...that looks out of place after previous commit.

Convert the IDL files in the udkapi module, and update the
types.rdb database.
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>

...as only "build" passes through instsetoo_native, where some files (pyuno.py
etc.) are copied into places where ooinstall expects to find them.

Fixes that if more than 84 empty but formatted rows at the end of a file
have been found the export skipped those rows.

The way alpha compositing took place in vcl's alpha vdev was subtly
wrong - it was supposed to implement porter-duff 'over', but didn't.
This is now fixed also for cases where the source surface contains
alpha, and needs to blend correctly into fully-transparent
background.

For reference: http://en.wikipedia.org/wiki/Alpha_compositing

3236c2c3 accidentally removed the gallery
item count which was used to differentiate the gallery symbols from
normal ones.

To allow my editor to do static analysis correctly.  Those IMPL_LINK
macros screw it up.

(As a low-hanging fruit to drop the shared library count by
one... Each library less counts.)

This reverts commit b5ca34fa.

829a9b9c "In Java, css.uno.RuntimeException is
not derived from css.uno.Exception" is a better fix.

which lead to Bridge::map_to_java (jni_data.cxx) copying the Context member of RuntimeException
twice, once with a fieldID for Exception, which fails on Dalvik.

...that are (also) used during the build (e.g., cd testtools/*/lib &&
./bridgetest_inprocess_java).  Broken by gbuild'ification of jvmfwk.