Age | Commit message (Collapse) | Author |
|
... used to build, since otherwise it'll break with "malformed object"
errors. Alo replace now unused XCRUN variable.
Change-Id: I58a20c34f903b7a7acb46597d94243389e6b3842
|
|
calling back() on empty vector results in undef behaviour
Crash originally reported here:
http://lists.freedesktop.org/archives/libreoffice/2013-September/055827.html
Change-Id: Ibefdc7e2495fc31d748c16fedd3cee5eb957bfa2
|
|
Change-Id: If77e1165efec7e8ef008beaeb4e6d17997ee416c
|
|
... and also export DEVELOPER_DIR so that xcrun returns CC and CXX
values that match each other and the found SDK too, which was not
previously the case on oddly set up machines with multiple SDKs.
Change-Id: Iaed362b369b558b706926f1d15d170db580a4724
|
|
Macro Expansion function is used in 'CustomAnimationPreset.cxx'
& 'TransitionPreset.cxx' to avoid code redundancy.Also appropriate
changes are to made to 'getExpandedFilePath' function.
Change-Id: I6b6ae068db3499fe2de2e78a884f9e0737577651
Reviewed-on: https://gerrit.libreoffice.org/5910
Reviewed-by: Thorsten Behrens <thb@documentfoundation.org>
Tested-by: Thorsten Behrens <thb@documentfoundation.org>
|
|
Change-Id: Iecb8104ea2891d5229e3ecace54e12287d9b0ac6
|
|
Change-Id: If2a5ce7d9163c9ef1862850ec3b89a5e600e9636
|
|
Change-Id: Ic573e5ae63fdfb3d4ff79eb6d81fe0f4b058d582
|
|
Change-Id: Icc55a6b485341e3be0d3e96f51f41a2bba1629c1
|
|
Change-Id: I467494c1731999c50883f96301b00a3dc1282c73
|
|
Change-Id: Ifafcbe6ed88a2cc18ecb1ff123adf974e10e5187
|
|
The typo causes spurious re-builds and error messages about the
dep-target.
Change-Id: Ia8bc236759c7ef13f81fe91c88f467d6e507bf29
|
|
Change-Id: I14d6a7e4a53e65070357a2a5ceab3b25498a9fd4
Reviewed-on: https://gerrit.libreoffice.org/5933
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
This removes nearly all ToLowerAscii() calls.
Conflicts:
linguistic/source/convdic.cxx
linguistic/source/convdiclist.cxx
linguistic/source/dlistimp.cxx
sc/source/filter/html/htmlexp.cxx
Change-Id: Iddcaacfb7383e1df3d2f13751a3c788eba953fdd
Reviewed-on: https://gerrit.libreoffice.org/5895
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I3a2f1ff990495c818a07e7c5b1d1ed509273f286
|
|
Change-Id: If1b4125fa0757fdcdff5466bf9b99791c930ffa6
|
|
Project: help cff95cb4ac050718c0ac7c9e37c5946ce4467c95
|
|
this one nearly killed me
Change-Id: I51f14017940a275cca33dacf6f42438da43f46bc
|
|
Change-Id: I8c071dd76e9ea84ee9aaf9cdd06b2103df4fd789
|
|
Change-Id: Id30805f09bc359c6f66d87f050427c0e586ec93d
|
|
not only one.
Change-Id: I1f611dda6a98fb6244409c0cd1fc87fc9dfaa8c3
|
|
Change-Id: I709305d4ec3c37d3fc1c2c911551174f8cfbb883
|
|
as opposed to table columns or other expressions.
So it makes no sense to slap a table name on them.
Notwithstanding HSQLDB 1.8 (our embedded database) bugs.
Change-Id: Ib5d0b1479e29b9efeafca9ebc2eb7ed8e0f42b79
|
|
Change-Id: I8660ae764c7dd51b8d780929effe895243e4fc4c
|
|
the layout changes for the basic IDE ( for the object browser and object
catalog ) seem flacky, I have seen since those changes have been introduced
some strange ( but random ) behaviour ( like the odd unrepeatable core ( maybe
related to this ) and also sometimes Modules appearing in the tree under the
wrong nodes etc. I'm no expert in the basic IDE code but this patch seems to
fix the problem.
However there is one drawback, in the core inducing scenario the tree view
( object catalog ) dissappears, this is because the patch suppresses the
problematic layout in this case ( as the layout seem not to be currently able
to deal with 'no-existent' (recently) deleted current window )
Probably in this scenario a fallback currentwin (instead of nil) could be set
this would behave better but ideally. Ultimately the layout class should
probably be modified ( possibly redesigned )
Change-Id: I9d1e23bd6fc4aae32aa78da8278c318f7051136a
|
|
...post 68e2a4e41d6e81a6e95a296d775c9ac8f5c97e8b "Revert 'Visibility doesn't
seem to work as we want in Apple's Clang.'"
Quoting <https://developer.apple.com/library/mac/documentation/developertools/
Conceptual/CppRuntimeEnv/Articles/SymbolVisibility.html> section "Symbol
Visibility and Objective-C:" "When building for x86_64 OS X or for iOS, symbol
visibility /does/ affect objective-C classes. [...] This means that if a given
class is intended to be usable outside the library or executable it's defined
in, you need to ensure proper symbol visibility."
The chosen syntax works at least with both --en/disable-64-bit "experimental"
(Clang-based) builds on OS X 10.8. Hopefully, it also works for baseline
builds. (Also, it could be that a more fine grained use of
SAL_DLLPUBLIC_EXPORT/SAL_DLLPRIVATE would be useful, but with the current setup
at least linking of Library_vcl against Library_AppleRemote works.)
