Change-Id: I5840d47047880636fb4e8f80ecef1a6f5377d786
Reviewed-on: https://gerrit.libreoffice.org/6840Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I4402b0b259c0dcd03a222993c861114028d6fb75

Change-Id: Ifb91f80cd9e8ea1cb9431b4f38d5afc9e7b9a86c

Change-Id: Ibf3be46e8619c2dce9636e8a90a6703f3e28ba33

Change-Id: Ie7f8d37b550c788fd6235a5de4bbcc51ac6f8150

Change-Id: Ifb32c09bee2e43f3412851e531175cd214bb420b

Don't use a mutex and condition in each ScFormulaCellGroup.

Not sure at all if this crack works, and I hate every moment spent on it, but
at least it is still behind an environment variable.

Change-Id: I70c917d47112e3a2ed010de4d123a4155f71c004

Commit 9389cf78 (cp#1000018 RTF import:
empty para at the end of footnote text got lost, 2013-11-15) assumed
that \par at the end of all substreams means an empty paragraph, but it
turns out this is only true for footnotes/endnotes, but not for e.g.
headers.

Additionally, that commit actually causes a memory leak, which is
detected by an assert in the SwIndexReg dtor, so crash-testing caught
this.

Change-Id: Idfa040bf3026a9515a120cd7afaf7d314553a131

...since 0bc89aac "cppumaker: Allow UNO
interface functions to throw std::exception."

Change-Id: Ia28f02ea10d45267fd951e80da01ffdfdb420885

...which it effectively had been already anyway.

Change-Id: I6c9ae34f59e8bd44a6be38e6e719ad0a96773c1c

Change-Id: I2f7afe172221dce2de4f5d9cbef84de95ab6078c

Change-Id: I3df112c840ccac71fb34d39900b3be3630b9c5a6

Change-Id: Ibbc34014f91c8f1ce73a6e9a8f61fa421fdb9144

Change-Id: If4ad335f4e645cc10cd7378d971b7389bb437f88

Change-Id: I76499e12720bd35793a8b616899c8d4a04852171

Change-Id: Ib3e43e14f9223bc53f20f1241d9b8c0eaa5470ad

Change-Id: If26bdd6ed0be59a4fff137ac3166b9d644e1aea7

This reverts commit a74b3e89.

Change-Id: Id670a2b66e61d3a4dceb9640d6ae34e9116691c0

Change-Id: I031409043f12bdfe4b8c838d74db8a60576fd507

Change-Id: Id26dcf8146dc7a41ea9995ad4b1f89925c08004f

Change-Id: If3d9feb25d825806cb690d001872f1d0ff76466c

Change-Id: Iae1fa695ea0266e97ae367db71fe6a82e6a3da1d

Change-Id: I4dcc7244e508fcdf484b73dc843b21f4b0a18a73

Change-Id: I954606a9d6d9b0fc66f2f3496e204c8e6496db88

Change-Id: I6d4d7cd2dc426134c108f1fbcf6e3c32ed4af88d

Change-Id: I48141ca77277c2de5b3526358cfadb12e3f4fbdd

Change-Id: Ic70aaea461e3d007e394c67c33a766d70aa4e92c

Change-Id: Ib3195870e708db49448d68d262611561c68ddcc6

Change-Id: I34299e0fc692c2ab43fe0c1fffaada348742d19f

Change-Id: I0d53dcf147a06f831120c5489b2975e42ea6120b

Change-Id: I885a222bf8a7ca4275867585ccf067a1ba7822c7

Change-Id: I09ffa455e79199e37ca6168753ec9ef4d5aa33dc

Change-Id: I81f98ca83b97bc2e3e419c7a37ad2a011932553b

It is pointlessly slow and the AccObject has a method for that.

Change-Id: I7ba1cc853255ac3b3b5008657d1cc79efc3a3f4b

It's disposed iff m_xAccessible is 0.

Change-Id: I3f8bc104784adf82be43bcaf43b7573a9897f82a

... and clear it in disposing().

Change-Id: I69cfe3cdcf6549a0471453960d1f935d9c3b7a0d

The manual acquire() call was apparently not coupled with a release()?

Change-Id: I069c969619e9afce2a4b836642cc1675025b11d0

Change-Id: I68e5c1785b73520cb59ca54a602ea176f8146f6f

Seems safer to store a Reference

Change-Id: I86ef9d18c03b2a58058d38e2ae43553ecd2c7be6

CoInitialize{Ex] always exist in all supported Win32 versions, so just
call them.

Change-Id: I45f227319aae811b6c94a3ac475467a06146b8e9