Age | Commit message (Collapse) | Author | Files | Lines |
|
Map between RelOrientation::PAGE_PRINT_AREA_TOP and
loext:vertical-rel="page-content-top".
Follow-up of commit 1c593e1916c9164c7db71da2017cfc26972f8e9f
(tdf#133045 sw: add shape alignment to the top page border).
See also commit 65b7873aab5deec7157328047e869a6385e0a74a
(sw from-bottom relative orientation: add ODF filter).
Change-Id: I9bff7a9507152262dda7d126fa3e7e1bee6a8276
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104554
Tested-by: Jenkins
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Change-Id: Ie099be1e8eb9a27463886ba62d14e7d0e8cf7296
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104810
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
Embedded charts of DOCX documents were lost or replaced
by images during conversion to native ODT format,
resulted by bad handling of XEmbedPersist objects in
EmbeddedObjectContainer.
Note: Add missing loext:external-data to ODF 1.3 schema
definition to fix ODF validation error in gerrit.
See commit 2054af83fefb955e20de2b40178a11726525057e
(fdo#72520 : Added property to store external data path in chart)
and commit a49a9dab3168c03a539adc131f2ade03236edb69
(fix one more ODF validation error).
Change-Id: I9edff9af3a79370ea447ffc6078da3520d0c6f63
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104104
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Change-Id: I6df242c276ceafb325d8d9fc609dae8bb449093b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102881
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
Change-Id: I0ec82de4932f4200aeb7cf778bf93dd9d1c28eda
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/102402
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
This patch depends on tdf#77794's
7.1 commit 7cc353df4f0993228984fcda3efb2c9181dddafb.
For more details about the issue in general,
see the verbose comments in this bug's previous
7.1 commit e4635544b816d1ca27bd1ebba60f51444b0a898e.
This patch is related to CompatibilityMode < 15.
Unfortunately, the previous patch didn't work
with older Word 2010 versions of the file,
which _shouldn't_ wrap non-LayoutInCell table-anchored flies.
Unfortunately, now that different behaviour is necessary
for different Word compat levels,
it no longer allows a nice way for Writer to handle
this natively. So since it would be very unlikely for a
user to create a document like this (since the necessary
"keep inside text boundaries" is off by default in Writer,
but is forced on by definition in Word 2013+),
I'm removing the compatibility flag I added in 7.1,
and its related unit test.
[To do this natively would probably require enabling
the IsFollowingTextFlow property by default in SW.
That sounds very dangerous since this property
is not restricted to IsInTable layout situations.
This property has been around since at least LO 3.5.]
Change-Id: I70da016cb68f515924ed6c17085bf73a9e1c5492
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/100684
Tested-by: Jenkins
Reviewed-by: Justin Luth <justin_luth@sil.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
This modifies the container over which iteration is performed.
Additionally, make sure that all nested table autostyles are
collected on the first phase.
Change-Id: I74c0bb1aaacad095226c21e6bf51cc8668133bb3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/101096
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Although extremely unlikely, the Left Page style can have a different
Left and FirstLeft. In the even more unlikely case that the document
starts off using a left-page-only style, then the first header
would not show on the first page, but only on all of the following
left-only pages.
So, of course, we want the very first visible page in the document
to show the defined First header/footer.
Change-Id: I7e74fdc085509fb8d6b80f36d1402309b9db9404
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/99862
Tested-by: Jenkins
Reviewed-by: Justin Luth <justin_luth@sil.org>
|
|
Change-Id: Ic309fc9b07186ce0b86ca54028d62e0fafd104fc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/99950
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Change-Id: I0dcd483aa74e3bf5a75a074f4fbd6793984815b7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/100157
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
This allows to call collectAutoStyles where required (e.g. when enumerating
used fonts), without side effect of writing table styles XML inside the call,
out of place.
Change-Id: Ida05e373eb8502590c43e2b0e85c3b0c1107c551
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/100153
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
You might have noticed that text in header/footers
will not wrap around fly frames, but just run underneath,
regardless of the wrap settings. Strange, eh?
[This is also true in footnotes.]
In an ancient effort to be compatible with MS strangeness,
OOo decided to do this as well for interoperability reasons.
http://openoffice.org/specs/writer/compatibility/adjust-text-wrapping.sxw
Apparently, flies in tables are exempt from that
rule in MSO, so this patch adds that exemption.
TABLE EXEMPTION IS AN EXPERIMENTAL ASSUMPTION
BASED ON VISUAL OBSERVATION FROM THIS BUG REPORT.
IT IS NOT BASED ON DOCUMENTATION.
I did look in DOC and DOCX manuals, and did a google
search, but found nothing.
A compat variable keeps older ODT files no-wrap,
so that we don't break layout of existing documents.
This variable is only read in the ODT import filter.
