which had served since d441fa1f8cc0a17a2bc166bf2f2a44f76b83a773;
we now have polymorhpic std::abs.
Change-Id: Ibe48ac51508f7d9f5a44d712550237db7f553df3
Reviewed-on: https://gerrit.libreoffice.org/32225
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Stephan Bergmann <sbergman@redhat.com>
The unit test's first blank page already started the paragraph
numbering. Setting it so that a contentless, single paragraph
section will not display numbering. In fact, no dummy paragraph
should have numbering, but that would get complicated.
This is in preparation for tdf#104710.
Change-Id: I98c9926cb3cd1e3fe3cd43bf46cab22bfbec7eba
Reviewed-on: https://gerrit.libreoffice.org/32235
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
This is a folloup for ef4b9032de, turns
out TRUTYPE_FONTTYPE does not cover CFF fonts like I thought (the
fonts that I was checking were actually TTF fonts).
Change-Id: I2d7b34b625aa03134421dc44a24e3c66be25df98
Reviewed-on: https://gerrit.libreoffice.org/32258
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Khaled Hosny <khaledhosny@eglug.org>
The unit test should have been showing a blank page before the
content page.
Non-section paragraphs should not remove the first/last status.
Frames were already excluded, now this patch also
excludes header/footer paragraphs.
This is in preparation for tdf#104710.
Change-Id: I686944843302160d6f6bbbcb73f01c836179aa9f
Reviewed-on: https://gerrit.libreoffice.org/32234
Reviewed-by: Justin Luth <justin_luth@sil.org>
Tested-by: Justin Luth <justin_luth@sil.org>
Check in SwModify::Remove() that unrelated clients aren't inserted into
an active SwClientIter.
Old value = (sw::WriterListener *) 0x0
New value = (SwDrawFrameFormat *) 0x72d83f0
SwModify::Remove (this=0x37fd850, pDepend=0x6ca85a8) at sw/source/core/attr/calbck.cxx:220
0 in SwModify::Remove(SwClient*) (this=0x37fd850, pDepend=0x6ca85a8) at sw/source/core/attr/calbck.cxx:220
1 in ClientModify(SwClient*, SfxPoolItem const*, SfxPoolItem const*) (pClient=0x6ca85a8, pOld=0x7ffe9d041bf0, pNew=0x7ffe9d041c10) at sw/source/core/unocore/unoobj2.cxx:326
2 in SwXShape::Modify(SfxPoolItem const*, SfxPoolItem const*) (this=0x6ca8540, pOld=0x7ffe9d041bf0, pNew=0x7ffe9d041c10) at sw/source/core/unocore/unodraw.cxx:2037
3 in SwClient::ModifyNotification(SfxPoolItem const*, SfxPoolItem const*) (this=0x6ca85a8, pOldValue=0x7ffe9d041bf0, pNewValue=0x7ffe9d041c10) at sw/inc/calbck.hxx:130
4 in SwFormat::~SwFormat() (this=0x39e6320, __in_chrg=<optimized out>) at sw/source/core/attr/format.cxx:239
5 in SwFrameFormat::~SwFrameFormat() (this=0x39e6320, __in_chrg=<optimized out>) at sw/source/core/layout/atrfrm.cxx:2519
So the SwXShape was a client of the SwFrameFormat that is deleted, then
is removed and added at the parent SwFrameFormat by ~SwFormat, then it
is removed *again* when its own Modify handles the SwFormatChg
notification, but this time it messes up the SwClientIter in ~SwFormat,
putting in a m_pPosition SwFrameFormat that is actually a client of the
parent SwFrameFormat, not the deleted one!
Thus the deleted SwFrameFormat is confused and sends a "sibling" of
itself the SwFormatChg notification and that then asserts with
dcontact.cxx:1468 "<SwDraw Contact::Modify(..)> - unhandled attribute?".
Change-Id: Ifba7fa8a1f41f09b1a3fcbb4cdf17f340e0fbe33
SfxTemplateManagerDlg now overrides behaviour of ModalDialog on
EventNotify by checking if the pressed key was esc and
the searchFilter is not empty and it was focused. In this case
instead of letting ModalDialog close the modal
it clears the searchFilter
Change-Id: I90d91a3b152f68b6c8780963d1b95258d9ccf23a
Reviewed-on: https://gerrit.libreoffice.org/31736
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
Tested-by: Tamás Zolnai <tamas.zolnai@collabora.com>
The calc macro test tries to select a range of cells by moving selection
right, but this fails when the UI direction is right-to-left as the
first cell will be the right most and moving right will do nothing.
Fix this by forcing en-US locale like we do with JunitTest.
Change-Id: I08d0bc50013b0936af85af75bca82a44b044d48a
Reviewed-on: https://gerrit.libreoffice.org/32243
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
...at least macOS doesn't like that; change from OSL_TRACE (effectively doing
nothing unless running a debug build with SAL_LOG=+INFO) to SAL_INFO (calling
std::stringstream -> malloc in debug builds, regardless of SAL_LOG env var) in
a005fbeddc "OSL_TRACE->SAL in framework..sal" made
the mistake obvious.
Change-Id: I43d35524a0861998a1ef58c6f448085a2fb01cd5
