of group area listeners.

Change-Id: I3a8d2ceb57462c2c2c4cddb6e34966d2465e149b

Change-Id: I75412790e0afc1dbe3255254ff13fd41c5e9934b

"static" keyword explains why first time the option is well taken into account and not the other times.
So let's remove this and the option will be taken into account all the times.

Change-Id: I9db84373e425dfe21c307b965b1796c781cd7863

Change-Id: I210760099eb630c1306630e49bf84a2f3bf7a3b6

Change-Id: I32a9d6cc625537ab6e0940d93bf9a7544df370c8

(regression from a2feb420)

Change-Id: I211910ba12fa1f2ce578bf467169a239ddf83e1f

Change-Id: I4c8aa46a02799dd8e8cedda0dd552b51b248a0e8

The change was broken. Luckily MSVC noticed: sw\inc\doc.hxx(1707) :
warning C4717: 'SwDoc::GetNumberFormatter' : recursive on all control
paths, function will cause runtime stack overflow.

This reverts commit 7ca3303a.

Change-Id: I6ccd990ba71867f0ff96ad06cf5a9e7c98c93657

Change-Id: Ic7ecfe85826d9148d14fce14e9de3f80a11eb775

Change-Id: Id134b05de3a81dd246baf8743c32fa8eb1e37935

Change-Id: I01f66eb78e357780f35cfd1df6722301cc5753b2

Change-Id: Ifb7e86b078bd549506a9cc1b9ce9fc22fffc5eec

Regression of 10fc1383 that exchanged
the PostPaint with SetDirty, but the paint is also needed.

Change-Id: Ic3e81fa106f405568effaed71ab33b575b4c5c5c

Change-Id: I86d1c3e0c177c671b280c05b47312c79389884ed

Add a work-around for Writer's usage of LONG_MAX to SvxColumnDescription:
on 64-bit platforms the LONG_MAX added in SwView::StateTabWin() for
SID_RULER_ROWS will overflow in vcl's LogicToPixel mapping and that
causes wrong positioning of the row highlight lines.

Probably Writer should use something other than LONG_MAX (no reason why
these types need to be bigger than 32-bit anyway) but that needs a
bigger cleanup.

(regression from 4c60f722)

Change-Id: I08147462356368d48959a85a85ef7dd8dcae0943

Change-Id: I009e3a506d9b3418215341cb7f3e232e7fcfa04b

Change-Id: I208bbe47d1134b6a362e50d54093e6d35e85f51b

Change-Id: I0b7ab0ed27ba23f9ad924819193dd30c70e48fed

Also tweak some of the assertions so they don't fire for character
borders.

Change-Id: Id63631864ab3e5f0366a514adc1ff73f01f57a84

Change-Id: I7f3ed82652e98da767cf7021bc5fc42e31e80b73

usage (5 test runs): vcldemo --test 5

Change-Id: Ib8c3db1e74f4a05f16a5359fe8645ffc61e891d5

Change-Id: I36ef1f24567b9e561f2bd5e981cbd60dba1343d1
Reviewed-on: https://gerrit.libreoffice.org/13103Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

the tests are done headless with dbgutil enabled and
this dialog contents dumped in dbgutil mode tries to
spit out the contents of the various "file format
not known", "file busted" dialogs with some bad casts
and falls over and dies

Change-Id: I353619f0209902d4c099b42064fd48c3c547f846

Change-Id: I89b3b175d75f27704dfcd1d11f953f0bfca5b60c

...and not only for instdir/, as was done in
ab67b79c "set OOo reference version to 4.1 for
incompatible MessageBox changes."

Change-Id: I58931982153e619c678adde6cf01ec461dc0d2f5

Because under Linux use text
"'"
and font "Nimbus Sans L"
and export as pdf and the output is garbled

If we do want to stick WinAnsiEncoding here then we'll presumably have have to
test that pEncoding actually matches WinAnsiEncoding and/or generate an
additional Encoding/Differences from WinAnsiEncoding for that font.

This reverts commit eea16cb3.

Change-Id: Ie84ad526f6756e996e1955610d0a9c627d1e0c8c
Reviewed-on: https://gerrit.libreoffice.org/13060Reviewed-by: Samuel Mehrbrodt <s.mehrbrodt@gmail.com>
Tested-by: Samuel Mehrbrodt <s.mehrbrodt@gmail.com>

Change-Id: I506c38dd5469e4301985de77f7f75632dfdde817

When undoing row deletion (and possibly other similar undo's).  And
write test for it.

Change-Id: I04b4fd9932f4236f124dcd25967355c6055dec33

Change-Id: I3f538668c9c5f1727b8cf49ab72fd5879e12f156

whether there *is* an id, or whether there is not an id.  An id of 0 seems to
be used as a flag that an interface is some sort of superclass which can be
subclassed by something else and have its toolbars etc reused.

Convert this to a bool so we don't need an a resource id for each one and just
drop the third arg for the normal "final" case and use a different define for
the inheritable case

Change-Id: I98380f03d73d57bf8cba02d339097e384518abaa

Change-Id: Icd9ff99616c452bd16ffc3dc8ab0981d75a386a2

(which now had a bogus value on Mac OS X anyway)

Change-Id: I5bf233506d6c658560dfadddf79a9ed8605c2a20

Change-Id: I04a446e6b8a8352be9f091980bca31842bb7e643

... as dialog buttons.

Change-Id: Ieea882b467c87093e7648590bbf14fe6dfa0b7ee
Reviewed-on: https://gerrit.libreoffice.org/13061Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

* Remove unnecessary containers and unhelpful frame
* Center main container
* Right-align labels to remove large gaps between controls
* A bit of separation for the Help button

This should serve as a model for similar dialogs.

Change-Id: I1964e8909159506484950df9d8ef7283809cce5f
Reviewed-on: https://gerrit.libreoffice.org/13057Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I55648417f56c28b0ac07ebff908805764bb2eb0c
Reviewed-on: https://gerrit.libreoffice.org/13059Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I0e26b9e13e69ed49e7bbbc7f2ac0b04941ee1da7
Reviewed-on: https://gerrit.libreoffice.org/12974Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: Ib9ce9a28eec4343850158541375608d63fda8d19
Reviewed-on: https://gerrit.libreoffice.org/13050Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>