Age | Commit message (Collapse) | Author |
|
Change-Id: Idbb9cb80fd842bc01ee857709a825821c2bcffd2
Reviewed-on: https://gerrit.libreoffice.org/69009
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
Tested-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Also changed the dialog's code a bit to work correctly with
the UI test too.
Change-Id: Idec85746f5ada979518d1fb0e09dc1c8e22ba7c1
Reviewed-on: https://gerrit.libreoffice.org/68967
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Change-Id: I44d25156c647e28c0b82f1c6da0de90a5a76261c
Reviewed-on: https://gerrit.libreoffice.org/68980
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I4b70eb2164032623a12f9e703c660eca1d6768ec
Reviewed-on: https://gerrit.libreoffice.org/68965
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Change-Id: I526faceab8eb1ce2f16d41ab1bed8cfb1bfcca24
Reviewed-on: https://gerrit.libreoffice.org/68966
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Change-Id: Idbb80d097b21386e7d2673290a318d98b92125d5
Reviewed-on: https://gerrit.libreoffice.org/68964
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
Tested-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
In the properties dialog, code expected that the ODF_FORMDROPDOWN_RESULT
is not set in case of an empty list. This caused a crash.
The field popup window code expected that the list is not specified
when it is empty, but DOC import code sets ODF_FORMDROPDOWN_LISTENTRY
even if the list is empty.
Change-Id: If4c86fc5a08cdc578150afaa42ad7e86bdba9150
Reviewed-on: https://gerrit.libreoffice.org/68963
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
Tested-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Change-Id: I66c0a7bad63d929ae346afe9d328d87dfa2c24ae
Reviewed-on: https://gerrit.libreoffice.org/68962
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
* Introduce a editing frame with a button for drop-down form field.
** The frame is mouse transparent.
** Pushing the button opens the popup window with the items of the field.
* The button is visible when the cursor is inside the field.
Change-Id: I5c7db138d14380899fee046c95a5afe14cfea213
Reviewed-on: https://gerrit.libreoffice.org/68961
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
If the AppLanguage is RTL, then it is likely that paragraph
styles inheriting from the environment will export as RTL.
So, if the paragraph is LTR (from a page style for example),
then it needs to be specified in order to override the paragraph style.
So, this is not just for RTL, even if it is a default value.
Exactly what the paragraph style will become is complicated, and
not necessarily the same across the different formats. Thus it
is safest to just spam the bidi attributes across any questionable
paragraphs.
A followup commit will similarly spam Justify in these situations,
and again the difference between doc and docx suggests it is
better to spam more, rather than less.
Change-Id: Id184a983675b147f051821e828583a4bf98b3211
Reviewed-on: https://gerrit.libreoffice.org/66922
Tested-by: Jenkins
Reviewed-by: Justin Luth <justin_luth@sil.org>
|
|
* It was weird anyway that a drop-down form field was represented
as an CheckboxFieldmark.
* It will be useful for later commits, to have a separate field type
for drop-down field.
* Needed to fix-up the API a bit because it was designed to specify
the field type after initialization. I solved it in a way to not break
the API behavior. Hopefully it's not very slow.
Change-Id: I3103e6b1c36289b27b62ab9ca7dfeebc14901c8a
Reviewed-on: https://gerrit.libreoffice.org/68960
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Change-Id: I0e9e8a0e9212cdf539caa163f9071bc2f21e4c9f
Reviewed-on: https://gerrit.libreoffice.org/68959
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Insertion of fields inside fields is not an allowed operation.
Change-Id: Icfdbc1add9c828227201b70cd545a83b6dcbc3e6
Reviewed-on: https://gerrit.libreoffice.org/68957
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Change-Id: I2b09bf30c3769f6b987409cabf24735647d3c31d
Reviewed-on: https://gerrit.libreoffice.org/68958
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Jenkins
|
|
Need to handle undo / redo explicitely for form fields, because
there is no a general working undo / redo mechanism for fieldmarks.
