Change-Id: I2d1aa79a67bf130c3844b3bccb34a1761e8e4dfc

Change-Id: I030321e8cb905be78b68f9e404926e22723632ef

This clobbers the functionality from commit:

    06d73142

    make idle timers actually activate only when idle

Since now all rendering and re-sizing is done in idle handlers it
does effectively the opposite of what was intended. A better solution
would allow special-casing the processing of just rendering,
re-sizing and window management to spin for eg. progress bar
rendering.

Also add helpful debugging labels to the idle & timeouts.

Also cleanup the Idle vs. Scheduler handling.

Also ensure that starting an Idle triggers a mainloop wakeup.

Also add a unit test.

Change-Id: Ifb0756714378fdb790be599b93c7a3ac1f9209e6

- ./logerrit testfeature will trigger a testbuild of the
  current feature branch as it is on gerrit
- ./logerrit testfeature $BRANCH will do the same for $BRANCH
- cloning to a tempdir and rm -rf'ing around isnt ideal, so
  consider this as a template for your own scripting

Change-Id: I50a3c80748af82d855522f245ddbff227c8adab2

Change-Id: I8ed4b4d221040ddf5f51d7e01134c855fc85a641

Stop pointlessly inlining these virtual functions.

Change-Id: I11d123bc704ec6cf54925a13d139fce70cc59566

This reverts commit 8761ab9b; thinko, as the
RTTI needs to be visible with GCC too in the test code.

which is what GetTransparency() returns. Iow, do what SvxBrushItem
does :)

Change-Id: Ief65ab75d8c0955c1ec264611a70dcc693412170

Change-Id: I402bd7319d31859510a06644eb7dc786d56b312e
Reviewed-on: https://gerrit.libreoffice.org/16213Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
Tested-by: Noel Grandin <noelgrandin@gmail.com>

Change-Id: I62d44e29cd4376f17270cef548d2cec6f4b621b9
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Change-Id: I7c41f1d596946706dde4adbeb5434ae0be58b04b
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Change-Id: I1937374b56b506681876d100112852b68f099038

The build relied on the presence of ANDROID_HOME envvar, now we pass
it explicitly.

Change-Id: If33ee2c6a22435151aad244378ee5aaf7b3cc913

The fix is twofold:
1. retrieve transparency from colour in SvxBackgroundColorItem
(add QueryValue, PutValue methods, use additional memberID to
retrieve alpha channel as a bool property)
2. add CharBackTransparent bool property to Draw [text] shapes,
xmloff needs it to be able to output 'transparent' string instead of
'#XXYYZZ' colour code in ODF format

Change-Id: Id830bd18fd6da5d9a46e2a94aa254d5d2a6b7ebb
Reviewed-on: https://gerrit.libreoffice.org/16182Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>

Change-Id: Ife3131eac8c653bb851f44f05be618289b5c5116

Fix the following cases:
- two names for same ID in sd/inc/app.hrc. Two ID defined.
- ID defined in include/sfx2/sfxsids.hrc and redefined in
  sd/inc/app.hrc. Only the sfx2 ID is kept
- ID defined in include/svx/svxids.hrc and redefined in
  sd/inc/app.hrc. Only the svx ID is kept

Change-Id: I21c7c25cb6db3c43ea2f9aac5aac75fabe61eaf0
Reviewed-on: https://gerrit.libreoffice.org/16209Reviewed-by: Philippe Jung <phil.jung@free.fr>
Tested-by: Philippe Jung <phil.jung@free.fr>

Otherwise the redlines are on wrong positions in the tiled rendering case.

Also - we are actually never call with something else than MAP_PIXEL, so
let's kill the resetting of the mapmode; let's see if some obscure corner case
bites us here.

Change-Id: I8da07dd5448b8863df3148882ec50e6746760cca

...to make typeid comparison in sw::UnoCursorPointer::SwClientNotify work also
under the Itanium C++ ABI/libc++ semantics (as used on Mac OS X), not only the
less strict GCC/libstc++ semantics (as typically used on Linux).  This caused
CppunitTest_sw_uiwriter's testUnoCursorPointer to fail on Mac OS X.
(DocDisposingHint and LegacyModifyHint need to be SW_DLLPUBLIC even though they
are nominally only used inside the sw library, because they are also used from
the CppunitTest_sw_uiwriter test code.)

Change-Id: Ideceade22ca3049ae287aea468de93c72bee5fdc

Change-Id: Ic79c0f1cf2816da60962f67c57efb75685d984f1

The items count of NumberingTypeMgr is limited by 8, but
some languages have more items, so remove that limitation.

Also fix the behavior of the 10th item. The old sidebar
popup had a "None" item, whose id was 10 (defined by
DEFAULT_NONE from include/svx/nbdtmg.hxx), and that causes
now the 10th item to behave as on/off toggle. Since we
don't have a "None" item anymore, get rid of that behavior
completely.

