This may happen when saving a currency with a changed symbol in the
current release, load it in an older release, save it there and load it
again in the newer release.
For example DKK
[$kr.-406] #.##0,00;[RED][$kr.-406] -#.##0,00
new in master/5.3 (changed symbol from 'kr' to 'kr.')
loaded and saved in earlier releases writes currency 'kr.' instead of 'DKK'
reloaded in master since there was no match as currency lead to number format
#.##0,00 [$kr.]
Change-Id: I5c8f165a2f420d77f5a2c0921c33a57797f92526
Undo/redo is limited to undo actions created by the same view in the LOK
case, this argument removes this limit. This can be used by a client for
"document repair" purposes, where undo/redo of others' changes is
intentional.
The sfx command dispatch has support for FASTCALL slots (a state
function is not called, the command is always enabled) and also has
support for state functions, but those functions only get the ID of the
slots, not its parameters. What is needed here is a command that's
disabled by default, but in case a Repair argument is used, then it's
unconditionally enabled. So handle that case in the sfx dispatcher
directly for now.
Change-Id: I96c1130bf51abcdd722684b1fa4a8277f92fd555
Which with
[$kr.-406] #.##0,00;[RED][$kr.-406] -#.##0,00
saved and reloaded gave
[$kr.-406] #.##0,00;[RED][$kr.-406]" -"#.##0,00
and thus an almost identical duplicated format with the built-in format.
Change-Id: Ie39b97576842bc29ac301d2fe64f7331f6c57fc4
There was not really a good enough reason for the change, and it
causes unfortunate regressions in 3rd-party code that previously
worked fine. (It also caused two regressions in LO's own code, but
those were already fixed.)
This reverts commit 5334ff287c65f028753171c7b38da7ffbcebc1cc.
Without this, comphelper::dispatchCommand() doesn't work on the current
frame. This causes e.g. undo not working when only a given view is
allowed to do undo.
In the desktop case vcl::Window::ImplGrabFocus() calls both
SfxViewFrame::MakeActive_Impl() and
framework::Desktop::setActiveFrame(), but in the LOK case the first was
called directly, that's how the active frame was outdated.
Change-Id: If97a9de316b4c8dd49e55cd273bdb2dfe9866f38
Reviewed-on: https://gerrit.libreoffice.org/28032
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>
Just to get this one file to compile. More errors come later. I just
spent a short time on this while waiting for something else.
Actually I have no idea what we should do on iOS nowadays. Do we want
to use cairo? Do we want to use OpenGL? Would it make sense to mimic
what we do on Android as much as possible? (But what do we do on
Android, and is that by choice or accident?) Even if that might mean
not using APIs native to iOS, but slower (not HW accelerated) FLOSS
alternatives that perform the same functionality, broadly speaking?
Change-Id: Id88a895b90f753417eced744141376656bcf72c3
This tokenizes some elements, de-tokenize while consuming
and emits elements through legacy interface. DummyTokenHandler
is just to test the correctness.
Change-Id: I1ea1e4d806ed4d426215f93b3f6b66a9776f6479
Reviewed-on: https://gerrit.libreoffice.org/27849
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
- Create style by example, tracked by undo;
- Fixed delete style, now is tracked by undo
Change-Id: Ic39b549b0b970b1b15001d527a82fb26e4a630aa
Reviewed-on: https://gerrit.libreoffice.org/27990
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
for a given site.
This patch uses the fact that getFolderByServerRelativeUrl API returns
site's root folder if used with empty path ('').
Hopefully this isn't otherwise context-dependent, so that it doesn't
return other folders in some unknown circumstances.
More correct would be instead to use the site's root path directly,
like '/sites/mysite'. But I don't know a reliable way to filter the
API part from the base URI.
To make use of the corrected SharePoint repository root id, the patch
also redirects all cases where "/" object is queried, to session's
getRootFolder().
libcmis part of the patch will hopefully soon be merged upstream.
Change-Id: I0f3b9930748de3fe4763eb27721739f097b9892d
Reviewed-on: https://gerrit.libreoffice.org/27987
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: David Tardon <dtardon@redhat.com>
Currently, entering a search term or clicking on
the category buttons on the Firefox Theme selection
dialog of Personalization option doesn't work at all.
With this url, it returns some results of firefox personas
which are applicable to LibreOffice.
Change-Id: Ie4da58cd987a1834dee969c09367229d3b69a47a
Reviewed-on: https://gerrit.libreoffice.org/27986
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>
Tested-by: jan iversen <jani@documentfoundation.org>
Project: translations 0f9380734b89eb362e97e4e23bbf32d924b39132
update translations for 5.2.1 rc1
and force-fix errors using pocheck
Change-Id: I49f23c37a0f656727ba869e616e9d0b5679b9b15
when we're not recording undo, then SdrUndoDelObj just deletes
the argument its passed. And this code assumes that it is transferred
to the SdrUndoDelObj and it still exists.
