Change-Id: I83a4332d9947d03382b10ea050f26bf3ed544299

Instead link against the library again.
Unit test is now closer to what the real app looks like.
Only 4 more functions exported because of unit test.

Change-Id: I6fd784725564e577658d68ff7fe0595a79cf7639

Change-Id: I46621589f2e1282fbaf369bdb387bb63c5768206

- killed GMailLoop, seems to work with g_main_context_iteration( NULL, TRUE);
- added test for sending and also receiving file

Change-Id: Ida11f875709dab71d404697e8c39fe4548f653b0

Change-Id: I7705d627d17f20c3c81de7aa4debbefc127354b0

Change-Id: I1d69be65e2d55f1de0a3d800cd920cc881e8a8ce

Change-Id: I424a3e91cd6163907e364c7d6f40e19d8bf12b48

The color setting was picked from SplitWindow::Paint()
for visual consistency with the sizers used in Impress
for example.

This commit also resurects presumably dead code.
In fact GDB let me see that only the short path
was taken.

This has an impact on Calc's window splitters and
on more visibly on the basic IDE which now looks
much nicer :)

EDITS:
 restricted this commit impact to IDE as a poc

Change-Id: I9a051fd894feccff4bbc7db7efe1160cb2c3a2c6

This rule failed if the MISCDUMMY file did not exist yet.

Change-Id: Ie9b1ecab2848f5ed2c459020a50013db691a9ca3

Change-Id: Ib291d5b053f9ed7b597f0ebbed725602430e36b7

Change-Id: I7162523a26488b8528a91ec7abd587f55647acbb

Two cases that 68331237 "Removed unused
SfxMedium::bDirect" did not catch.

Change-Id: I33b60bd27b97ac086d0927a6877d055561dde1f6

Change-Id: I048b0d4f28a92c23523e8f53993632238ebfa622

Looks to me like the intent was to call SfxViewFrame::GetFirst(NULL, false)
rather than SfxViewFrame::GetFirst(NULL, true) to which
SfxViewFrame::GetFirst(sal_False) would translate.  The call was in that form
effectively since it was introduced with
7c896369 "Obsolete Objects/Classes removed."

Change-Id: Ib7d9ce75d946291bdfd43420bdd8a1de533cdc98

Change-Id: I20d0bd3eeb7661b51193ba01042e15ef62629ed3

Change-Id: I57abfb3b3d659edcd130788d173c2df3b79101f0

Change-Id: Idf2ec39ba286d359903edb6be60f2692c6a9837b

Change-Id: I4b877ab97259f7e031e34daf7ce93e1b59417a08

Change-Id: I156255a5691beedec0cd00db0c772dac8ecfa564

Change-Id: Ia6f5bf7c3372515c2796d239aeb610fbe6f29f8d

Change-Id: I521e2c816b3dbc4aba1770e5a3b5eb47d5542ac1

Change-Id: I87875922d952387d9ffd2667dd621d5cd94b4ea4

Change-Id: Id40b2fb6f8e1667bcb5164e483ea7c6949b627a1

Change-Id: Ifee45475f8cfeee6b2a4f880f0ff1cc7f9024715

Change-Id: Ic3bbb20c446545d67682a6c547fb69e2255eb0b7

Change-Id: Icb567a848162bcee072476e817cc82278bddfd99

It appears this came in to make it work with JREs < 1.4, i#58308.

Change-Id: I6974e0fe7b94ecf1b577af064dcce8f017d0a31d

Apparently Xalan was introduced years ago to have a DOM implementation
with JRE < 1.4 (i#37179), which then got automatically converted to
saxon years later.

Change-Id: Ibc13b84331854926965f35e3afe25e923bad821b

Change-Id: Ib62473d841bc9a66acde6529d3f63e3fd1a00928

Change-Id: I9c038e6fd1c60e4ac77740c55a45439d813524e0

With gb_Jar_add_jar and gb_Jar_add_system_jar adding to the manifest
classpath automatically it is no longer necessary to call
gb_Jar_set_jarclasspath manually except for the URE jars, which
are apparently not supposed to be added automatically.

Change-Id: I1e743e7ecb9cb5651e02005aa09e127bea1b0a29

Let gb_JavaClassSet_use_customtarget add the customtarget workdir to the
classpath.

Change-Id: I836e890b43bb2ca06d19cf9f83a5fa8f735cf963

Change-Id: I549888fcfe08d83101e414de86e8c9db1357d6eb

Using a system jar should really put the system jar onto the manifest
class-path, but that is currently difficult because externals are used
by JavaClassSet and not Jar; fortunately it seems nobody uses
gb_JavaClassSet_use_external directly so move the implementation to
gb_Jar_use_external.

Change-Id: Ieac480b3dec0ad39d2404b0a64b641eaf73c1e4a

Change-Id: If3db4dea84448e9d7ee26d797a25f7e4aa4d5941

Refer to this thread for details.
http://lists.freedesktop.org/archives/libreoffice/2012-August/036852.html

Change-Id: Id2115732a1e1bcd78c13445f5aaefe1ec7cfa0c2

Change-Id: I770ff1d77102cfbe63849babed6bd9406d958b09

This is done by removing caching in the AbstractCoverFlowImageAdapter.

Change-Id: Ib6df57a73545f8a98b4104a6e2917121e9d7afa8

discovered 2 bugs with the previous patch,
a ) need to properly calculate the topleftmost position for the combined shapes
 in the group
b) when you have groups within groups then the offsets need to be applied to
the topmost parent

Change-Id: I2edc0cccca439b3103b61c8116d1cde1536519da

Change-Id: I93195124aafd8472671a2cf4edddd82c37789016