During the insertion of the fieldmark, text insertion also happens which
generates an interfering undo action, so we need to disable undoing
temporary.
Also need to invalidta SID_UNDO slot to make the undo toolbar item updated
after we insert a form field.
Change-Id: I358c2704cb30212a38f8a998888a36f72fa404e5
Reviewed-on: https://gerrit.libreoffice.org/68956
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Change-Id: I28a8d96d3ff31a53512af6d61c58594da7b9f73e
Reviewed-on: https://gerrit.libreoffice.org/68008
Tested-by: Jenkins
Reviewed-by: Tamás Zolnai <tamas.zolnai@collabora.com>
|
|
Change-Id: I7f9ff0fc58adf51eae7fef5ce925b91b8d1f4922
Reviewed-on: https://gerrit.libreoffice.org/68940
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ifb2c9a2d1d7dcc0ed3e8458c1a13933ccababd4c
Reviewed-on: https://gerrit.libreoffice.org/68939
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I242f56f2bc5dc8fce6db1ba8f1f8269da4fec0fa
Reviewed-on: https://gerrit.libreoffice.org/68938
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I1fcde14b66ce80157cbebe7f3ec8c5ced1e13143
Reviewed-on: https://gerrit.libreoffice.org/68937
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I89bea33e3a07655a25f1951109679188532001dd
Reviewed-on: https://gerrit.libreoffice.org/68943
Tested-by: Jenkins
Reviewed-by: Zdenek Crhonek <zcrhonek@gmail.com>
|
|
Change-Id: Ib08cca8faf2822e5a62f745c4213cf0ca2cc09d7
Reviewed-on: https://gerrit.libreoffice.org/68923
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
This is a test on a random file to see if all compilers accept this.
Change-Id: I412aef57730d07c8e6e2710fee71d5eded326d48
Reviewed-on: https://gerrit.libreoffice.org/68896
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
|
|
This should be more maintainable.
Change-Id: I60ee68cd79f7a0627bcaf8fd34d3036124ea0f9e
Reviewed-on: https://gerrit.libreoffice.org/68914
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
... in SplitNode.
The problem is that first the new anchor position is set in the
SwFrameFormat, then SwDrawContact::DisconnectFromLayout() is called
(implicitly from its SwClientNotify()).
This then cause the a11y wrapper to be disposed and an event to be sent,
which then ATKListener::notifyEvent() immediately processes by
retrieving all of the children of the anchor SwTextFrame.
At this point, we get an assert from SwTextFrame::MapModelToView,
because the layout anchor frame is still the old one, but the model
format already has the new position, so the frame can't map the
anchor position.
Avoid this by explicitly disconnecting from the layout before setting
the anchor on the SwFrameFormat.
Change-Id: Iba8960729dd041e13de4963d1b2ab6b223c8a427
Reviewed-on: https://gerrit.libreoffice.org/68880
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
In SwXMLImport::SetConfigurationSettings() there is a list of settings
that will be ignored/not loaded form the file if IsLoadUserSettings()
is false, which is presumably not the default.
The beginning of this list is obviously settings that can be set via
Tools->Options, but the end is mostly a bunch of Word-compat layout
settings that cannot be set in the UI.
Let's assume some cargo-culting happened there (this function is tricky
and underdocumented anyway), and replace the end of the list with those
settings that have been added to Tools->Options->Writer->Compatibility
in the last years, since those can be called "user settings".
Change-Id: I6ea4a5da4033ca3512d71e4123b47d46e20bf596
Reviewed-on: https://gerrit.libreoffice.org/68803
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
SwXMLImport::SetConfigurationSettings() needs to handle only those
settings for which the default is different vs. old OOo/LO versions;
DisableOffPagePositioning however defaults to false, so we can just rely
on that default. The inverted bDisableOffPagePositioning check is
wrong.