Change-Id: I5546ca637f9951cc093a7bbf8fda11dc35341d4e

Change-Id: Ia7f99123625df28cd5459ee6d488e02b20d386f0

When Base is embedded in a Writer document, we still need this when the
document is saved for the first time after embedding.

Change-Id: I1103da01838abd4ac1b03dc4d44e10db3ce650bc

Accidently removed in commit f01f3120
(SwDBManager: extract StoreEmbeddedDataSource() from registration code,
2015-06-09).

Change-Id: I27e47249b92b8fd19dcd09a1226f5e80184ea8db

The CPython changelog says for version 2.4:

- Python no longer relies on the LC_NUMERIC locale setting to be
  the "C" locale; as a result, it no longer tries to prevent changing
  the LC_NUMERIC category.

Change-Id: I6d63c4dedca48c99bf81135d69285d9116a19740

Change-Id: I7caa540d069addb610310830c26ecd647afc6538

Change-Id: I9490d888ad148caecba034ef12b7398ae5c66a3f

Change-Id: I0081113bf312287fe6e14749a569709ee892d8dd

Change-Id: I78478de91473681fd8d836293dabe7b790dda516

the enhanced-path contains decimal points so the
last element is a double not an integer

Change-Id: I90be76a2d4cb90ee7a904aa72fe65770c675fc53

Change-Id: I97d74eebcb45f413dbc73ab96db447228eeb158f

Change-Id: Id3cf58660bfcaf7424d027773894ce6888df788a

...that was introduced in b88a62cc "CWS-TOOLING:
integrate CWS dbaperf2," apparently to speed up later activity by pre-loading a
JVM or Calc.

The problem is that any spawned FastLoader thread was never joined, so could
still run during exit.  That caused many spurious failures of
CppunitTest_sw_uiwriter after df726c8b
"CppunitTest_sw_uiwriter: add embedded data source definition testcase"
triggered spawning such FastLoader threads there.

Adding code to join any spawned FastLoader thread is not easy, though.  For one,
it is not clear whether there would not be very many (terminated, waiting to be
joined, thus still holding up system resources) such treads when doing the join
in ~DatasourceURLListener or DatasourceURLListener::disposing.  For another, a
naive attempt at joining in ~DatasourceURLListener leads to deadlock, as it is
called with the SolarMutex locked and the E_CALC branch in FastLoader::run
blocks waiting to acquire it.

Thus, the best approach appears to be to remove this optimization.  Hopefully,
that does not have too dire performance impacts on typcial user scenarios.

Change-Id: Ifeab013ddd8acee1289423facba0c120684e9325

This old-school table has three rows in it, but the second row is of 0 height
so is indistinguisable from the third row by layout positioning, so the
WW8TableNodeInfo view of the table is that it has two rows, and the comparison
of being on the last row is done with the m_xTableWrt->GetRows view which
considers it to have 3, so the table end marks are never output.

add a new finalEndOfLine property that the WW8TableNodeInfo sets on the last
end of row of the table as it sees it, which should resolve this.

old style tables are a cess pit, which is why they were replaced

Change-Id: I996aa59a338a594487f49ec0f228af3fb3032d15

Change-Id: I689783e176921ea03475d3e113426563bf20b9d7

This has become necessary after edb38d70 "update
to liborcus 0.9.1; Includes switching to dynamic libs."

For one, pass -Wl,-rpath,$ORIGIN in via LDFLAGS, as is done for various external
projects.

For another, libtool (as internally used by liborcus) unhelpfully adds further
-Wl,-rpath arguments of its own, to make the resulting lib in its presumed
location find its dependent libs in their presumed locations.  But we override
all that to the single instdir/program/ directory and corresponding RPATH
$ORIGIN anyway, so the simplest fix to get a clean RPATH appears to outsmart
libtool's attempt add adding further -Wl,-rpath arguments by setting
hardcode_libdir_flag_spec_CXX to empty (see the internals of libtool for why
that works).  Other external projects (like redland) suffer from this problem
too and currently have a "dirty" RPATH; fix pending.  Until libreoffice-4-1 that
was generally not an issue, presumably because having all libs in solver
happened to cause libtool not to emit any additional -Wl,-rpath arguments.

Change-Id: I80dff5141a33c198f30aecb7c51b41bc06145f1c

Change-Id: Ia73ae54068d208d6b2db3439f38402664bfa8d29

Change-Id: Ieff88d5de119db30c3fbd2f58ba52696eeb3d327

Change-Id: I60c7540241c41f5063736f2a3d4817371411c8dc

Change-Id: Ic25e9aca7a2a61ac677720427fae7a18a8a179fa

Let's hope passing an empty NSDictionary works fine.

Change-Id: I90a9b09d421ae8cd202b602db97b2c08b149fd07