009-02-23 20:57:05 +0100 fs r268376 : #i98162# don't hold the values as strings, but as ORowSetValue, this way preserving their type, and being agnostic to different result/rowset implementations doing different to-string-conversations 2009-02-23 20:55:44 +0100 fs r268375 : getObject: throwFunctionNotSupportedException, instead of silently returning NULL 2009-02-23 20:55:20 +0100 fs r268374 : #i98162# some more supported types 2009-02-23 20:54:43 +0100 fs r268373 : #i98162# +operator != 2009-02-20 09:35:39 +0100 fs r268306 : #i99422# for a font, display the font name, the style, and the size 2009-02-20 09:33:45 +0100 fs r268305 : #i99422# in the property browser, FONT supersedes CHARFONTNAME: the aggregated FormComponentHandler displays them more nicely now 2009-02-19 16:12:06 +0100 fs r268293 : #i99372# recognize DataType::FLOAT as numeric 2009-02-19 15:43:12 +0100 fs r268291 : #i99415# 2009-02-19 15:40:15 +0100 fs r268290 : #i99242# lcl_firstFocussableControl: take disabled controls into account 2009-02-19 15:34:36 +0100 fs r268289 : #i99396# properly decode the base name of the URL when using it as title 2009-02-19 15:19:05 +0100 fs r268287 : #i98247#
2009-06-05 09:47:55 +00:00
break;
}
case TypeClass_STRUCT:
{
::com::sun::star::util::Date aDate;
::com::sun::star::util::Time aTime;
::com::sun::star::util::DateTime aDateTime;
if ( _rValue >>= aDate )
{
(*this) = aDate;
}
else if ( _rValue >>= aTime )
{
(*this) = aTime;
}
else if ( _rValue >>= aDateTime )
{
(*this) = aDateTime;
}
else
SAL_WARN( "connectivity.commontools", "ORowSetValue::fill: unsupported structure!" );
CWS-TOOLING: integrate CWS dba32b 2009-06-03 14:58:08 +0200 fs r272581 : #i102439# 2009-05-29 13:56:18 +0200 fs r272456 : remove the sub form when the 'add subform' setting changes from <true/> to <false/>, not only its controls 2009-05-29 13:19:27 +0200 fs r272454 : display '(Default)' instead of an empty string when a control has the default font 2009-05-28 20:49:18 +0200 fs r272428 : #i98162# getFirstSelectedValue: do not return reference to a temporary 2009-05-27 15:30:22 +0200 msc r272353 : #102303# 2009-05-26 13:03:06 +0200 fs r272295 : spelling 2009-05-26 12:59:54 +0200 fs r272294 : merge m48 version to get latest cygwin related fixes 2009-05-25 14:02:06 +0200 fs r272239 : remove references to local files, needed for debugging sessions only 2009-05-25 14:01:16 +0200 fs r272238 : #i102021# ensure members such as bNumberFormat are initialized before actually returning them 2009-05-25 13:10:20 +0200 fs r272236 : #i10000# reset ENABLE_EVOAB2 2009-05-22 06:44:45 +0200 oj r272167 : #i99104# add import handler for calc-sett 2009-05-22 06:42:27 +0200 oj r272166 : #i99104# impl NullDate as member 2009-05-22 06:36:22 +0200 oj r272165 : #i99104# handle nulldate from parent model 2009-05-22 06:33:13 +0200 oj r272164 : #i99104# export and import calculation-settings and nulldate 2009-05-22 06:27:50 +0200 oj r272163 : #i99104# export and import calculation-settings and nulldate 2009-05-20 13:29:58 +0200 oj r272118 : #i99911# handle invalid name 2009-05-20 13:28:49 +0200 oj r272117 : #i101261# handle different rowsets 2009-05-20 11:29:55 +0200 msc r272111 : #i100000# 2009-05-20 11:28:27 +0200 msc r272110 : merge in change from dba32a 2009-05-20 11:27:38 +0200 msc r272109 : #102082# CTRL + C does not work 2009-05-20 09:43:36 +0200 oj r272106 : clean up includes 2009-05-20 09:32:15 +0200 oj r272105 : #i99060# merge error resovled now VERTICAL_ALIGN is in both stmt 2009-05-20 08:37:21 +0200 msc r272104 : add workaround for issue #102010# 2009-05-20 08:10:12 +0200 oj r272103 : #i99104# use numberformatkey 2009-05-20 08:07:02 +0200 oj r272102 : #i99104# use column info from rowset 2009-05-20 08:04:43 +0200 oj r272101 : #i102032# use a special column type where prec and scale are the values currently set at the column 2009-05-20 08:03:04 +0200 oj r272100 : #i102032# correct type info, we have to use SQL defined