If it doesn't exist for ODT, it is set to false.
By default it is true, so it automatically is
enabled for anything that doesn't modify it in its
import filter, including all DOC/DOCX/RTF etc,
and newly created ODT documents.
In other words, allowing wrapping in the header for table-anchors
is the new default behaviour unless an import filter turns it off.
Headers/footers are the most common example. I also tested with
footnotes, and found that Word 2016 does wrap in that case as well,
even though the UI only allows AS_CHAR anchoring.
FYI: Allowing wrapping at ALL times
can be set with the Writer compatibility option
"Use OpenOffice.org 1.1 text wrapping around objects".
Change-Id: I9ad0c82df4af794079cce86fad9e401ea4575e59
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/92378
Tested-by: Jenkins
Reviewed-by: Justin Luth <justin_luth@sil.org>
|
|
Change-Id: Ia4094f4f411e3f0652fbcc019594ca99ce2428cf
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/95368
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
Associate a style of family "drawing-page" with a style:master-page.
This fixes the small part of the draw:fill attribute problem that is
covered by OFFICE-3937 in ODF 1.3.
This is the import part.
Change-Id: I4c86fa24c36407b64ce33f0890e5da8c26c5292a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/93670
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@cib.de>
|
|
This adjusts my LO 5.2 code for tdf#91083 that tried to emulate the
table's keep-with-next property which doesn't have a matching
counterpart in MS formats.
I always confused myself trying to understand what my year-long coding
attempt was trying to do. This seems much understandable, and efficient.
The big clue was that it affected non-MS formats - which was unintended.
Change-Id: I7886e52430cb34799e25f7fcf73500e28bbe2a55
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/93443
Tested-by: Jenkins
Reviewed-by: Justin Luth <justin_luth@sil.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Map between RelOrientation::PAGE_PRINT_AREA_BOTTOM and
loext:vertical-rel="page-content-bottom".
Change-Id: I1d614bf7c82a76285f4268b8008e08c25ef9b7f0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/93120
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Change-Id: I3a6f437493caf8b4edde10703b7b7bb67ec1f848
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/92684
Tested-by: Jenkins
Reviewed-by: Xisco Faulí <xiscofauli@libreoffice.org>
|
|
Change-Id: Iea9d1c5972534eb7fef17464fda88f559e9a75f6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/92365
Tested-by: Jenkins
Reviewed-by: Xisco Faulí <xiscofauli@libreoffice.org>
|
|
Change-Id: Ic276a4ad53920c7f1e8bb8f7bcefe580ef88a446
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/92346
Tested-by: Jenkins
Reviewed-by: Xisco Faulí <xiscofauli@libreoffice.org>
|
|
Change-Id: I70d0d70c62b92ff65aadebc0952a922ce21b56d2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/92315
Tested-by: Jenkins
Reviewed-by: Xisco Faulí <xiscofauli@libreoffice.org>
|
|
This is the last padded numbering type that is supported by Word but was
not supported by Writer.
Change-Id: Ica1a0843897c61a4b569105fd21e5bfe7b5012cb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90912
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
This makes the UI work as well.
Change-Id: I4e94b85097cc359b257b07ba7517edfab3011093
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90827
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
This makes the UI work as well.
Change-Id: I8d64b88e57ba3e4fd61afba892f0ee2267f1c8b2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90683
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
ODF allows any string as style:num-format="...", go with "01, 02, 03,
...", because that seems to be consistent with both
DefaultNumberingProvider::makeNumberingIdentifier()'s fallback mechanism
and with OOXML (which uses "001, 002, 003, ..." for the "pad to 3"
case).
Change-Id: I5c5c7ee5bd61175afc3e682276e69344852106d5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/89891
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
There were two problems with this attribute:
1. It was written in style:table-cell-properties instead of
in style:style.
2. It was referencing a number format id, instead of a style
name. Moreover, the data style wasn't even exported as part
of office:styles (if at all).
Both import and export were affected. For export, it was easily
possible to reuse some related stuff from Calc, so that stuff
was moved into xmloff and used from there (there are no logic
changes for Calc). For import, loading of the invalid attribute
was kept for compat reasons. Although it's only useful for
automatic number formats, as the data styles weren't exported
properly anyway (e.g. see the document attached in bugzilla).
Change-Id: I8b70ad205972fada6f3845837d6ed5928d7d6406
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/89551
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@cib.de>
|
|
As it turns out, UI names of table styles are
leaking into documents, and changing those names
actually breaks the import of documents from
previous versions. The problem is that a table
style itself is saved using its programmatic name,
but is referenced by tables using its UI name. So
after changing the UI name, these no longer map.
It's still possible to manually reapply the style,
but if not doing this and just saving, the style
and its child cell styles will be silently lost.
Moreover, if the given document is of fodt type,
it's not even possible to save it (even not as
"save as" to odt).
