Now the cache should be capable of detecting what needs
to stay the same for having the same result, so it should
be enough to have just one cache that can reuse results
even between callers from different places.
Change-Id: Ibdc0303f5b727d1a1d7be91d61db9465ed95e1c5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132673
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
It's only used for the DXArray argument in ImplLayout(), so if we
ignore that one when caching, the position can be ignored too.
Change-Id: Id2a7b452e2b75139177c1e73c48eaa466742a59c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132672
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
Some things like font could change, so it needs to be saved and compared.
If would be even better to avoid comparing OutputDevice pointers
and instead compare only what actually matters, in order to reuse
the same glyphs used for different output devices, but I can't
find all the relevant data in OutputDevice.
Change-Id: Ib8634165b5b312d1db2c9fc28f8d7fdf7b552d25
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132671
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
Wrap the text portions inside the content control in a
<loext:content-control> XML element. Also map the (so far) single UNO
property of it to <loext:content-control
loext:showing-place-holder="...">.
This is just initial export for inline text content controls, more
properties are to be added in follow-up commits.
Change-Id: I5d928255b925ed7e08fb635ba39f546e9a4879de
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132717
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
In a next step to get the Diagram mechanism/ModelData independent
from oox, I moved the Text ModelData to svx, using a TextBody
struct. This is a 1st move that covers most of what the
algorithms to handle Diagram re-layout and other functionality
use. This will potentially have to be extended accordingly
when missing data is detected. It is potentially much more
simple as the oox TextBody, by purpose.
Due to functionality using that data I could now massively move
more of it to svx.
Change-Id: I6d6e6c572f119aeefa4e91eff56f58f3ceb6a31e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132523
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
SbxVariable::NameToCaseInsensitiveName might then be used wherever
such names may be needed (see e.g. tdf#148358).
Change-Id: I7749a12a05225398848cbf72700c6f0bc119bc22
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132561
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
1. Make BasicCollection::implGetIndexForName take OUString again,
after commit f7de7de118 made it take
std::u16string_view. All call sites pass OUStrings, and commit
ef32c3b4f9 made it create OUString
from the argument.
2. Have SbxVariable cache a case-insensitive variant of the name.
It is currently only used in the Collection implementation, but
may be used in other places; Tthe names are case-insensitive in
Basic, and VBA allows non-ASCII characters in names, so this
caching might be useful elsewhere.
3. Skip non-ASCII characters when calculating name hash, to allow
non-ASCII-containing strings still have some hash variance, when
at least some of the first 6 characters are ASCII.
Change-Id: If90ccea2c4b44c34967e6b764b6fab45b2976c40
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132493
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
Regression introduced into LIBO_INTERNAL_ONLY code with
1da6908173 "add string_view variants of methods to
O[U]StringBuffer".
And add some tests. Assigning an OStringChar to an OStringBuffer also uses the
std::string_view assignment operator, so also test that with testOStringChar,
even though there it is relatively likely that the OStringChar temporary is
followed by null bytes, which could make the test happen to erroneously succeed.
But at least tools like ASan or Valgrind could catch that. On the other hand,
assigning an OUStringChar to an OUStringBuffer does not use the
std::u16string_view assignment operator (and rather uses a
ConstCharArrayDetector-based one, which was similarly broken and has been fixed
in b66387574e "Fix -Werror=array-bounds"), so
there's no test for that here.
Change-Id: I7cf10ee5ce0e4280a91d116cd82d4871a0f44af6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132363
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
> In file included from svtools/source/svrtf/parrtf.cxx:28:
> In member function ‘typename rtl::libreoffice_internal::ConstCharArrayDetector<T, rtl::OUStringBuffer&>::TypeUtf16 rtl::OUStringBuffer::operator=(T&) [with T = const rtl::OUStringChar_]’,
> inlined from ‘virtual int SvRTFParser::GetNextToken_()’ at svtools/source/svrtf/parrtf.cxx:183:94:
> include/rtl/ustrbuf.hxx:352:20: error: array subscript ‘unsigned int[0]’ is partly outside array bounds of ‘rtl::OUStringChar [1]’ {aka ‘const rtl::OUStringChar_ [1]’} [-Werror=array-bounds]
> 352 | std::memcpy(
> | ~~~~~~~~~~~^
> 353 | pData->buffer,
> | ~~~~~~~~~~~~~~
> 354 | libreoffice_internal::ConstCharArrayDetector<T>::toPointer(literal),
> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 355 | (n + 1) * sizeof (sal_Unicode)); //TODO: check for overflow
> | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> svtools/source/svrtf/parrtf.cxx: In member function ‘virtual int SvRTFParser::GetNextToken_()’:
> svtools/source/svrtf/parrtf.cxx:183:94: note: object ‘<anonymous>’ of size 2
> 183 | aToken = OUStringChar( static_cast<sal_Unicode>(nTokenValue) );
> | ^
as seen with recent GCC 12 trunk in an --enable-optimized build.