type names 2009-05-19 10:27:02 +0200 oj r272061 : #i99104# export null-date 2009-05-19 08:26:53 +0200 oj r272056 : #i99104# export null-date 2009-05-18 13:15:10 +0200 msc r272014 : add issue #102019# 2009-05-18 11:33:07 +0200 msc r272005 : add issue #102019# 2009-05-18 08:59:45 +0200 msc r271996 : add workaroud for issue #102010# 2009-05-15 10:21:24 +0200 msc r271929 : #101944# 2009-05-11 21:18:30 +0200 fs r271792 : #i99914# 2009-05-08 13:52:06 +0200 oj r271715 : #i96423# remember column span 2009-05-08 11:26:19 +0200 oj r271708 : #i98605# impl new scale mode 2009-05-08 10:33:35 +0200 fs r271706 : SendUserCall: only call into the shape notification routine for UserCall types where this is necessary (performance issue) 2009-05-07 20:52:44 +0200 fs r271698 : outsource ShapeProperty from shapepropertynotifier.hxx 2009-05-07 20:43:33 +0200 fs r271697 : #i99056# use notifyShapePropertyChange, instead of getShapePropertyChangeNotifier - the latter throws if no shape exists, yet 2009-05-07 20:33:58 +0200 fs r271696 : #i99056# +notifyShapePropertyChange: allow notifying chages without checking whether there actually already exists an SvxShape 2009-05-07 16:22:15 +0200 fs r271679 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:21:37 +0200 fs r271678 : #i10000# cygwin needs quotes around the classpath 2009-05-07 16:01:11 +0200 oj r271677 : #i99914# notify parent handler 2009-05-07 15:54:54 +0200 fs r271676 : #i10000# cygwin needs some quoting 2009-05-07 14:49:48 +0200 oj r271672 : #i99277# quote alias name 2009-05-07 14:48:12 +0200 oj r271671 : #i92538# add ~ in front of type 2009-05-07 14:37:13 +0200 oj r271667 : #i99118# change type from char to varchar 2009-05-07 14:36:23 +0200 oj r271666 : #i99118# clear dest columns when changing to create new table 2009-05-07 13:35:32 +0200 oj r271657 : #i94467# handle type 0 as double as well 2009-05-07 13:20:49 +0200 oj r271655 : i99743# setNull when varchar is no text 2009-05-07 12:58:06 +0200 fs r271651 : initialize the SdrObject's property change notifier after the ctor, if necessary 2009-05-07 11:47:18 +0200 fs r271647 : #i10000# 2009-05-07 10:57:16 +0200 fs r271639 : OPropertyBrowserController::propertyChange: care for the current property state, too, and properly forward it to the UI 2009-05-07 10:18:14 +0200 fs r271636 : onNewComponent: do not ask the map for grid columns, it will throw 2009-05-07 10:09:55 +0200 fs r271634 : #i101623# 2009-05-07 09:53:44 +0200 fs r271631 : #i101622# 2009-05-06 21:55:53 +0200 fs r271615 : #i10000# 2009-05-06 21:10:42 +0200 fs r271611 : #i10000# 2009-05-06 13:11:48 +0200 fs r271583 : #i10000# 2009-05-05 22:29:31 +0200 fs r271559 : proper assertion message 2009-05-05 22:29:03 +0200 fs r271558 : diagnostics 2009-05-05 22:16:16 +0200 fs r271557 : #i10000# 2009-05-05 13:50:32 +0200 fs r271513 : #i10000# 2009-05-05 10:21:50 +0200 fs r271503 : #i10000# 2009-05-05 09:30:26 +0200 fs r271501 : why did those survive the rebase -C step? 2009-05-05 09:18:12 +0200 fs r271500 : #i10000# 2009-05-04 17:08:17 +0200 fs r271475 : CWS-TOOLING: rebase CWS dba32b to trunk@271427 (milestone: DEV300:m47) 2009-05-04 14:51:26 +0200 fs r271456 : line ends 2009-04-30 15:55:27 +0200 fs r271418 : NewURL -> PublicConnectionURL 2009-04-22 21:18:34 +0200 fs r271141 : #i100944# 2009-04-22 09:12:26 +0200 oj r271071 : #i101261# little code change 2009-04-22 09:11:43 +0200 oj r271070 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:11:25 +0200 oj r271069 : #i101261# new grabage container for nodes 2009-04-22 09:11:02 +0200 oj r271068 : #i101261# new grabage container for nodes 2009-04-22 09:10:44 +0200 oj r271067 : #i101261# new grabage container for nodes 2009-04-22 09:10:21 +0200 oj r271066 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:08:24 +0200 oj r271065 : #i101261# only ask for parameters which aren't set before 2009-04-22 09:07:25 +0200 oj r271064 : #i101261# only ask for parameters which aren't set before 