Try to set another command from the dropdown, if there is any
enabled. Otherwise give up and disable the button altogether.
Change-Id: I7fa46a946826056d04851faee9c7ad85f00c3678
If a second arg is passed in the "Value" property, it will be
treated as boolean. If true - the button will be
ToolBoxItemBits::DROPDOWN, and will keep replacing the main
command with the last selected one from the dropdown. It will
also respond to status updates of that command, currently for
enabled/disabled and boolean (true treated as pressed) states.
Change-Id: I09a5c20e6d2a010867037754f036096246749ec4
... to popup menu controllers it creates, similar to what
is does for child MenuBarManager instances. That way a
child ResourceMenuController placed inside a popup menu
won't mistakenly think it's inside the main menu.
Change-Id: I75eea28eeb38f16bfaaa5526752a957859ef0d32
We should also just use the string literals as such in the code (some
newer parts of this module already does that), and not hide them
behind preprocessor macros. An Easy Hack?
Change-Id: I3ab5793ff265a4aceff2fed1562a54bfb0ce3ccb
Because some stuff wants to multiple-inherit from VclAbstractDialog and
OutputDevice-subclasses, and we'd prefer to keep all the lifetime
management through a single smart pointer class (VclPtr)
The change in msgbox.cxx and window.cxx is to workaround a bug in
VS2013 to do with virtual inheritance and delegating constructors.
Change-Id: I178e8983b7d20a7d2790aa283be838dca5d14773
Reviewed-on: https://gerrit.libreoffice.org/29140
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
- do concat for json in C++, everything else seems fragile on Windows
- have APPEND vars separately
- check that gbuildtoide work on modules without a full build (modulo
some blacklisted "creative" ones)
Change-Id: I6fe267fee7d1b77d758072303729387dfeb8e6c8
Reviewed-on: https://gerrit.libreoffice.org/30293
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>
Bunch of these were setting C++ or Make modes and icky tabs...
Also, reportedly Emacs can figure out to enable python-mode
automatically.
Change-Id: I50072488fb92cb4d27aa3f74f717a28ae3967543
In the case that the field data is not a SvxDateTimeField or SvxDateField
then the old "aFieldInfo.pFieldItem = nullptr" call will mean that
the pFieldItem is leaked for that element.
The only thing needed later is the existance of a date field, and the field's
position, so encode those two facts as a bool and EPosition and
stop abusing aFieldInfo.pFieldItem to serve as a flag
Change-Id: I6a1c46d32406dbd6865cfcf6d23a3863bada6810
Traceback (most recent call last):
File "C:cygwinhometdflodejenkinsworkspacelo_gerritConfigwindows_msc_dbgutil_32solenvqapythongbuildtoide.py", line 32, in test_gbuildtoide
del(os.environ['LD_LIBRARY_PATH']) # built with ASAN; prevent that
File "C:cygwinhometdflodejenkinsworkspacelo_gerritConfigwindows_msc_dbgutil_32instdirprogram\python-core-3.3.0libos.py", line 692, in __delitem__
raise KeyError(key) from None
KeyError: 'LD_LIBRARY_PATH'
Change-Id: I1575caf90cafffd71dd28fedc46b5e3570848be8
Reviewed-on: https://gerrit.libreoffice.org/30304
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Michael Stahl <mstahl@redhat.com>
Previously we only managed to verify a signature in case the certificate
was already imported in the local NSS db. Don't depend on that by
(temporarily) importing certificates from the PDF signature.
Also adjust a test file that failed previously (the test DB has only an
"Alice" cert imported, intentionally sign the file as "Bob" as well).
Change-Id: Id8440acc31915f5a1718ea48129b950bb67e7486
PDF signatures are always chained, so when removing a signature not only
the item at a given position should be removed on the UI, but the whole
position-last range.
Change-Id: I76b14308885267cdac994fa957218a8b7df6b3cf
We can mandate that the byte range end is the end of the file for the
last signature only.
With this, signing a previously unsigned file multiple times works, so
add a matching testcase for that as well.
Change-Id: I8fe5482890fca4dab8da6305aa7fc7f60df612d8
This reverts commit e2fa592c770222344fccb0e39db3d500fc7690b5, sould no longer be
necessary after 3be8c4c2bb919d69a2812c76edbaeaa41d149edb "The badvectorinit
plug-in probably can be retired".
Change-Id: I26fadaf002e8e21144a7e888a80bb40eb017a64b
Reviewed-on: https://gerrit.libreoffice.org/30296
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Stephan Bergmann <sbergman@redhat.com>
In a build with -fsanitize=address, this fails with:
Change-Id: Ida0d4445d7f829545b493e9dd4c2c4ef33960850
make: symbol lookup error: instdir/program/libfreebl3.so: undefined symbol: __asan_option_detect_stack_use_after_return
... during DeleteCells() and InsertCells(), instead of one Undo per
UnmergeCells() call. And actually create Undo only if bRecord requested.
Change-Id: I4f1747c3f42f36e16be81f989f0af5d048ba9d9f
We used to just dump all the object offsets in the xref of the
incremental update, but Adobe Acrobat doesn't like that, and considers
that a second signature invalidates the first. If we properly only
mention new and changed objects in the xref, then this doesn't happen.
This requires actually parsing incremental updates, the previous code
depended on LO writing not-really-incremental xrefs at the end of
incremental updates.
Change-Id: Icdd73fe0a3eab16f8c5a62f1355edbb49f6e73de
Reviewed-on: https://gerrit.libreoffice.org/30288
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>
With this commit, NSS debug mode is enabled only if you compile
with --enable-dbgutil using ./autogen.sh command.
Only for the Linux platform builds.
Change-Id: Ia8626f36692fd38e8b1dbd774e4bc0ccedc59072
Reviewed-on: https://gerrit.libreoffice.org/29781
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Giuseppe Castagno <giuseppe.castagno@acca-esse.eu>
Reviewed-by: Michael Stahl <mstahl@redhat.com>
