Seems poppler changed outputdev virt func signatures with version
12, so with system poppler, we never got any image import functions
called. Added override annotation to prevent this from happening in
the future.

Change-Id: I565de80ce218e831e276d3a1d88ded3baee876e7

We shouldn't be trying to free the font table buffer when destroying the
blob, so pass NULL for destroy function instead of free(), and also use
HB_MEMORY_MODE_READONLY just to be safe.

Change-Id: I85b5a575249b4efc0f5799db205ee17cbeb66d22

Change-Id: Id7cb705b2b719cdd3bf16b8aba858c0319686599

Change-Id: I7a6ee4dce338b7529fa7500bb0c8341a4f2ebc22

Change-Id: I25794499711845cb8183664d1345e5513587d81c
Reviewed-on: https://gerrit.libreoffice.org/3679Reviewed-by: Tor Lillqvist <tml@iki.fi>
Tested-by: Tor Lillqvist <tml@iki.fi>

Change-Id: I6ddeff7ae72693fd0951d71bd7b5444938bc7286

Change-Id: Iaec745dc1c8eb0614cc2fe1d70a94a00d18cc934

Add "$COMMONPROGRAMFILES\Merge Modules" (on current versions of Windows
that probably is "C:\Program Files (x86)\Common Files\Merge Modules") as
fallback, since for example on Windows Server 2012, the registry entry
has not been created when installing Visual Studio Express

Change-Id: I16f58b8a1d453c8628c6eb66c2cea37ab3da913a
Reviewed-on: https://gerrit.libreoffice.org/3680Reviewed-by: Tor Lillqvist <tml@iki.fi>
Tested-by: Tor Lillqvist <tml@iki.fi>

Project: dictionaries  acf0ae3fdec65df75f4983d2b943582f9ab99fa2

The things are not even enums, so there's no error checking either.

Change-Id: Ib33cd96b49a840cbe486dbed9a7c370827b54c2d

And another step closer to opening my test file correctly in Excel.

Change-Id: Ib6aa8ddb2f2792513c7263d86865e08fe5f1483d

Change-Id: I12ddd18c1b4476d42c097704c2af2d0af4e15a1e

One step closer to opening my simple test file correctly in Excel.

Change-Id: I1bfa76a46081ba478e44d8d90d8a91790a9a9ff3

The extra semi-colons were confusing clang. Just remove the macro,
it's not adding any real value.

Change-Id: I7f7c312d2c0167a17e31624e197ae75b77fa0c2f
Reviewed-on: https://gerrit.libreoffice.org/3677Reviewed-by: Tor Lillqvist <tml@iki.fi>
Tested-by: Tor Lillqvist <tml@iki.fi>

...since 81af84c0 "gbuild: let JunitTests run
with MSVC."

Change-Id: I4ba97431c3439c0bed74149fe66cbb76db5317ce

Change-Id: I36c62fc677de70c08d0d59e45debc6a86375d3b7

...rather than a TypeDescriptionManager service that is no longer available.

Change-Id: I1054ab0a658597a60d11e2f9aa0960df25d14bc4

Change-Id: I7c750debaeee5d9804be378e0f65bec13ac296eb

Project: help  eddac30b5afa84c2420a553d2e822722dc857228

Let foo.odb be a database file that has a macro that connects to the
Database on "Open Document" event (and needs to prompt user for
user/password).

There was a race condition between two actions:

 1) the asynchronous treatment of "OnFirstControllerConnected" in dbaui::OApplicationController,
    which tries to get dbaui::OApplicationController's mutex

 2) the StarBasic macro calling dbaui::OApplicationController::connect
    which needs to display a dialog (to get username and password),
    and thus puts that dialog in the main thread's event queue
    and waits for it ... with dbaui::OApplicationController's mutex held

Now, if "1)" is before "2)" in the event queue of the the main thread,
*but* "1)" is executed *after* "2)" has taken the lock, there is a deadlock.

Fix:

 1) Make OnFirstControllerConnected synchronous.
    Make sure (by taking mutex in dbaui::OApplicationController::attachFrame, its ancestor in the call graph)
    that nothing else will happen with the OApplicationController as long as it is not finished.
    ---> it does not need to take mutex itself anymore

    This avoids the "order in the asynchronous events" dependency.