2009-04-22 08:49:07 +0200 oj r271062 : #i77501# preview only when needed 2009-04-22 08:45:44 +0200 oj r271061 : #i101261# new prop max rows 2009-04-22 08:44:18 +0200 oj r271060 : #i101261# create dataprovider earlier to avoid the wrong legend in chart 2009-04-22 08:42:48 +0200 oj r271059 : #i101261# handle parameter 2009-04-17 21:00:23 +0200 fs r270954 : #i98350# 2009-04-17 13:54:19 +0200 fs r270942 : #i99565# 2009-04-17 13:51:34 +0200 fs r270940 : #i101153# only localize the (potentially) localizable properties when there really is support at the control model 2009-04-17 11:43:14 +0200 fs r270932 : removed superfluous include 2009-04-17 10:10:15 +0200 fs r270926 : #i10000# 2009-04-17 10:02:36 +0200 fs r270925 : #i10000# 2009-04-17 09:15:13 +0200 fs r270918 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:14:56 +0200 fs r270917 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 09:13:25 +0200 fs r270915 : #i99056# some more refactoring of the recently introduced property change notification mechanism for UNO shapes 2009-04-17 08:30:34 +0200 fs r270914 : removed unotools/servicehelper.hxx in favour of the (duplicated) comphelper/servicehelper.hxx 2009-04-16 21:05:25 +0200 fs r270903 : #i10000# 2009-04-16 20:43:43 +0200 fs r270902 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-16 15:57:36 +0200 fs r270891 : #i99056# make SetUnoControlModel virtual 2009-04-16 15:44:02 +0200 fs r270890 : #i99056# let the ScShapeObj notify changes in its Anchor property 2009-04-16 15:36:47 +0200 fs r270889 : #i99056# enable the sheet anchor type property, too 2009-04-16 15:33:45 +0200 fs r270887 : #i99056# shape notification outsourced to the SdrObject, this is what all other shape implementations (which only aggregate an SvxShape) have access to, too 2009-04-15 14:53:13 +0200 fs r270844 : #i10000# 2009-04-15 13:08:29 +0200 fs r270836 : #i10000# 2009-04-15 12:28:14 +0200 fs r270832 : #i10000# 2009-04-15 10:59:14 +0200 fs r270827 : #i10000# 2009-04-15 09:41:08 +0200 oj r270823 : fix issues found with findbugs and pmd 2009-04-14 21:08:04 +0200 fs r270808 : #i99056# implement SheetAnchorType - now the only thing missing to enable it is the proper notification when it is modified 2009-04-14 17:09:00 +0200 fs r270799 : #i99056# implement XServiceInfo for the ScShapeObj 2009-04-14 17:07:55 +0200 fs r270798 : #i99056# implement TextAnchorType, partially implement SheetAnchorType 2009-04-14 15:54:05 +0200 fs r270786 : #i99056# SwXShape: notify changes of the AnchorType property 2009-04-14 15:47:32 +0200 fs r270785 : #i99056# deliver shapepropertynotifier.hxx 2009-04-14 15:46:54 +0200 fs r270784 : diagnostics 2009-04-14 15:08:28 +0200 fs r270781 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:41:09 +0200 fs r270773 : #i99056# outsourced the SvxShape's property change notification code into a dedicated class 2009-04-14 14:37:23 +0200 fs r270772 : in dtor, remove the properties from the temporary component context 2009-04-14 14:36:34 +0200 fs r270771 : getWeakUnoShape made const 2009-04-14 12:23:08 +0200 oj r270757 : #i101064# add missing braces 2009-04-14 12:21:25 +0200 oj r270756 : #i101065# add braces for gcc 4.3.2 2009-04-14 12:17:45 +0200 oj r270755 : #i101059# add dep for manifest 2009-04-09 12:06:58 +0200 oj r270686 : #i93100# use OptimalSize from control to get height 2009-04-08 09:56:55 +0200 oj r270619 : #i92537# handle calculations in the select columns as well 2009-04-08 09:30:39 +0200 oj r270615 : #i96657# throw error message when the key doesn't have any columns 2009-04-07 12:08:26 +0200 oj r270592 : #i77501# impl preview of the executed report 2009-04-07 12:01:56 +0200 oj r270591 : #i77501# impl preview of the executed report 2009-04-07 11:41:03 +0200 oj r270590 : #i77501# impl preview of the executed report 2009-04-07 11:39:32 +0200 oj r270589 : #i77501# impl preview of the executed report 2009-04-07 11:29:25 +0200 oj r270588 : #i77501# convert dos to unix