Commit Graph

520 Commits

Author SHA1 Message Date
6d78c490b0 Get rid of GRADIENT_* defines, they conflict with Windows GDI types.
Change-Id: Ia0bc37f81b7213b20fc5093beb46d7a05b1ac931
2012-06-12 16:31:43 +02:00
3858ba3585 targetted string re-work
Change-Id: I9ed4df2a25c8a2a868f83de48472715a6cabe2be
2012-05-22 18:21:18 +01:00
b1c3e8ae28 make gbuild the default assumption of build.pl
this removes dmake completely out of the build for migrated modules
build.pl now assumes modules to be gbuild, unless there is a
prj/dmake file

Change-Id: I674a036b182ee13c5ec093e83cb3d38133112d3b
2012-04-29 03:50:46 +02:00
41c4585504 tune the guessed position a bit, so that it works better with test documents 2012-04-18 19:17:11 +02:00
e14b7cdd54 implemented EmfPlusRecordTypeDrawString record 2012-04-18 19:08:43 +02:00
8e92501541 added debug message for unsupported cases 2012-04-18 15:40:05 +02:00
695d82b63d SrcRect for SrcUnit = 2 is in pixels, don't Map it 2012-04-18 15:37:50 +02:00
fc435133c3 ImplText never instantiated 2012-04-18 13:03:21 +01:00
e94e669bd3 set minimal width if pen width is set to zero 2012-04-13 16:25:31 +02:00
35fb870b5f fix crash on invalid/missing pen 2012-04-13 13:15:26 +01:00
830724367d fix busted indent 2012-04-13 13:15:26 +01:00
4c50f23f87 LinkTarget.mk: remove gb_LinkTarget_add_package_headers 2012-04-08 01:05:53 +02:00
c923f7d2c2 gbuild: "use" vs. "add":
Naming convention for gbuild methods:
- "add" is used for stuff that is logically a part of the target
  (i.e. not registered at the Module, but defined in the target's makefile)
- "use" is used for stuff that is logically a different target
  (i.e. it is registered at the Module, has it's own makefile, may be
  in a different module than the target)
2012-04-08 01:05:52 +02:00
fe630f3b09 callcatcher: update list, add README, remove some newly unused code 2012-04-05 13:59:05 +01:00
66aed6f470 New try to sort out the overloaded virtual method weirdness
Revert "Clang WaE: -Woverloaded-virtual weirdness, this seems to help"
as that broke tinderboxes. Try another way instead, renaming two
versions of the overloaded render() method, so that there is no longer
any overloading.

Compiles -Werror-clean with Clang, hopefully no problem with other
compilers either.

This reverts commit 86b99ab408c7235604a1a02d7bc531fe5d5f97f6.
2012-04-04 08:27:23 +03:00
86b99ab408 Clang WaE: -Woverloaded-virtual weirdness, this seems to help 2012-04-04 01:05:50 +03:00
6f087d103a removed duplicate includes in cppcanvas / cppuhelper / cui 2012-03-30 00:42:35 +09:00
e443216ce2 readUnicodeString can just return the string
seeing as no-one checks the return value, we don't need
the intermediate temporary string
2012-03-26 12:18:09 +01:00
80fe015a86 Add COMPHELPER_SERVICEDECL_EXPORTS* back in, after libmerge changes
Did I mention I hate boilerplate copied all over the place? ;)
2012-03-16 17:02:23 +01:00
af1514abf2 Performance: avoid scaling just-generated bitmap 2nd time
We're creating the bitmap to exactly match our output size -
so why rounding off total transformation again, incurring another
expensive bitmap scale down in the canvas?
2012-03-16 17:02:20 +01:00
84b3f7b345 gbuild: get rid of realpath in gb_Foo_set_include 2012-03-10 21:23:20 +01:00
24b449b7fc cppcanvas: processEMFPlus: fix valgrind warning:
When reading EmfPlusRecordTypeDrawImagePoints, handle the variant where
points are stored as 16-bit integers instead of floats correctly.
Found via valgrind on loading bugdoc from fdo#42252.
2012-03-09 15:40:00 +01:00
0d396171b6 workaround for buggy metafile images 2012-03-07 16:59:53 +01:00
7b348b9f67 CRLF to LF, no code changed 2012-02-27 21:38:50 +07:00
eed2695cd5 Removed unused code from cppcanvas 2012-02-17 15:48:23 +00:00
8963cc01f5 EasyHack: Removed gcc conditionals 2012-02-13 17:33:15 +00:00
8b9615be88 Added (and improved) READMEs for modules which used to be in libs-gui 2012-02-08 07:46:57 +01:00
c857dff7fc switch to include-based build rather than sourced-based build 2012-02-05 19:34:05 -06:00
04d58ef75a String->rtl::OUString 2012-01-24 16:53:55 +00:00
2669d048bc make VectorOfOutDevStates a class and refactor this a little 2012-01-05 09:18:17 +00:00
3b8bf60a04 Revert "callcatcher: Remove unused code"
This reverts commit 070eff8cf1ad7763b8b730336f11032893b77049.
2011-12-21 06:46:56 -05:00
070eff8cf1 callcatcher: Remove unused code 2011-12-21 06:25:31 -05:00
d498a3c2a1 callcatcher: Remove unused code 2011-12-19 23:19:45 -05:00
2acd8fbac9 in modules, when we have a env we are in stage gbuild 2011-12-07 15:53:06 +01:00
e10fc18fb2 typo fix: explicitely -> explicitly 2011-12-01 23:41:42 +02:00
d73e574d8a move reconfigure into gbuild 2011-11-29 17:25:14 +01:00
d169860411 more debug info for emf+ rendering - path 2011-11-28 10:23:51 +01:00
c0bed9c72e remove precompiled_xxx.hxx/cxx 2011-11-27 18:07:55 -06:00
3f2ff54778 remove pch from the include list 2011-11-27 15:24:45 -06:00
c0f6713f94 remove include of pch header in cppcanvas 2011-11-27 13:02:39 -06:00
992757cc5e remove pre-compiled header support in gbuild and gbuildified module 2011-11-27 11:49:38 -06:00
f2f0dbd273 make gbuild makefiles run independant of pwd again 2011-11-25 12:04:40 +01:00
d63d64c4de empty a bunch of d.lst files 2011-11-21 22:46:17 +01:00
828abf1e06 Revive gradient clipping debug visualization code. 2011-11-21 16:37:02 +01:00
469c9ac2f9 Enforce a "lib" prefix for UNO components for Android
This commit for the old build system. (Don't bother for components not
relevant for Android.)

The Android package installer (as invoked through "adb install", from
"ant debug install") silently ignores native libraries in app packages
(.apk files) whose names don't start with "lib" and end with ".so".

The package builder (as invoked through "ant debug") in the SDK gladly
includes also thusly named native libraries in the .apk, though. Yay
for consistency.
2011-11-17 15:22:59 +02:00
0ea8c014ef tweak gbuild standart Makefile to allow partial build in unsourced env
This allow to run make in a module wihtout the need to source
Env.Host.sh.
2011-11-16 06:18:54 -06:00
76e37b7f79 adapt naming scheme 2011-11-01 09:27:24 +00:00
67cf9cbeeb add and use an OString::equalsIgnoreAsciiCaseAscii equivalent to OUString one 2011-11-01 09:27:22 +00:00
8e9ad03830 Related: fdo#38580 NULL object 2011-10-10 16:51:55 +01:00
946090359b Added the unreachable code back in, even for GCC 4.0, so Mac OS X would not fail. 2011-10-08 23:47:34 +02:00