The code passes adjusted position for pixel data in order to create
a sub-rect of the pixmap without copying, but at least on Intel Mac
the GC* functions then try to read the entire pixel row even though
it shouldn't be necessary. Since that may cause reading past the
pixmap data if the last line is involved, use entire pixmap width
to avoid that.
Change-Id: Ia08059b363d39dd465b2cadde5138ee045628f59
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/136003
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
* Update helpcontent2 from branch 'master'
to 9f035d777141d5f039a85d6f64b627eaf73b8079
- (related)tdf#138844, (related)tdf#149018 "Type" -> "Position and Size"
tdf#138844 changed the "Type" tab in the Properties dialog for Image
and Frame to be labeled "Position and Size" and tdf#149018 did the
same change for OLE Object. This patch updates relevant help pages to
reflect the change in the tab name.
Change-Id: Ieb8f1f6947b59638a713e583d2bb9a174b8815a2
Reviewed-on: https://gerrit.libreoffice.org/c/help/+/136026
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
1fb53a6375 "pyuno_loader::CreateInstance: delete
the initial PyThreadState" had added the PyThreadState_Delete for claimed
benefits but whose details appear lost to history (cf. the comment thread
starting at
<https://gerrit.libreoffice.org/c/core/+/3452/2#message-602ff52abdd1c95fd5c13cfe405b5fadd0048c5f>
"pyuno_loader::CreateInstance: delete the initial PyThreadState"). And at least
a recent master Linux --enable-python=fully-internal build with the bundled
Python 3.8.12 appears to succeed `make check` just fine with the
PyThreadState_Delete temporarily removed.
But on the other hand, building against upcoming Python 3.11 now started to make
CppunitTest_services fail with
> Fatal Python error: init_threadstate: thread state already initialized
> Python runtime state: initialized
> Thread 0x0000ffff81c8b020 (most recent call first):
> <no Python frame>
> Fatal exception: Signal 6
> Stack:
> /builddir/build/BUILD/libreoffice-7.3.4.2/instdir/program/libuno_sal.so.3(+0x37c28)[0xffff81be7c28]
> /builddir/build/BUILD/libreoffice-7.3.4.2/instdir/program/libuno_sal.so.3(+0x37e40)[0xffff81be7e40]
> linux-vdso.so.1(__kernel_rt_sigreturn+0x0)[0xffff81ccb7ec]
> /lib64/libc.so.6(+0x82878)[0xffff81742878]
> /lib64/libc.so.6(raise+0x20)[0xffff816fae00]
> /lib64/libc.so.6(abort+0xe8)[0xffff816e72b8]
> /lib64/libpython3.11.so.1.0(+0x104e28)[0xfffee4de4e28]
> /lib64/libpython3.11.so.1.0(+0x105200)[0xfffee4de5200]
> /lib64/libpython3.11.so.1.0(PyThread_get_thread_native_id+0x0)[0xfffee4ed6764]
> /lib64/libpython3.11.so.1.0(PyThreadState_New+0x14)[0xfffee4ed6628]
> /builddir/build/BUILD/libreoffice-7.3.4.2/instdir/program/libpyuno.so(_ZN5pyuno14PyThreadAttachC2EP3_is+0x78)[0xfffee4c8c52c]
> /builddir/build/BUILD/libreoffice-7.3.4.2/instdir/program/libpythonloaderlo.so(pyuno_Loader_get_implementation+0x5c)[0xfffee5243060]
> /builddir/build/BUILD/libreoffice-7.3.4.2/instdir/program/libuno_cppuhelpergcc3.so.3(+0x544b4)[0xffff815544b4]
because of the PyThreadState_Delete. (The deleted PyThreadState, while not
reused again directly, is still recorded in the state obtained from
PyInterpreterState_Head() later.)
So conservatively keep the PyThreadState_Delete of unclear benefit for older
Python versions and only drop it for 3.11 where it is known to have negative
effects now.
Change-Id: I9b99f1e947f0b165ddc95c2bfbd764858dda39db
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/136006
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
... instead of std::vector to not get invalidated iterators when
the container is resized.
Regression from
commit f8defe59ff
CommitDate: Wed Sep 6 22:45:10 2017 +0200
Replace some lists by vectors in unoobj (sc)
which was bad for this case.
Change-Id: I8d3a001242865cadc82b359a3198906d26373a41
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/136007
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
* Update helpcontent2 from branch 'master'
to 51cfc9a957eec7be595e8b667a528cbe749992a0
- (related) tdf#149414 change in Impress Format menu command name
Change-Id: I80d37dc341f35c89fe73d815df900178a3552c74
Reviewed-on: https://gerrit.libreoffice.org/c/help/+/136017
Tested-by: Jenkins
Reviewed-by: Seth Chaiklin <sdc.blanco@youmail.dk>
Setting new items but not clearing existing items smells like an
obvious omission. Why would we want a union of two nodes' items?
This improves things a bit for the bugdoc, but now it ends up with a
different wrong list on the node 254.
Change-Id: I3f5db0a4f81ac6e9af56d1faaefc2cb107dff597
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131841
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
* Update helpcontent2 from branch 'master'
to c7fba791de12f34179bb355fedbc111f4112bde8
- tdf#149572 update "Left Paragraph Border" and "Right Paragraph Border
tdf#149242 changed the Horizontal positioning labels in the
"to" control from:
"Left paragraph border" -> "Left of paragraph text area" and
"Right paragraph border" -> "Right of paragraph text area"
This patch updates the option names and makes corrections to
in the description of the referent of these options.
Change-Id: I604cb4354f239d476375791fa57865d10b1d0d24
Reviewed-on: https://gerrit.libreoffice.org/c/help/+/135881
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
maybe a problem since
commit 46419cd7a2
Date: Tue Jun 18 15:11:30 2013 -0400
ScFormulaCell is no longer a child class of ScBaseCell.
Change-Id: Id33072f193045e2eaf51373b47dac803f9a5d52c
presumably the cloned ScColorScaleEntry should end up with
a ScFormulaCell for the destination document and not the source one
Change-Id: I451d5827be183198b61116ab8c582cfda03b2031
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135998
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>
There is a general mismatch between XHTML and Writer lists: XHTML can
only contain list items (for ordered or unordered lists), while Writer
can contain list headers and list items. List headers have no bullet or
number at the start, list items are the normal text nodes.
Commit 8c2607ae3c (sw HTML export, XHTML
mode: fix lost </li> when last list item is not numbered, 2022-02-21)
fixed the list item end side of this problem: if all text nodes in a
list are headers, then don't write ul/ol at all, otherwise end list
headers with </li> as well to make sure the output XML is valid.
However, this created a mis-match, the starting <li> for list headers in
a list which have non-header text nodes at as was not adapted.
Fix the problem by extending OutHTML_SwFormat() so list headers in a
list with non-header text nodes always have a <li> and </li>, and this
condition is the same on the start/end side.
Calculating if at least one text node is non-header in a list may not be
cheap, so reuse the already calculated info from
OutHTML_NumberBulletListStart() in OutHTML_SwFormat().
Change-Id: I3817a489f16166fc5b4c33ee64e2283c41a4402c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135999
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
It's the same data for all threads, they access it as read-only,
so it doesn't make sense for each thread to build its own copy.
Change-Id: Ia1559c61d976bcce78661cae7e030bb8430aed7c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135794
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
for non-CJK scripts. Non Asian text was centered within as many cells
as required according to ODF1.2 20.301. But it also required that
text should be snapped to grid if style:layout-grid-snap-to is
true ( i.e. IsSnapToChars() is false ) per ODF1.2 20.305. That means
ODF1.2 20.301 ruled.
This patch changes the width of inserted SwKernGridPortion so that
non-CJK text aligns to the edge of text grid to make ODF1.2 20.301
rule, an provide more consistent layout result as expected.
Change-Id: Id6b4d2b965e20670eb9ddf657d36b8a49073f32a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135654
Tested-by: Jenkins
Reviewed-by: Mark Hung <marklh9@gmail.com>
* Update helpcontent2 from branch 'master'
to a17fbdee73a7f567637c56a1b6354c0fef3b3a10
- tdf#149353 positioning objects in multipage and multicolumn paragraphs
Presented as <note> in "to paragraph" (shared/01/05260200.xhp),
while only the text is embedded as a paragraph in "to character"
(shared/01/05260300.xhp) because there is already a <tip> on that
page, without much other text.
<note> added to swriter/guide/anchor_object.xhp that positioning
is only on the page where the anchor is located.
Change-Id: I8173050620fbeb8fc78d747ad05d8ab1f74f1346
Reviewed-on: https://gerrit.libreoffice.org/c/help/+/135169
Tested-by: Jenkins
Reviewed-by: Seth Chaiklin <sdc.blanco@youmail.dk>
... resulting format code of an existing format as not attempting
to add a duplicate. It makes no sense to insist on strictness if
the client can't know the rules..
Change-Id: I0c8ca215984bf84487036ccf2b570128b1694d54
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135898
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>
Once you enter a checkbox or picture content control, the selection is
read-only to avoid loosing content the next time the content control is
updated. But this means you can't launch the properties dialog to e.g.
enable placeholder mode.
Fix this by relaxing the restriction and allow the properties dialog on
protected cursor as well, matching Word behavior.
Change-Id: Ifef9af66d077682c46c852c9ecba2df5f17234fc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135965
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
...as dereferencing a past-the-end iterator would be fatal, as seen with
<https://ci.libreoffice.org/job/lo_ubsan/2432/> during UITest_sw_fieldDialog at
> warn:legacy.tools:19040:19040:vcl/source/treelist/treelist.cxx:1353: Entry not in model or wrong view
> /opt/rh/devtoolset-7/root/usr/lib/gcc/x86_64-redhat-linux/7/../../../../include/c++/7/debug/safe_iterator.h:283:
> error:
> attempt
> to
> dereference
> a
> past-the-end
>
> iterator
> .
>
> Objects involved in the operation:
> iterator "this" @ 0x0x7fc0a42e1fe8 {
> type = Hٷm�␡ (mutable iterator);
> state = past-the-end;
> references sequence with type `NSt7__debug13unordered_mapIP15SvTreeListEntrySt10unique_ptrI15SvViewDataEntrySt14default_deleteIS4_EESt4hashIS2_ESt8equal_toIS2_ESaISt4pairIKS2_S7_EEEE' @ 0x0x7fc0a42e1fe8
> }
>
>
> Fatal exception: Signal 6
Change-Id: Ibd547536849b15d56711e065262197a2e168db3d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135961
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
just do the default cursor move, toggle can be triggered with
space/return or directly with 0-9 (+ a for 10)
Change-Id: I654ca64e49b3a9d4672668553095e971a5c9d1e8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135921
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
The legacy iterative process doesn't need this, but the newer
straightforward one does.
Change-Id: I33932f37d5ad7bae0ffe5f23abdf2c4d82b84598
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/135920
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Luboš Luňák