lineends 2009-04-07 11:28:23 +0200 oj r270587 : #i77501# impl preview of the executed report 2009-04-07 11:16:50 +0200 oj r270586 : #i77501# impl preview of the executed report 2009-04-07 11:16:00 +0200 oj r270585 : #i77501# impl preview of the executed report 2009-04-07 11:15:44 +0200 oj r270584 : #i77501# impl preview of the executed report 2009-04-07 11:15:28 +0200 oj r270583 : #i77501# impl preview of the executed report 2009-04-07 11:15:04 +0200 oj r270582 : #i77501# impl preview of the executed report 2009-04-06 15:38:54 +0200 fs r270559 : merge changes from CWS dba32a herein 2009-04-03 15:56:16 +0200 fs r270494 : ImpSvMEdit::Resize: do multiple iterations, if necessary 2009-04-03 14:35:49 +0200 fs r270487 : #i10000# 2009-04-03 13:17:16 +0200 fs r270476 : #i99056# display geometry information for controls, too 2009-04-03 13:16:37 +0200 fs r270475 : #i99056# better ordering of the geometry properties 2009-04-03 13:16:07 +0200 fs r270473 : #i99056# now that SvxShape supports property change listeners, forward add/remove requests to it 2009-04-03 13:13:18 +0200 fs r270472 : #i99056# at SvxShape, allow for PropertyChangeListeners for Size/Position 2009-04-03 09:29:27 +0200 oj r270456 : #i94571# use correct prop name 2009-04-03 09:14:54 +0200 fs r270451 : merge changes from CWS dba32a herein 2009-04-02 17:00:51 +0200 fs r270424 : better diagnostics 2009-04-02 16:35:19 +0200 fs r270421 : diagnostics 2009-04-02 16:34:50 +0200 fs r270420 : #i99056# mxUnoShape not accessible anymore, use impl_setUnoShape instead 2009-04-02 16:32:48 +0200 fs r270419 : #i99056# make getUnoShape cheaper: keep the pointer to the SvxShape all the time, so there's no need to ask for it in getUnoShape. As a consequence, we will later be able to use the pointer in scenarious where performance (potentially) matters 2009-04-02 16:31:13 +0200 fs r270417 : merge changes from CWS dba32a herein 2009-04-02 16:23:16 +0200 fs r270414 : merge changes from CWS dba32a herein 2009-04-02 14:10:35 +0200 fs r270405 : #i10000# 2009-04-02 14:06:26 +0200 fs r270404 : merge changes from CWS dba32a herein 2009-04-02 14:03:03 +0200 fs r270401 : #i10000# 2009-04-02 13:58:13 +0200 fs r270400 : #i10000# 2009-04-02 12:59:44 +0200 fs r270397 : merge changes from CWS dba32a herein 2009-04-02 12:46:30 +0200 fs r270396 : #i99056# let the form page maintain a mapping between control models and control shapes 2009-04-02 12:44:07 +0200 fs r270395 : merge changes from CWS dba32a herein 2009-04-02 12:42:06 +0200 fs r270394 : merge changes from CWS dba32a herein 2009-04-02 12:35:20 +0200 fs r270393 : #i10000# precompiled header 2009-04-02 12:05:31 +0200 fs r270392 : merge changes from CWS dba32a herein 2009-04-02 12:00:42 +0200 fs r270391 : merge changes from CWS dba32a herein 2009-04-02 11:47:26 +0200 fs r270390 : merge changes from CWS dba32a herein 2009-04-02 11:39:15 +0200 oj r270389 : #i94467# foxpro impl several new types 2009-04-02 11:35:58 +0200 fs r270387 : merge changes from CWS dba32a herein 2009-04-01 14:10:51 +0200 fs r270329 : merge changes from CWS dba32a herein 2009-03-31 17:29:50 +0200 fs r270290 : merge changes from CWS dba32a herein 2009-03-30 14:53:56 +0200 fs r270233 : #i100417# don't set grid column widths to 0, but to <void/> 2009-03-30 12:31:03 +0200 oj r270213 : #i100552# wrong orb used 2009-03-30 12:19:20 +0200 oj r270212 : #i98303# convertlike corrected to sal_Unicode 2009-03-30 11:58:25 +0200 fs r270210 : merge changes from CWS dba32a herein 2009-03-30 11:38:16 +0200 oj r270205 : remove duplicate code from merge 2009-03-30 11:02:27 +0200 fs r270202 : merge changes from CWS dba32a herein 2009-03-30 11:02:19 +0200 fs r270201 : merge changes from CWS dba32a herein 2009-03-30 10:31:26 +0200 oj r270200 : #i100665# only throw exception and do not drop table 2009-03-30 09:36:24 +0200 fs r270195 : assertion text 2009-03-28 20:21:58 +0100 fs r270187 : #ii10000# 2009-03-28 20:19:54 +0100 fs r270186 : removed unused