 2) Change dbaui::OApplicationController::ensureConnection to do the user prompting
    WITHOUT HOLDING the mutex, and use the mutex "only" to protect actually assigning
    the connection to m_xDataSourceConnection.

    Theoretically, in some race condition, we could connect twice and then discard one connection <shrug>.
    ensureConnection will never return the discarded connection, though.

    (I think I got that right with respect to http://www.cs.umd.edu/~pugh/java/memoryModel/DoubleCheckedLocking.html)

    This keeps it from locking on another condition while holding the mutex.

Change-Id: Iab1bbec5d5df12bb89d027d43e498c78c92ffc32
Reviewed-on: https://gerrit.libreoffice.org/3310Reviewed-by: David Tardon <dtardon@redhat.com>
Tested-by: David Tardon <dtardon@redhat.com>

Change-Id: I2555dc113b2fb4c77bec187b15f0bb545811b07b

Change-Id: I7a4f782235a23f50bceaef041a92e42e47860a78

Change-Id: I55447aba5abcc8205543c7ca64763b5c99854837

See the comment in climaker_app.cxx main about room for improvement.

Change-Id: I11cc2f62c9d42269d5c96253198ee4e730496e75

Change-Id: I2ff65979a9da7da12ad6ac97dc090898913a2e8c

Change-Id: I181c87f8406f095bd4dc499b47d91081f58aaf59

Project: help  7584ed08efd8e9cea733a782544a1e1eeb48433f

Build tools do not link against merged libraries but individual small
ones we need for this (gb_BUILD_HELPER_LIBS).

Change-Id: Ia765e11a93ed05d126334a3e146fb88e368954ac

Change-Id: I801f4af269efcea91bc3bafbfc9e0b9033488f2d
Reviewed-on: https://gerrit.libreoffice.org/3645Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: If6caa7abed25ce95f1c62f9673619426eb49b5a8

Change-Id: I47f1a72fac3e6678b5dfe880f2f189ffc436333b

IMHO there is no reason why they should be 0444. I have found no
explanation for it, either in the commit that introduced the bundled
dictionaries or in the related bug.

Change-Id: Ia42218a0d579ced5f17248a092eab2c61cb9005f

Change-Id: I3edbae98c9bb51899e5592fae1cfe9f3632ad535

It turned out that ApplyDXArray() is need to apply advance width
adjustments after justification, so we can't just bypass it.

So I just copied GenericSalLayout::ApplyDXArray() and stripped it of
ICUism so it does not break with HarfBuzz, but I had to make
m_GlyphItems non-private, so I'm not sure this is the right approach.

Change-Id: I66d647c3590fdf912c39d0cf23ac72bcc7ca72c9

The 3rd parameter to hb_buffer_add_utf() should be the length of the
whole text not the current run, so that HarfBuzz can take the context
into account when shaping.

Change-Id: I9e4e928d40078c3e3667cfdb6d8f24bf6e58263d

No more second guessing if text width, we know that information already,
so pass it around instead of trying to re-create it.

Change-Id: I19faacbc309d38753c3c9f7214dfa0bf59cc66b5

Not that I'm a fan of Hungarian notation, but for the sake of
consistency. Fix some placement of opening braces along the way.

Change-Id: Id6ea758fd438a4040e7451430a0f3a166efdec43

Change-Id: I78efebb576dffa8d39e98283feb9aab2186b5a39

GenericSalLayout::GetTextWidth() uses GlyphItem's mnNewWidth when
calculating text width, and though this seems logical (it is after all
the actual with the glyph is contributing to the all over advance
width), it results in shorter width calculation whenever glyph width
adjustment is involved, no idea why!

The #ifdef is there so that the ICU code path is not changed in anyway,
but all of this should be merged into GenericSalLayout when ICU is
gone.

Change-Id: I7cbde1675b78e87c142513eb52a13ee5fdc60617

Use local variables instead of altering the returned glyph positions
array, looks more cleaner this way.

Change-Id: Ibbcced57777010bd045668a99d7beb0618abe226