And add a test, even though it is relatively likely that the OUStringChar
temporary is followed by null bytes, which would make the test happen to
erroneously succeed. But at least tools like ASan or Valgrind could catch that.
(For the corresponding OStringChar and OStringBuffer scenario, this issue does
not arise, as OStringChar is not covered by ConstCharArrayDetector, so the
correpsonding OStringBuffer assignment operator is OK memcpy'ing n+1 elements.
There /are/ similar issues with string_view assignment operators for both
O[U]StringBuffer, which will be addressed in a later commit.)
Change-Id: Ia131d763aa5f8df45b9625f296408cc935df96ac
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132354
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Splitted and moved parts of DiagramData class to svx as peparation
to access from there. Done as pure virtual class so that no
incarnations will be possible, also made the constructor protected.
The derived class in oox hosts all functionality/data which
involves usage/modification of oox::Shape class. That way we get
closer to get the Diagram DataModel isloated/seperated.
Not-yet moved is the String/Text holding data, it's still in oox.
Moving that one will be next, that will allow to migrate quite
some more functionalty to svx.
Change-Id: I389dbf3ebf6171b8175cf30be7bbc8c20d9a38e1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132303
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
Impress and Calc used to dump the same image file
as many times as it was featured in the document,
resulting redundant, sometimes huge documents.
Note: using only checksum to recognize image duplication
is a regression, because checksum collision results
image loss. This is a very unlikely event, and
the following commits have got the same problem.
The solution is comparing the images with the same
checksum byte for byte.
See also commit b484e9814c
"tdf#83227 oox: reuse RelId in DML/VML export for the same graphic"
and commit 797fef3861
"tdf#118535 DOCX export: save header image once".
Change-Id: I9f233d521941381746634cf4f9b5991da0dadda9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131928
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
already in its ctor
which changed since...
commit 9009663deb
Date: Sun Feb 26 22:48:06 2017 +0100
tdf#83257 [API-CHANGE] Pivot chart implementation
which defers it, take a probably safely backportable approach here to
fix it.
Change-Id: I1863791af7ae737f337993833a9cf1429a3a8028
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132286
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
This is a follow up to commit 2029b2f6dd.
That commit has increased the line count of WriteCustomGeometry to more
than 500. This patch splits it to a main part of about 230 lines and
a new method for the previous 'switch' block of about 300 lines. That
makes the loops in the main part better readable.
Change-Id: Ied4378f54e7c8dc7965a5b1db15baf0b35f63f59
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132274
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
...once we have a C++20 baseline, which would render uses of the consteval
O[U]StringLiteral ctors ill-formed if we accidentally failed to write to all of
buffer's elements. This had been broken (and the TODO comments had become
misleading) with 21584b304b "make
OUString(OUStringLiteral) constructor constexpr" and
bca539e889 "make OString(OStringLiteral)
constructor constexpr".
Also add corresponding test code.
Change-Id: I2bc680282c717d403a681ff4b9396580c8382de1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132275
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Line connectors were imported as plain shapes, losing
their functionality during editing, i.e. keeping
connections of boxes and other shapes.
Change-Id: I0f1562be2dbcce0e45eb209c6ca4e035731039e1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131303
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
This reverts commit 5b53dd0514.
The using directive was not pointless, it was necessary to correctly call the
std::swap customization point.
Change-Id: I642587fe2a8fc281a6079837edce3a8a6c47ea33
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132184
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
As preparation for own im/export move data classes for
which that is now possible to svx
Made TypeConstant more safe, also sorted it. Corrected an
error there. Secured TypeConstant usage by identifying
usages (mnType -> mnXMLType). Had to convert from
oox::OptValue to std::optional since 1st is not available
in svx in datamodelcontext.cxx. That makes the test work
and the test-file load corrrectly.