help ids 2009-03-28 20:19:40 +0100 fs r270185 : removed unused help ids 2009-03-28 20:19:10 +0100 fs r270184 : #i100237# +DefaultState/XReset 2009-03-28 00:29:29 +0100 fs r270177 : CWS-TOOLING: rebase CWS dba32b to trunk@270033 (milestone: DEV300:m45) 2009-03-27 22:56:46 +0100 fs r270173 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:55:52 +0100 fs r270172 : #i100237# DefaultState property for buttons, enabled only when Toggle=Yes 2009-03-27 22:54:15 +0100 fs r270171 : #i100237# DEFAULT_CHECKED -> DEFAULT_STATE 2009-03-27 22:53:54 +0100 fs r270170 : #i100237# introduce a DefaultState property for buttons, which implies buttongs supporting XReset, which needed some refactoring 2009-03-27 13:31:41 +0100 fs r270152 : ignore output paths 2009-03-27 11:23:44 +0100 fs r270139 : tuned behavior with respect to invalid keys/values 2009-03-27 09:57:14 +0100 fs r270136 : don't allow Double.NaN 2009-03-27 09:56:16 +0100 fs r270135 : talk about Double.NaN 2009-03-26 12:14:30 +0100 fs r270067 : removed unused parameter 2009-03-26 12:14:02 +0100 fs r270066 : removed widening conversion when checking keys 2009-03-26 09:17:34 +0100 fs r270053 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:17:11 +0100 fs r270052 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-26 09:16:49 +0100 fs r270051 : separated the enumerator functionality into a dedicated interface, this way not burdening XMap with it 2009-03-25 21:55:20 +0100 fs r270044 : #i100541# properly calculat bNeedScrollBox 2009-03-25 12:56:17 +0100 fs r270019 : renamed the previously introduced error condition 2009-03-25 12:11:48 +0100 fs r270015 : #i100095# when the error messages contain non-trivial details (SQLState/ErrorCode), then always display the 'Details' button 2009-03-25 12:10:05 +0100 fs r270012 : renamed the previously introduced error condition 2009-03-25 12:01:04 +0100 fs r270011 : #i100095# when no address book (or respective profile) can be found, then use a dedicated ErrorCondition 2009-03-25 10:29:07 +0100 fs r270003 : add missing localization 2009-03-25 10:23:12 +0100 fs r270001 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-25 10:21:55 +0100 fs r270000 : in preparation of #i1000095#: rework the error handling, allow using css.sdb.ErrorCondition values, plus allow propagating the nsresult 2009-03-20 23:05:38 +0100 fs r269829 : XMap::getSize removed 2009-03-20 23:05:19 +0100 fs r269828 : changes as suggested by sb 2009-03-20 23:04:56 +0100 fs r269827 : enhanced documentation 2009-03-20 15:31:40 +0100 fs r269815 : changes as suggested by sb 2009-03-20 13:23:52 +0100 oj r269804 : #i92538# correct the zorder that fixedtext is infront of control 2009-03-20 12:59:38 +0100 oj r269801 : #i94571# paraadjust now supports BLOCK 2009-03-20 12:58:11 +0100 oj r269800 : #i94571# export style with data style 2009-03-20 12:57:05 +0100 oj r269799 : #i94571# text align is now paraadjust 2009-03-20 12:37:02 +0100 fs r269797 : enum keys only accepted if they have the exact type, not if they have *any* enum type 2009-03-20 12:28:31 +0100 fs r269794 : some changes requested by sb 2009-03-20 08:52:47 +0100 fs r269780 : doc 2009-03-20 07:37:31 +0100 oj r269779 : #i99913# only notifiy when values are different 2009-03-20 07:36:58 +0100 oj r269778 : #i99913# add undoenv as listener at the dataprovider 2009-03-19 22:52:52 +0100 fs r269771 : added comment 2009-03-19 22:40:06 +0100 fs r269770 : +testEnumeration 2009-03-19 22:39:41 +0100 fs r269769 : implemented enumeration, getKeySet, and getValues. Should be finished now. 2009-03-19 14:01:01 +0100 oj r269743 : #i99913# reset the modified state when selecting an object 2009-03-19 12:19:54 +0100 lla r269739 : #i72390# cleanups 2009-03-19 09:25:27 +0100 fs r269727 : #i10000# 2009-03-18 23:37:02 +0100 fs r269708 : extended checks for value type acceptance 2009-03-18 23:36:41 +0100 fs r269707 : fixed value type checks 2009-03-18 14:59:56 +0100 fs r269678 : initial complex