(regression from fe3d5766fa3c42f6cf8d1ea47af820e0b1c1cf48)
Change-Id: I300fa597f58b586b49089bb555a00a6923862abe
Reviewed-on: https://gerrit.libreoffice.org/68801
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
By implementing BTLR support in
SwTextFrame::SwitchVerticalToHorizontal() (Point version).
Cursor traveling now looks good: all of up/down/left/right direction
have the correct paragraph and character position for all the lrtb, tbrl
and btlr cases.
As a side effect this also fixes mouse click, where clicking above the
paragraph positioned the cursor at the bottom of the paragraph.
Explicitly add a test for the mouse case as well, given that I initially
planned to fix the keyboard part directly in SwCursor::UpDown(), where
the keyboard test would pass, but not the mouse one.
Change-Id: Iabeded3f03a64416cfcaf58e0438c4a1a793e662
Reviewed-on: https://gerrit.libreoffice.org/68886
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Change-Id: I8791d0f2cc33813c6034b771fa1f699143c3cfac
Reviewed-on: https://gerrit.libreoffice.org/68877
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
Change-Id: Ic6d8082251246f3ca8c00b090679a0e590b0a8d0
Reviewed-on: https://gerrit.libreoffice.org/68876
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
detect static variables that can be made const.
Thanks to mike kaganski for suggesting this.
Here I introduce a new plugin feature - using markers
in nearby comments to disable the plugin for specific
vars.
Some of this stuff was old debugging code. I removed the stuff
that was older than 5 years.
Change-Id: I6ec7742a7fdadf28fd128b592fcdf6da8257585c
Reviewed-on: https://gerrit.libreoffice.org/68807
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
This causes asserts from std::upper_bound in ooo26330-3.sxw (without the
fix to the initial expansion, when the fields all expand empty)
Change-Id: I1005e68d27f53986b9216541e0b26ce52082cf55
|
|
Found with bin/find-unneeded-includes
Only removal proposals are dealt with here.
Change-Id: I9dd7984affc4343f148c66077feaac19176adf51
Reviewed-on: https://gerrit.libreoffice.org/68769
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
If nStartPos == size() and !bUseFirst.
(regression from d0e9cc832d19b622532f01580d9cf78ee0b215db)
Change-Id: I571d24ec5e9d4f2780e7c6d5c8cee09baeaffcc1
Reviewed-on: https://gerrit.libreoffice.org/68821
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
After import, SwGetExpField are empty if ShowRedlineChanges=false
because the content from the document is lost in SwGetExpField::Copy().
Example is ooo26330-3.sxw, which then asserts once fields are updated,
but that is a different problem.
(regression from 57f358f1ac4f3e05fcdf54326388ba8a19f0f2b3)
Change-Id: I1bf939d69e599919c3a5c5631091ca0a0074d139
Reviewed-on: https://gerrit.libreoffice.org/68820
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
The problem was seen here:
https://bugzilla.suse.com/show_bug.cgi?id=1127857.
One needs to call SwGlobals::ensure in order to initialize
SfxApplication via SfxApplication::SetModule(SfxToolsModule::Writer.
Change-Id: I3add24b56dc44a93ad81c1b52539da821f6244fc
Reviewed-on: https://gerrit.libreoffice.org/68839
Tested-by: Jenkins
Reviewed-by: Tomáš Chvátal <tchvatal@suse.cz>
|
|
See bt https://bugs.documentfoundation.org/attachment.cgi?id=149389
It seems the method from cui/numpages.cxx takes more things into account
At least, it doesn't crash for the attachment of this tracker
Change-Id: I05ca0b019ac483890bea4acdbb695d73b660e638
Reviewed-on: https://gerrit.libreoffice.org/68033
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
Change-Id: Idd125c18bee1a39b9ea8cc4f8c55cddfd37c33e1
Reviewed-on: https://gerrit.libreoffice.org/68579
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Instead of showing the tracked deletion as untracked
normal text, losing the original and actual text contents,
now w:del within a same size w:ins is imported as
tracked deletion correctly.