Obviously, the issue isn't just with renaming.
The same happens also with documents created
with a different UI language (even English).
Fortunately, up to now English UI names were
identical to the programmatic ones. So the first
thing we can do is to accept both kinds of names
for table:template-name. This way, we solved the
problem for documents created in an English UI,
and in addition made them work in non-English UI
(unlike before). As for export, we want to always
writes programmatic names, so newly edited
documents will continue to work regardless of
future UI changes or UI language switching (and
also stay compatible with older versions).
For the fodt export failure, changed the order of
things in SwXTextTableStyle::replaceByName, as
setting a new box breaks SwXTextCellStyle::getName
in this specific case. Also changed cell styles
to be named using the parent style's programmatic
name, so new documents won't have this problem
when opened in older versions. This also fixed
part of the PythonTest_sw_python failure.
The remaining PythonTest_sw_python failure was
about the "TableTemplateName" UNO API property
of a table, which didn't work with programmatic
names. That's a real bug by itself, and was
fixed. Also an explicit test was added, to make
sure the API always returns the programmatic name.
Finally, an odf export test was added. It tests
files with both old-style UI names, and new-style
programmatic names. Styles should be correctly
imported, used by the table, and survive export.
Change-Id: I45dfda193813fea184dc42e5e75544ebc05d4a92
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87826
Reviewed-by: Maxim Monastirsky <momonasmon@gmail.com>
Tested-by: Maxim Monastirsky <momonasmon@gmail.com>
|
|
* Add checkbox to pagraph dialog
* Store property in paragraph model
* Move docx import/export from grabbag to paragraph model
* Add ODF import/export
* Add ODF unit test
* Add layout test
Change-Id: Id4e7c5a0ad145c042f862995d227c31ae2aa0abd
Reviewed-on: https://gerrit.libreoffice.org/83979
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@cib.de>
|
|
During DOCX import, not highlighted hyperlinks, ie. without hyperlink
character styles, set the Visited/Unvisited character style text
attributes to "Default Style" to avoid saving them with the
default highlighted hyperlink character styles in ODT.
Regression from the commit 576611895e51186d38ddefa10ed8d66075d9de37
"tdf#127741 DOCX import: format hyperlink with Default character style".
Change-Id: I5ffbb107e6704b285bc3d1546e08a324c386a0ab
Reviewed-on: https://gerrit.libreoffice.org/82205
Reviewed-by: László Németh <nemeth@numbertext.org>
Tested-by: László Németh <nemeth@numbertext.org>
|
|
... in sections with "Collect at end of text" set when "Track
Changes->Show" is off; the pre-increment erroneously became a
post-increment.
(regression from fe1d3328997741b55202aca7b3dc566ca833a5f4)
Change-Id: Ie438418883bdf91a519d553c10e8d9952a94a52d
Reviewed-on: https://gerrit.libreoffice.org/81234
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@cib.de>
|
|
Expose the AllowOverlap shape property as <style:graphic-properties
loext:allow-overlap="..."/>.
Change-Id: I6b6e08c67224ac7d4fb87046ea8accf94cdb583f
Reviewed-on: https://gerrit.libreoffice.org/79462
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
Change-Id: I6c1ae63a89d5ed34d2fa245279d4552949bb64a7
Reviewed-on: https://gerrit.libreoffice.org/74853
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
Change-Id: I069d84b1e6b80731c5d13a1f8b06f4ed9df0844a
Reviewed-on: https://gerrit.libreoffice.org/77153
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
This is similar to commit 6bb241ccc61c6904efec95978fa17e33c0eb1df3 (ODT
export: fix lost <text:user-field-decl> for fields in tables in headers,
2019-05-29), but here the container we want to ignore (between the
header and the field) is a text frame, not a table cell.
Change-Id: I6e8006fbd666802070cfeb88ca4528c66cc6d559
Reviewed-on: https://gerrit.libreoffice.org/73205
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
The problem was that XMLTextFieldExport::ExportFieldAutoStyle() assumed
that the text of a field anchor is always the toplevel XText, which is
true in case of body vs header text, but false in case header text vs
text-in-table-in-header.
So add an UNO property which exposes the parent of a table cell, this
way text in header (regardless of it's in a table or not) will have the
same XText, leading to writing the necessary <text:user-field-decl>
element for the matching <text:user-field-get> definition.
Change-Id: I077b8d7e9dfae4062539894318637e266b925382
Reviewed-on: https://gerrit.libreoffice.org/73176
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Layout is still not yet correct right after the import, though.
Change-Id: Icdba2e8d608f35b6b5b43b88ffb223f779af1b89
Reviewed-on: https://gerrit.libreoffice.org/71552
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
An easy way would be to just extend aXML_WritingDirection_Enum, but then
we would write the new attribute value to a non-extension namespace.
