See tdf#74608 for motivation.
And move the code out of a header file into a .cxx file.
Change-Id: I37b88e1d5173c9064f1b733b5d38fc40a463e2a8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98677
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Regression from commit 682e0488df
(tdf#134099 sw: fix textbox anchors on copy-paste and undo, 2020-06-29),
the problem was that setting the anchor of a frame format triggers
"modify" notifications, but the handlers of those notifications expect a
consistent layout. This invariant is not held while we're still in
SwTextNode::SplitContentNode().
Fix the problem by updating the textbox's anchor the same way "normal"
fly frames are handled, i.e. add/remove the frame format to the new/old
text node manually and block notifications.
Change-Id: If1f07d4230540796a81d9ed46a932b67d5995462
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98690
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Someone might be tempted to just delete any
border on a merged cell, hoping/expecting that
the adjoining cell would have a corresponding
border.
Granted, this test was hand-crafted and is not
necessarily something that Word would ever
create. At least be aware that any fix
that breaks this example is not a complete
solution.
Change-Id: I438c835b54ec8b33eb48de55f09a7cf2965f6440
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/95533
Tested-by: Jenkins
Reviewed-by: Justin Luth <justin_luth@sil.org>
The purpose of DesktopLOKTest::testGetSignatureState_Signed() is to make
sure that in case we don't trust a cert and we add the necessary CAs,
then we trust the cert.
So just return early when we trust the cert already, it's not an
interesting failure.
Change-Id: I32f24b5464a0e14156292de2acda562025ded73f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98673
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
According to [1]:
> Changed in version 3.6: Unrecognized escape sequences produce a DeprecationWarning.
> In a future Python version they will be a SyntaxWarning and eventually a SyntaxError.
[1] https://docs.python.org/3/reference/lexical_analysis.html#string-and-bytes-literals
Change-Id: Ia4f79f17ccb121f423f35b1e1306d5ae285e8762
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98321
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
A natural mistake - especially when there are no existing
functions - is to assume that cell number defines column
number, and so it is valid to compare between rows.
This patch provides the functions, and switches this bug's
earlier patch to properly use columns instead of cells.
This commit depends on two prior patches in bug 129452:
Change-Id: Ie305477f0e3468a4a923095d76f520d97fe99ffe
Change-Id: Ibfdac336bbb1f7303c7e585a85c94be37ad6f916
I hope that this implementation covers all the bases.
This code is dreadful to understand,
as witnessed by comments from those much smarter than I.
P.S. I also cancelled a vertical merge if the cell
is not there (in a gridAfter situation).
If a row is shorter than the previous ones,
then a vertically merged cell should not be considered
to be able to span that gap. If the cells below are still
merged, it would be considered a new merged cell.
Change-Id: I63158ac73b1bf86d9f75dd3c1299d1b1a3f08064
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/97434
Tested-by: Justin Luth <justin_luth@sil.org>
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: Justin Luth <justin_luth@sil.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
This adds TEXTLANGUAGE to "unsupported" actions, so that it is
not logged as a warning, when we get a metafile with such action.
Change-Id: I1afb3ce48e4526d7c9e623fe4571b8953f9a8ebb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98629
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
This prepares the test for semi-transparent text, but the assert
is not yet enabled until the bug gets fixed.
Change-Id: I31a241910fd7bdf27579f291a497b76292eac775
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98628
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
drawinglayer::tools causes conflicts with ::tools namespace, so
better to move Primitive2dXmlDump out from this namespace as for
inspecting purposes it ccan be injected into existing code which
could break make it conflict and break compilation.
Change-Id: I5f9e594b2b8b1dd24067fa5f5ca3cfc4eefb85cd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98627
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
In PSWriter::ImplWriteF(), (nCount + 1) - nLen used to be unsigned long
but now it's unsigned 32-bit int so on 64-bit platform values > 2^31
no longer init nStSize to negative values.
(regression from cf82475c78)
Change-Id: Ib17537cf67c80883f10cf2a3b7e38d6a8f8dfcb7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98654
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@cib.de>
to make the ongoing convert-to-UNO-constructor work easier
Change-Id: I99596561832b10538505cf4dfdabaf11d6e4d7cb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98664
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
* Update translations from branch 'master'
to 09580786c1190b9ac378ba3077785d046afb88af
- update translations for master/7.0
and force-fix errors using pocheck
Change-Id: Ie0de173892c097e353130fb3a466849cc2fd9b15
The relative orientation of OLE objects was
not copied from the replacement object to
OLE, resulting bad position.
Co-authored-by: Attila Bánhegyi (NISZ)
Change-Id: If62124e5a40218a224e047efbe86a09606b44af5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98493
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
Recheck after 7-0 branchoff
Found with bin/find-unneeded-includes
Only removal proposals are dealt with here.
Change-Id: I7f5c35474b671facbda85a9538462dd4291add91
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/97655
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Set length of the horizontal axis title box is about
to 80% of the length of the chart area, to avoid of
collision of axis titles, like MSO does.
Change-Id: Ie01308e2ddc8c1c4d19f812231ede21ff939cb98
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98484
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
...at same position.
The problem is that in this case the as-char fly was written before the
at-char fly but the positioning of the at-char fly can be relative to
its character position, i.e. before the as-char fly.
Apparently as-char flys are written in
DocxAttributeOutput::EndRunProperties() via WritePostponedDMLDrawing(),
wheras at-char flys are written earlier, in SwWW8AttrIter::OutFlys() via
DocxAttributeOutput::OutputFlyFrame_Impl().
So this undoes the swap that these undergo via the magic of the mark
stack.
Change-Id: I83a72bb2affbf321fc4dea4e7fb37bdb43cea2e7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98543
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@cib.de>
Follow DomainMapper_Impl::CloseFieldCommand() and just don't waste
effort creating a fieldmark that doesn't provide any benefit.
This should avoid any fieldmark related problems introduced in
e511a0ca5dde6d731bb126bbfe21768867890102..d9030ad6298e2f49ee63489d6158ea6ad23c0111
Change-Id: I6688dcda1e3b41ac648f3d69740f05d34bb46191
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98542
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@cib.de>
This patch now uses AttrChangedNotify instead
of NotifyItemUpdate to check for cursor change
Change-Id: Ibdbd1565d72097d2f78ecb8194a201585951ce9d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98490
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
In situations when no command line params are available (for
example in Windows Store msix packages they are still missing)
let's try to use another executable shortcut for soffice
Change-Id: I6d083912dbed1166d2d68efa5eb0096b73cb58c0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98382
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
Regression from commit: d4190685ac
(tdf#114836, only set changed SfxItemSet properties)
Change-Id: If33fa39019bbd36632d15eb3cc0606727e58b111
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98241
Tested-by: Jenkins
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
property to true, if we do not have text property attribute.
Change-Id: I1a416f74c3f034f902fd583790f9db307f5e8881
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98252
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
This was introduced with da62b0feb6 to find
the best paper size for user defined page sizes.
However, this leads to problems with Impress: Slides have screen sizes which
have no relation to paper sizes. For the slide site "Screen (16:9)" the
page size "Executive" is the nearest matching size.
However, this paper is (though supported by printers), very uncommon.
We do not want the nearest matching paper size the printer supports, but
rather we want to fit the slide on the printer default paper.
Note that finding a matching print paper still works (and it even has some
fuzzy matching), so some sort of "nearest paper matching" still exists,
(but with sane limits).
Change-Id: Ie60023d64b251954aa50e8bbdd36f6a290b9f278
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/98396
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