test case for the new css.container.Map implementation 2009-03-18 14:59:24 +0100 fs r269677 : verifyExpectedException moved to base class (in complexlib), and renamed to assureException for consistency 2009-03-18 14:58:35 +0100 fs r269676 : removed unused imports 2009-03-18 14:58:03 +0100 fs r269675 : first implementation of the new css.container.Map service (not completed, yet) 2009-03-18 14:57:17 +0100 fs r269674 : base class for UNO components, freeing you from some repeating work 2009-03-18 14:55:53 +0100 fs r269672 : +assureException: call a given method with given parameters on a given object, ensure that a given exception is thrown by the method implementation 2009-03-18 14:54:58 +0100 fs r269671 : +getComponentContext 2009-03-18 14:54:00 +0100 fs r269670 : isEmpty returns a boolean, not a long 2009-03-18 14:14:43 +0100 oj r269663 : #i99743# now text also supports null 2009-03-18 13:54:14 +0100 oj r269660 : #i99223# remove check for 2 params 2009-03-18 13:33:35 +0100 oj r269659 : #i99060# replace text::ParagraphVertAlign with style::VerticalAlignment 2009-03-18 13:32:18 +0100 oj r269658 : #i99060# don't set void property when void isn't allowed 2009-03-18 13:31:11 +0100 oj r269657 : #i99060# handle vertical alignment 2009-03-18 13:28:28 +0100 oj r269656 : #i99060# remove unused elements from sytle 2009-03-18 09:35:42 +0100 lla r269639 : #i72390# cleanups 2009-03-18 09:31:20 +0100 lla r269638 : #i72390# add ButtonList 2009-03-18 09:30:46 +0100 lla r269637 : #i72390# renamed interface 2009-03-18 09:30:15 +0100 lla r269636 : #i72390# use ButtonList instead of ImageList 2009-03-18 09:29:05 +0100 lla r269635 : #i72390# new ButtonList, cleanups 2009-03-18 09:26:34 +0100 lla r269634 : #i72390# cleanups 2009-03-17 12:21:20 +0100 oj r269590 : #i99222# remove assertion 2009-03-17 12:17:22 +0100 oj r269589 : #i98605# impl scale mode 2009-03-17 12:10:42 +0100 oj r269588 : #i98605# impl scale mode 2009-03-17 11:40:15 +0100 oj r269584 : #i96944# doesn't create equation for shapes 2009-03-17 11:33:16 +0100 oj r269583 : #i96423# switch calc from float to long 2009-03-16 15:19:18 +0100 fs r269550 : #i41930# enable zoom for embedded/outplace documents 2009-03-16 14:25:54 +0100 oj r269542 : #i93734# remove ContextSensitive 2009-03-16 14:21:58 +0100 oj r269541 : #i99274# page header before group header 2009-03-16 14:18:23 +0100 oj r269539 : #i99110# fix value type 2009-03-16 14:14:16 +0100 fs r269537 : line ends 2009-03-16 14:11:06 +0100 fs r269535 : line ends 2009-03-16 14:08:34 +0100 fs r269534 : #i100087# (provided my np): allow for polymorphic types with more than one parameter 2009-03-16 12:30:31 +0100 oj r269521 : compile error 2009-03-16 12:19:12 +0100 oj r269519 : compile error 2009-03-16 10:39:28 +0100 oj r269511 : compile error under linux with swap 2009-03-13 10:33:04 +0100 oj r269462 : CWS-TOOLING: rebase CWS dba32b to trunk@269297 (milestone: DEV300:m43) 2009-03-12 14:37:25 +0100 fs r269416 : interface SequenceOutputStreamTest is unneeded, and pollutes the namespace here :) 2009-03-12 14:35:07 +0100 fs r269414 : not needed 2009-03-12 14:34:15 +0100 fs r269413 : preparation for multiple tests in this module 2009-03-12 14:33:02 +0100 fs r269412 : ShowTargets was moved from module integration.forms to module complexlib 2009-03-12 14:32:48 +0100 fs r269411 : helper class for projects containing multiple complex test cases (and following a certain structure) 2009-03-12 14:00:14 +0100 fs r269407 : proper module after the move 2009-03-12 13:59:10 +0100 fs r269406 : superseded by ../makefile.mk 2009-03-12 13:47:38 +0100 fs r269403 : not needed anymore 2009-03-12 13:45:46 +0100 fs r269402 : moved, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:45:07 +0100 fs r269401 : moved from ../ 2009-03-12 13:43:59 +0100 fs r269400 : moved to ./comphelper, in preparation of adding more test cases here, with a common infrastructure 2009-03-12 13:29:47 +0100 oj r269396 : #i99914# set