So special case the new attribute value during both import and export
(and only for table cells as a start).
Change-Id: I431bf99693c4a3452e91f241bd2f0fcfc72c68fd
Reviewed-on: https://gerrit.libreoffice.org/67770
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
OLE objects by importing their VisibleArea settings
This also reverts commit 5c1a6c9adb5ccfbb869a0a7ac730d8860a1bf405
"tdf#99631 DOCX import: set 1:1 scale in embedded XLSX".
Change-Id: I73dc945c86d0200e72767810b2ff39f233729080
Reviewed-on: https://gerrit.libreoffice.org/65343
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
based on the OLE object size, instead of applying
different scales for the bad (non-imported) VisibleArea
settings.
Change-Id: I3f246b779afd145fe260af83173c1944df21fb1a
Reviewed-on: https://gerrit.libreoffice.org/65244
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
in words, for example “az Ipsum”, not “a Ipsum”.
This bug was reported by Gellért Gyuris.
Complete commit 1037e3759bf178b52d16c12a811717f94ab9950a
(tdf#115319 references with Hungarian articles)
Change-Id: If930feb11a0308246d2512f0093bcacdc8675d0b
Reviewed-on: https://gerrit.libreoffice.org/63637
Tested-by: Jenkins
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Change-Id: If08e90a11e57698f98af15a3f35648a98b03e062
Reviewed-on: https://gerrit.libreoffice.org/60771
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: Iaf9c8e2ed72115e1f82d2541ae2a1d4803795a46
Reviewed-on: https://gerrit.libreoffice.org/60752
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
Change-Id: If47a2e4953e4b98f41c9115779522a755eea8192
Reviewed-on: https://gerrit.libreoffice.org/56522
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
|
|
Change-Id: I1a92d2001e58751c5bbe41f6480f4c46dcc8c9e7
Reviewed-on: https://gerrit.libreoffice.org/59766
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
|
|
Change-Id: I4d67c8b06aeb7ef14471b1d881f985484eab9af0
Reviewed-on: https://gerrit.libreoffice.org/57079
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I094f91c2af2d171067e3c37a8d52276835d36e9c
(cherry picked from commit 61150f1c37744457e7a1a1c1e684612b6adf0298)
Reviewed-on: https://gerrit.libreoffice.org/49424
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
Change-Id: Iad92cbe0b0376725c3078d26a4f3201d072c3c16
Reviewed-on: https://gerrit.libreoffice.org/54554
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
This change has multiple parts:
- Move "BulletAsImage" test from ODT only to globalfilter and
run it on ODT, DOC, DOCX, RTF formats and extend checks of
the XGraphic used for the bullets and the size.
- Check if GIF is animated as we need to know this in unloaded
graphic or bullets aren't rendered correctly if we assume
they are animated.
- Use "Graphic" property in writerfilter to get the graphic from
a XShape and not the "Bitmap" property which returns a Graphic
as a MetaFile and not the original Graphic.
- Make sure "GraphicBitmap" is filled with XBitmap and not with
XGraphic.
- Change "testFDO74215" to use the expected bullet size as it
is in the original document. Looks like the initial bug was
just asserting the bullet size is set to a value (non-zero).
Change-Id: I6b151c0bf9f426669e07522f0fc699fbb652046b
Reviewed-on: https://gerrit.libreoffice.org/54477
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
ODF TC decided to rename the element to meta:creator-initials, so adapt
the import so it can read this in addition to the 2 element names
produced by current and past LO versions.
Also add a test.
Unfortunately it turned out that the ODF validator had a bug in checking
character data in foreign elements, which is triggered by the test
document, see https://issues.apache.org/jira/browse/ODFTOOLKIT-475
... so update the validator jar as well.
Change-Id: I1de1e8772b41f8937f043d9a0d150e169f25ffd4
Reviewed-on: https://gerrit.libreoffice.org/53979
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>
|
|
chmod -x for doc, xlsx, png, hxx, cxx, xls, ppt, hpp
Change-Id: I52aed261e318cfd765e9adb3ed8edd226c8a59d8
Reviewed-on: https://gerrit.libreoffice.org/52569
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
|
|
in page number, chapter and outline numbering
in ~30 languages by integrating libnumbertext library.
- offapi: add linguistic2::NumberText
New NumberingType constants:
- ordinal indicators (1st, 2nd, 3rd...)
- cardinal number names (One, Two, Three...)
- ordinal number names (First, Second, Third...)
Note: these numberings are parts of OOXML, too.
Plain text files of Libnumbertext's language data
are installed in share/numbertext (similar to
share/fingerprint), allowing further customization.
Change-Id: I4034da0a40a8c926f14a3f591749a89a8d807d5a
Reviewed-on: https://gerrit.libreoffice.org/53313
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
|