Change-Id: I7390df34f557664860a63779c5f972d5013ecc99
Reviewed-on: https://gerrit.libreoffice.org/68779
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Change-Id: I556d61256bed8f607eed5a9c6e0ed3fcfd63af2c
Reviewed-on: https://gerrit.libreoffice.org/68802
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I321233437b573d254ca63aaf7274b54334ec066a
Reviewed-on: https://gerrit.libreoffice.org/68786
Tested-by: Jenkins
Reviewed-by: andreas_kainz <kainz.a@gmail.com>
|
|
Commit 56b2cf0c10d9caa01ebae1d80465e342d046a85c introduced a "feature"
which would hide an empty line after a table and only make it visible
when the cursor is in it.
So when loading an ODF doc, only enable this feature for which have the
CollapseEmptyCellPara setting set.
Change-Id: Ib4dfbbe8d45eb57547c51c7eee2f81331fc5300e
Reviewed-on: https://gerrit.libreoffice.org/68742
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
Change-Id: I0b1c04b846ff5e6197fce59bd3c4325528169a7c
Reviewed-on: https://gerrit.libreoffice.org/68787
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I304621018cb1e2a47e478e86df4229bcf2176741
Reviewed-on: https://gerrit.libreoffice.org/68757
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Due to sw_redlinehide the SwTextNode::MakeFrame() is skipped on
SwTextNodes that are merged due to redlining. But then the
IgnoreStart/IgnoreEnd flags on the hints are not initialised and some
assert like assert(pHt->IsFormatIgnoreStart()) fails.
tdf90056-1.odt is an example document.
There doesn't appear to be a convenient place to initialise it per-node
as it is finished; the ODF import inserts APPEND_PARAGRAPH before
inserting the hints themselves.
So remove the initialisation from MakeFrame() and just do it in
SwReader::Read().
Change-Id: Ib33fe3033fc05bd2f5ef2ac8d059d587642ccf48
Reviewed-on: https://gerrit.libreoffice.org/68748
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
The problem is that the ToX must be updated with the same layout redline
setting as it was originally created, so that subsequent Redo actions
see the expected node indexes.
Unfortunately it's not enough to just pass a flag to the ToX update
functions, because they check GetTextNodeForParaProps() so we need a
layout corresponding to the layout setting; if there isn't one, the
existing one is temporarily toggled.
This could be much better if the MergedPara would be independent of the
layout and always exist, but with the various SwModify design issues
that looks tricky to do...
(regression from 80cedb5dcb6a7dd6c01349b93fab49ecee5f6594)
Change-Id: Ibdc5b4ace54ace27e5223a25ecaf39bb493fb69b
Reviewed-on: https://gerrit.libreoffice.org/68704
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
Tested-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
This reverts commit c9bb48386bad7d2a40e6958883328145ae439cad,
and adds a bunch more fixes.
Change-Id: Ib584d302a73125528eba85fa1e722cb6fc41538a
Reviewed-on: https://gerrit.libreoffice.org/68680
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ia6f79e50b93004b8bf896e24450ce8666b303037
Reviewed-on: https://gerrit.libreoffice.org/66856
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
|
|
...given that they get assigned with values of those types. This change has no
actual effect on the results, but silences Clang
-fsanitize=implicit-signed-integer-truncation warnings like "implicit conversion
from type 'long' of value -671088640 (64-bit, signed) to type 'sal_uInt32' (aka
'unsigned int') changed the value to 3623878656 (32-bit, unsigned)".
Change-Id: I763a9d85b9e40797312688551206c308fc3d36e3
Reviewed-on: https://gerrit.libreoffice.org/68705
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
This change has no actual effect on the results, but silences benign Clang
-fsanitize=implicit-signed-integer-truncation warnings.
Change-Id: I0e96d998f9030c5bb4431dc0ccd689c8f2f3aeea
Reviewed-on: https://gerrit.libreoffice.org/68711
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|