Change-Id: Iff4fe9e50d1400c83879f62fe29b35bd19d58eb8
|
|
Change-Id: Icd832483228812cf3c8ddff3c0d56090f3b17856
|
|
When exporting a Frame that contains an eDrawing, check whether it's
a diagram and export it properly if so.
Exporting the diagram means adding the relations for the four
mandatory files (data, layout, quickStyle and colors), and the
optional one (drawing), and also writing those from the imported
XDocuments to the respective files.
If the drawing file exists, the data XDocument must be updated before
writing it to disk, as it contains a reference to the drawing file
that must be updated with the new relation id.
With this change, the files saved to docx that contain SmartArt will
be properly opened by MS Word without data loss.
LibreOffice won't be able to properly open saved files though, as
it requires the theme file to be present, and it's not being
exported yet. This is planned to be fixed in an upcoming patch.
Change-Id: If69ca9815dd076597a37937fe2af8ed4d1c7acc5
Reviewed-on: https://gerrit.libreoffice.org/5896
Reviewed-by: Thorsten Behrens <thb@documentfoundation.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>
|
|
Change-Id: I8484be77353b288b6dfaa0d86fa4b325fd1187a5
|
|
Change-Id: Ie3e2f6c9d1b8ff75b8450e8c377073821d0f1b37
|
|
Change-Id: Ib9a706a7ecb20bbee283e566f94fd22d412027e7
|
|
Change-Id: I516a4e9da4e670f0fdbbfa0f793ccfb308189534
|
|
Change-Id: Ib71adb544c02b764fe85b2fd32fda2097efc41d0
|
|
Change-Id: I879a47720f337b57038ac3207cb466aa42d0beeb
|
|
Because mtvelements.hxx is very slow to parse by the compiler, and
cellvalue.hxx is included everywhere. Hopefully this will speed up the
compilation time of sc...
Change-Id: Ic9a9b8483c8325e4a91021f071f2391db8b57806
|
|
Change-Id: I6c7cfd8c1e0d88ad99f5e2ca15758eab033ec50c
|
|
Should help to avoid potential memory leaks.
Change-Id: Ic9518af17d818eb439358cb86afa03e80ab9799e
|
|
Replace an events loop at the communication service with single-shot
connection workers. This should help to avoid deadlock behaviour in situations
when user quits the connection to one server and tries to connect to another
one. The events loop required synchronization blocks which cause ANR in
the situation described above.
Change the server connection interface to contain a connecting stage
in a separate method. This is more just-in-case measure for situations
when the connection constructor shouldn't be blocking.
Change-Id: I941a4b67d965f6b1f76bc9975818e82aea12bf00
|
|
Currently, when the sidebar is showing, and we hold the handle to make
it wider/narrower, if we move the mouse much to the left (so that it
can't become any wider), and now move mouse pointer to the right (while
still holding the handle), it immediately begins to shrink.
Ideally how it should behave when we move the mouse much to the left is
that- the sidebar should begin shrinking once we have the mouse pointer
over the handle again and not immediately.
Change-Id: Id17dc19e6e1ad02fb7879ace9a2e092ac0128693
|
|
Triggered by aa9af08b389a106fcfb53842ac7669b208a27205 which explicitly sets
rSet.Put( SdrTextAutoGrowHeightItem(FALSE) ); so there is something set on the
style which is being overwritten.
The code here resets the style to the default of "true" before going on to set
it to the explicit "false" again. In that window of time the master shapes
listen to the property change, on being set to autoheight they resize and on
being unset, they remain stuck on their autoheight calculated size.
Change-Id: I567a791b2bbbcb3a1a111633fabf509142984645
Reviewed-on: https://gerrit.libreoffice.org/5887
Reviewed-by: mhofmann <borim7@web.de>
Reviewed-by: Thorsten Behrens <thb@documentfoundation.org>
Tested-by: Thorsten Behrens <thb@documentfoundation.org>
|
|
Change-Id: I7495f86cb35b6f712cfb7d603f022f6f6c407266
|
|
Change-Id: I2eaa32e0ec09b239e03d3efa776f5b47c2fb5c6d
|
|
Would have caused eternal recursion, probably was intended to use
throwGenericSQLException or similar, but throwing the correct
exception is probably cleaner anyway.
Change-Id: Ic4afa623bfcd57eb68ef6cfbf737862fd40eaaa2
|
|
Currently this causes some (all?) gcc to break.
Change-Id: If6d802f5a763904d06107fa99731dd4512f18052
|
|
Change-Id: I3fd7508a3b9c362661ad1bfa66901be9f938b8e6
|
|
... where everything is inside some artificial "LibreOffice 4"
directory that is set as WINDOWSBASISROOTNAME in openoffice.lst.in
Change-Id: Ib04f84a8064739e0ea9d11b3b79cc1fa167a06e5
|
|
Change-Id: I81909e4542bd5e8d1f8ae182c3c17f9bbea9745d
Reviewed-on: https://gerrit.libreoffice.org/5881
Reviewed-by: Fridrich Strba <fridrich@documentfoundation.org>
Tested-by: Fridrich Strba <fridrich@documentfoundation.org>
|
|
Change-Id: Ib1074e3025680306c0a8bf7dcff651cefdcb90ba
|
|
Change-Id: I02881a2a7ab3178521388b76e2413b7e1cd6c443
|
|
Change-Id: Ia4fa44c2382f6d1f5a6b3fed533f401bcb03be53
|