regression from
commit 57207cab00
Author: Ariel Constenla-Haile <arielch@apache.org>
Date: Tue May 28 10:01:32 2013 +0000
Add missing copy&paste of SfxAppToolBoxControl_Impl::Select
(cherry picked from commit 47204c74d9e52f54f6983af19b66af2a96b42e61)
Conflicts:
framework/inc/uielement/popuptoolbarcontroller.hxx
svtools/inc/svtools/toolboxcontroller.hxx
The menu shown in the ui is a copy of a menu. The original menu has UserData
set on it which contains the target for the command, but the copy doesn't, so
looking up the target always fails and ses the "_default" frame and not the
correct "_self" required for the template menu
Change-Id: Ic1887467bf15f081d27286bd284929c17a2b19ad
...as GCC 4.6 reports "error: invalid cast from type 'pthread_t {aka long
unsigned int}' to type 'long unsigned int'" despite [expr.reinterpret.cast]/2:
"An expression of integral, enumeration, pointer, or pointer-to-member type can
be explicitly converted to its own type; such a cast yields the value of its
operand."
Change-Id: I26f1a3e6df2f0e55c665ad4f49185e0ebf432e95
Preset shape text rotation values are read in a few places and stored in
moRotation. With these changes, moRotation is always read unchanged, but
multiplied by -1 when it's used.
Change-Id: I633d665c21daa69e15fa828a43300f10d2bf2054
Reviewed-on: https://gerrit.libreoffice.org/13561
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>
...otherwise, wouldn't make sense to have non-const/const overloads anyway.
Needs to introduce one const_cast into doctxm.cxx (which already contains quite
a handful), otherwise the required changes would turn into too much of an
avalanche.
Change-Id: I694c7a1df7089d83ee95d35ed09bb864e9703f7c
so docs cannot be toggled r/o r/w from properties page or the file name
changed. Unkink the logic a bit based on that and move now static widget
settings to .ui and remove unused checkbox
Change-Id: Ic55e41be8ffd02e214a233a2eb90736c070b5418
Trying to figure out whether the other 2 asserts could trigger, the idea
to just delay everything that could potentially re-sort the array and
thereby invalidate the current iteration index seemed appealing.
Should be faster too: the main loop now looks at each hint only once,
but deleting a hint now requires a O(log(n)) GetStartOf() lookup.
Change-Id: I5c7b8493b35ee71c8ebdc284d2508c9812dc29cf
The assert for case 3 is wrong and fires when importing ooo44732-2.doc
but there is also a bug here where a hint could be skipped.
Change-Id: I028d2d5df9e80cf0001d9bc11aa7fabcd01e83bb
regression from
commit 1049511c0c
Date: Tue Jan 13 21:43:27 2015 +0900
infobar: add buttons one by one with addButton
Change-Id: Icf758a9671191836bbe5bc7ca7d978574cb74479
SwParaSelection sets a mark on the cursor, but there is no need to do that.
DelFullPara will delete the SwTxtNode anyway.
Change-Id: I99b8dd637bd4d2b49a555e147514a2def19fa0bf
Reviewed-on: https://gerrit.libreoffice.org/14054
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
4500afcc1b was incomplete in that the
result of GetVersionEx() is actually stored and other functions do
manual checks on it.
Change-Id: I1f66a13e98e6bfc40bc35fa5382c2f6257f905d6
where odt checks for a null pTxtCrsr and skips the object and fodt
just ploughed ahead. i.e. make fodt equally useless as odt (fdo#84942)
Change-Id: I915cc1f7c065ea83377e895a932636c82698cf1f
We don't want such a mode to affect other uses of randomness, though. Thus use
a separate random number generator object for these two functions, and use a
fixed seed for it if the SC_RAND_REPEATABLE environment variable is set.
As RAND() is implemented in sc, and RANDBETWEEN() is implemented in scaddins,
it was a bit hard to figure out where to add the new functions needed, without
having to over-engineer things with UNO. (This functionality is totally
Calc-specific, but neither sc nor scaddins has any public (non-UNO) API.)
Caolan suggested the formula module, which seems like a good enough place to
me.
Change-Id: I4b0cb327392e51a18bce28478af91b0174d6b726
on this and about 30 other documents where the NaN result is cast back to a
long, giving rise to an assert in vcl/source/outdev/map.cxx:391 of
long int ImplLogicToPixel(long int, long int, long int, long int, long int):
Assertion `std::abs(n) < std::numeric_limits<long>::max() / nMapNum / nDPI'
failed.
Change-Id: Ia87e4ee6bbf0f10b7eba513e9a5ba3c42440d181
in the label for date number format
(cherry picked from commit 2c11fc40e0500da7fc217e1bc1290adfdccab82f)
Conflicts:
officecfg/registry/data/org/openoffice/Office/UI/GenericCommands.xcu
Change-Id: I8e5b6a141948fbce8361b74a3f29db3cdf7d71f4
- a inline function not use
- a variable will be set but not use
Change-Id: I67674a6cd05ddb058d11d46188d5c9d039ed40fb
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>
Also used by VCL image plugins - defer the rest to an easy-hack.
Mostly impacts mobile which uses scaddins linked without a map file.
Change-Id: I49fd40146905f4de42c9b0e7355ab89689c776bf