parent on dataprovider 2009-03-12 13:10:35 +0100 oj r269393 : #i99832# check thrown exception and show error 2009-03-12 13:08:10 +0100 fs r269392 : reorganizing tests 2009-03-12 12:52:55 +0100 oj r269390 : #i99118# convert formatkey in numberformat 2009-03-12 12:34:53 +0100 fs r269388 : new API tests 2009-03-12 12:29:05 +0100 fs r269386 : Map not yet committed 2009-03-12 12:28:36 +0100 fs r269385 : oops, forgot the SequenceInputStream during the previous refactoring 2009-03-12 12:12:39 +0100 oj r269384 : #i99104# set HasCategories prop 2009-03-12 12:12:08 +0100 oj r269383 : #i99104# check HasCategories even for internal dataprovider 2009-03-12 12:10:40 +0100 oj r269382 : #i99104# set HasCategories prop 2009-03-12 10:51:49 +0100 fs r269373 : #i10000# exception specifications 2009-03-12 10:49:18 +0100 fs r269372 : #i10000# exception specifications 2009-03-12 10:44:02 +0100 fs r269371 : #i10000# exception specifications 2009-03-12 10:30:55 +0100 fs r269368 : refactored the UNO service registration in this module, using the helper classes provided by comphelper itself, so you have less effort when extending the list of to-be-registered components 2009-03-12 10:30:37 +0100 fs r269367 : module-local includes 2009-03-12 07:05:54 +0100 oj r269357 : #i99104# database dataprovider doesn't need dataranges and diagramdata 2009-03-11 10:58:28 +0100 oj r269306 : #i99911# check if name of the report is a valid file name 2009-03-11 10:03:23 +0100 oj r269299 : #i99666# the report is new when the HierarchicalDocumentName is empty 2009-03-10 11:32:45 +0100 oj r269258 : #i99221# use fallback for language 2009-03-10 10:48:40 +0100 oj r269255 : #i99433# now use OStringBuffer 2009-03-10 10:36:21 +0100 fs r269252 : initial version of (X)Map 2009-03-10 09:52:23 +0100 oj r269246 : #i99433# now use OStringBuffer 2009-03-10 08:56:13 +0100 oj r269240 : #i99655# patch applied 2009-03-09 07:35:33 +0100 lla r269058 : #i10000# wrong variable assignment fixed 2009-03-06 17:20:40 +0100 fs r269030 : some explicit defaults 2009-03-06 17:20:30 +0100 fs r269029 : #i98600# 2009-03-06 14:40:34 +0100 fs r269009 : #i87692# during reload, prevent the document being modified just because of some control content changes ... 2009-03-06 12:52:20 +0100 lla r268997 : #i10000# ambigous problem with FontWeight fixed 2009-03-06 11:39:32 +0100 fs r268989 : #i10000# (approved by pl): use --without-t1-library configure option 2009-03-06 10:55:43 +0100 fs r268986 : #i99953# depends on xmlscript module now 2009-03-06 10:54:04 +0100 fs r268985 : #i99953# also adjust the event names found in dialogs embedded in the forms 2009-03-06 09:53:41 +0100 fs r268977 : #i10000# 2009-03-06 09:30:41 +0100 lla r268973 : #i10000# merge problems 2009-03-05 17:52:34 +0100 fs r268932 : #i98593# for sub components which are actually controlled by a DocumentDefinition (aka XComponentSupplier aka XCommandProcessor), close them by executing the 'close' command, not by suspending/closing the controller (which cannot be intercepted) 2009-03-05 11:41:56 +0100 fs r268889 : default the drop down line count for list/combo boxes to 20 2009-03-05 11:39:10 +0100 fs r268887 : do not display empty error messages 2009-03-02 10:13:57 +0100 lla r268639 : #i91541# CWS rebase m41 to m42 2009-03-02 09:06:27 +0100 lla r268635 : #i10000# add ';' to strings 2009-02-26 11:18:00 +0100 fs r268492 : reportdesign depends on REPORTBUILDER, not REPORTDESIGN 2009-02-26 10:11:38 +0100 lla r268489 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-26 09:04:11 +0100 lla r268488 : CWS-TOOLING: rebase CWS dba32b to trunk@268395 (milestone: DEV300:m42) 2009-02-24 12:09:13 +0100 lla r268392 : #i91541# #i91542# cleanups 2009-02-24 08:08:06 +0100 lla r268382 : merge all dba32a changes into dba32b 2009-02-24 07:14:55 +0100 lla r268381 : merge all dba32a changes into dba32b 2009-02-23 21:44:28 +0100 fs r268377 : oops ... don't tamper with m_aListSourceValues at the end of