Use the same pattern as ScPostIt::RemoveCaption for this situation
Invalid read of size 8
at 0xB1713B6: SdrObject::GetOrdNum() const (svdobj.cxx:777)
by 0x3C9E029C: ScDetectiveFunc::DeleteArrowsAt(short, int, bool) (detfunc.cxx:695)
by 0x3C9E1D4C: ScDetectiveFunc::FindSuccLevel(short, int, short, int, unsigned short, unsigned short) (detfunc.cxx:1118)
by 0x3C9E2405: ScDetectiveFunc::DeleteSucc(short, int) (detfunc.cxx:1207)
by 0x3CF812BA: ScDocFunc::DetectiveRefresh(bool) (docfunc.cxx:480)
by 0x3CFC9FAF: ScDocShell::DoHardRecalc(bool) (docsh4.cxx:1250)
Address 0x50992d50 is 112 bytes inside a block of size 464 free'd
at 0x4C2D22A: operator delete(void*) (vg_replace_malloc.c:576)
by 0xB1D04BC: SdrPathObj::~SdrPathObj() (svdopath.cxx:1681)
by 0xB1701F5: SdrObject::Free(SdrObject*&) (svdobj.cxx:394)
by 0xB258666: SdrUndoObjList::~SdrUndoObjList() (svdundo.cxx:720)
by 0xB258A39: SdrUndoRemoveObj::~SdrUndoRemoveObj() (svdundo.cxx:774)
by 0xB25E29B: SdrUndoDelObj::~SdrUndoDelObj() (svdundo.hxx:298)
by 0xB25E2B7: SdrUndoDelObj::~SdrUndoDelObj() (svdundo.hxx:298)
by 0x3C7E5D0E: ScDrawLayer::AddCalcUndo(SdrUndoAction*) (drwlayer.cxx:1120)
by 0x3C9E0238: ScDetectiveFunc::DeleteArrowsAt(short, int, bool) (detfunc.cxx:692)
Change-Id: Idc6d1f8e9ad8b203dac55630f8c100e74d3e017b
Project: help 4c58b1b195d7df86ea6ccee322478064f7f8a10a
clarify on CEILING,CEILING.MATH,CEILING.XCL and FLOOR,FLOOR.MATH,FLOOR.XCL
Change-Id: I557b2be54fad33e540017e093d3261e430e9a152
So one view can't undo the changes of an other view by accident. If
this is found to be useful in the desktop case, perhaps a dedicated
config option can be added for it; for now the behavior is LOK-only.
Change-Id: I7ff505d021bd6f6be36953ecc8f8bb971ce8927e
Reviewed-on: https://gerrit.libreoffice.org/28007
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>
Test number formats of bug reports
Test odd date formats which do not follow Excel detection
Change-Id: Iaa6d25a0103cae5e3ddd76075a324240aa255c35
Reviewed-on: https://gerrit.libreoffice.org/27494
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
so if the SwView it nominally belongs to is deleted while a
mail merge wizard is open it still exists while the wizard
has a ref to it
Change-Id: I4f5acbcc1ea6404851cf6e0179fd3c795342f726
officeotron is a OOXML documents validator which can be used for
verification of export files in tests. Previously it was possible
to use it only from external repository with extra build steps.
Change-Id: Iae6ec3695fdfaf738d180405047ff04b2024c906
Reviewed-on: https://gerrit.libreoffice.org/27388
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
Edit state = false disabled undo, then edit state = true enabled the
undo button, even the last LOK callback was '.uno:Undo=disabled'. Fix
this by storing the LOK state in a map, and using it when edit is
enabled.
With this, clicking on the Edit button right after loading a document
results disabled undo/redo buttons as expected.
Change-Id: Id6023f976f135555a43486f71603c823e59d8d60
Reviewed-on: https://gerrit.libreoffice.org/28003
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>
The general cleanup of wiki development, removed a page, that configure.ac uses.
Page restored, and script updated to point at correct page.
Change-Id: Ie9a7fb6d9079506171b6ca318736626bfc742b98
They both end up as the same locale, but with LANGUAGE_SYSTEM the LanguageTag
still has the mbSystemLocale=true "this is the system locale" context.
Change-Id: I7ff8a368e4dc9c6c471dd78be744c842b188478b
When loading older documents that calculated a number format on the fly an
inherited but default format of a type could had been applied using the fixed
resolved locale instead of the default system locale, which then was stored
upon save and remained sticky. This because a format the formula depends on
already was applied using the resolved system locale.
http://bugs.documentfoundation.org/attachment.cgi?id=78559 of tdf#63267
exhibits the behavior on the hidden sheet 'Festwerte' when unprotected and
inspecting number formats in column A under Datum.
Change-Id: If23908f259458e988c5164cc5e268bfc9a6a6bcd
'strcpy' is insecure as it does not provide bounding of the memory
buffer in general, so let's avoid it even here.
Change-Id: If39319a2df7ddd9297938bc0be67fe5f8a2af962
Reviewed-on: https://gerrit.libreoffice.org/27999
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>