Change-Id: Ifee008caf486ddb6b52a347dc08e7e053c49ef49
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132195
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
...which appears to have become even less relevant with
db89f53c31 "remove OpenGL VCL backend code". And
the vcl/unx/glxtest.cxx machinery that it is based on is (a) known to cause
issues like <https://gitlab.freedesktop.org/mesa/mesa/-/issues/3957>
"LibreOffice's OpenGL version detection code hangs when running inside a flatpak
container with a different mesa version", and (b) is one of the two reasons why
an soffice that uses Wayland nevertheless also requires Xwayland during startup
(the other reason being oosplash). So getting rid of the glxtest machinery is
beneficial.
The remaining two potential uses of OpenGL on X11/Wayland are the obscure
css.rendering.SpriteCanvas.OGL service implementation (about which
db89f53c31 states that "it seems has never been
finished or enabled (or so it most probably should be dumped too)") and some
slideshow transitions. About the latter, Caolán stated on IRC: "I think we
grew this set of stuff to check for dodgy opengl primarily for the case where
vcl used opengl for ordinary UI optimizations; but I think that use is gone now
so I wonder does it make sense to just drop all of that entirely; for just slide
transitions we apparently survived fine without the denylist for ages". (And in
any case there is still the WatchdogThread support with OpenGLZone::hardDisable
in VCLExceptionSignal_impl, vcl/source/app/svmain.cxx, should any OpenGL code
run into problems.)
(The removal of gb_LinkTarget_use_glxtest from gb_LinkTarget_use_vclmain, which
indirectly brought in gb_LinkTarget_use_libraries,*,vcl, revealed that an
explicit use of vcl was missing from various Executables etc., which thus had to
be added now.)
Change-Id: Ifa5220fd09910a4459ca546d9655e479a2e38f1e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131943
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
...and implement export of all missing commands,
use existing viewBox if suitable,
use one EnhancedCustomShape2d
move WriteCustomGeometryPoint to protected,
make GetCustomGeometryPointValue local
The fix solves tdf#100390 too.
Without the fix the entire enhanced-path was exported as one element
<a:path>. The command F was applied to the whole drawing but should
affect only the subpath. The implementation is changed so that each
subpath gets its own element <a:path> and command F acts only on its
subpath.
Support for export of handles and equations is still a long way off.
Thus there is no reason to tread shapes with and without handles
different. Shapes with handles had used method WritePolyPolygon, but
that is not able to handle subpaths. So have desided to use method
WriteCustomGeometry for all cases.
To get shapes exported regardless of path commands I have added the
export for the missing commands.
I have removed the no longer used method WritePolyPolygon.
The special treatment of shapes "moon" and "mso-spt89" (right up arrow)
in export is no longer needed. Related code parts are removed. The
unit test testFlipAndRotateCustomShape is adapted.
In case the method WriteCustomGeometry fails, the enhanced-path is
invalid. In that case an minimal custGeom is written in case of docx.
Shapes whose drawing does not touch all edges of the snap rectangle
were exported with wrong position and size of the drawing inside the
snap rectangle. That is fixed by using an existing ViewBox for the
OOXML path size. The old way of creating a path size from point
coordinates is only used if the shape has no suitable ViewBox.
The point values in unit test SdOOXMLExportTest2::testTdf111798 are
adapted to path size 21600 x 21600 and traverse direction of the points
is corrected. The resulting shape outline is still the same as before.
The expected xml is updated for file tdf92001.odp in
SdImportTest::testDocumentLayout. The resulting shape outline is the
same, because the shape touches the edges of the snap rectangle.
The case, that the shape outline does not touch a edge of the snap
rectangle is tested in SdOOXMLExportTest3::testEnhancedPathViewBox.
Still missing is the case, that ViewBox has other left,top than 0,0.
In that case all coordinates would have to be shifted because the path
size in OOXML has only width and height but not left,top. That will
not be included in this patch.
Change-Id: Ib1736d6a08371f4d98411d2769275f0580cd0030
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131837
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
As preparations to use the Diagram ModelData further isolate
it from the oox-library-only drawingML Shape used for import.
It is necessary to completely isolate the Diagram ModelData
from the Diagram import mechanism as a preparation to be
able to re-create that Shapes on-demand anytime if needed
for re-layout(s).
Also removed one unused loadDiagram implementation and
streamlined the AdvancedDiagramHelper some more.
Change-Id: I7a7c55389e0d00f70c02db73ce2c3ff9ce7a5b22
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/132058
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
Do this only in case there are no effects on the color, as that is not
yet handled.
Note that the theme color was already stored in the grab-bag, so this is
primarily interesting in case the theme color was changed or the source
format was ODP.
Change-Id: Ia4995be68d5f243875632eec4aaf8afbb8f4d5cc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/131984
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Noel Grandin