loadData 2009-02-23 20:57:05 +0100 fs r268376 : #i98162# don't hold the values as strings, but as ORowSetValue, this way preserving their type, and being agnostic to different result/rowset implementations doing different to-string-conversations 2009-02-23 20:55:44 +0100 fs r268375 : getObject: throwFunctionNotSupportedException, instead of silently returning NULL 2009-02-23 20:55:20 +0100 fs r268374 : #i98162# some more supported types 2009-02-23 20:54:43 +0100 fs r268373 : #i98162# +operator != 2009-02-20 09:35:39 +0100 fs r268306 : #i99422# for a font, display the font name, the style, and the size 2009-02-20 09:33:45 +0100 fs r268305 : #i99422# in the property browser, FONT supersedes CHARFONTNAME: the aggregated FormComponentHandler displays them more nicely now 2009-02-19 16:12:06 +0100 fs r268293 : #i99372# recognize DataType::FLOAT as numeric 2009-02-19 15:43:12 +0100 fs r268291 : #i99415# 2009-02-19 15:40:15 +0100 fs r268290 : #i99242# lcl_firstFocussableControl: take disabled controls into account 2009-02-19 15:34:36 +0100 fs r268289 : #i99396# properly decode the base name of the URL when using it as title 2009-02-19 15:19:05 +0100 fs r268287 : #i98247#
2009-06-05 09:47:55 +00:00
break;
}
2009-11-11 13:30:02 +01:00
case TypeClass_INTERFACE:
{
Reference< XClob > xClob;
if ( _rValue >>= xClob )
{
(*this) = _rValue;
setTypeKind(DataType::CLOB);
}
else
{
Reference< XBlob > xBlob;
if ( _rValue >>= xBlob )
{
(*this) = _rValue;
setTypeKind(DataType::BLOB);
}
else
{
(*this) = _rValue;
}
2009-11-11 13:30:02 +01:00
}
}
break;
2001-03-28 10:26:47 +00:00
default:
SAL_WARN( "connectivity.commontools","Unknown type");
break;
}
}
CWS-TOOLING: integrate CWS dba32j 2009-11-26 09:30:10 +0100 msc r277648 : #100000 fix urgent testscript error 2009-11-25 10:27:56 +0100 msc r277625 : #i100000 fix urgent testscript error 2009-11-17 22:16:46 +0100 fs r277544 : CWS-TOOLING: rebase CWS dba32j to branches/OOO320@277531 (milestone: OOO320:m5) 2009-11-17 14:26:47 +0100 fs r277535 : ooops, didn't mean to commit this uncommented line 2009-11-14 20:50:29 +0100 fs r277505 : add a --disable-pango switch to SM's configure options, and add a patch which makes certain code respect it (well, respect it better than currently, by not using some pango_x_* functions, which are not always available, even when pango itself is present) 2009-11-14 20:47:41 +0100 fs r277504 : spelling 2009-11-13 14:04:00 +0100 fs r277502 : #i100764# (commit approved by ab@openoffice.org): getModelFromBasic: do not start with the parent's parent when looking for ThisComponent, but walk up the anchestor chain, starting with the immediate parent 2009-11-13 11:04:15 +0100 fs r277496 : #i100764# better heuristics for determining whether or not to participate in the ThisComponent game 2009-11-13 11:02:30 +0100 fs r277495 : #i100764# set the WB_EXT_DOCUMENT style at the backing component's container window, when creating it without the TaskCreator (which would normally do this) 2009-11-11 13:49:11 +0100 fs r277452 : #i106816# 2009-11-11 13:48:53 +0100 fs r277451 : fix the CREATETARBAL target, which is expected to create the zips used as prebuilts. The *inc.zip missed the NSS files in case NSS was built externally 2009-11-11 12:15:34 +0100 fs r277449 : update ignore list 2009-11-11 12:09:23 +0100 fs r277448 : add a link to the Mozilla build tools download location 2009-11-11 12:07:16 +0100 fs r277447 : add a link to the Mozilla build tools download location 2009-11-06 16:23:12 +0100 fs r277393 : #i106643# 2009-11-03 23:20:29 +0100 fs r277328 : #i106574# reverted the recent fix for issue #i105235#, and implemented a better one. 2009-11-02 12:59:48 +0100 fs r277294 : #i106550# errorOccured: also display the error when we're not in a nested form action - form actions are allowed to be triggered by other instances as well
2009-12-01 09:53:35 +00:00
} // namespace connectivity
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */