Age | Commit message (Collapse) | Author |
|
Change-Id: Id096bd25cd6df5941eaec45f7573f5887ae51d79
|
|
This introduces a new thread that takes care of handling the IPC
command and result for executing the file picker
('Commands::Execute'), which is the only command in the file
picker that may block for a longer time until it receives its
result from the kde5 side (namely, when the dialog has been
closed).
While the file dialog is being executed, activate a dummy GTK dialog
that also takes care of handling events in its main loop as long
as the file dialog is shown. The dummy dialog is closed together
with the KDE file dialog.
(Since the actual KDE file dialog is run as a separate process,
this one is mostly "transparent" to the soffice process from the
point of view of a dialog.)
This allows dropping the custom event processing previously done
in 'Gtk3KDE5FilePickerIpc::readResponse()' that had the potential
to cause all kinds of problems, e.g. when another event related to
the file picker was triggered from a Java process via UNO.
Change-Id: I3d663253f09320f7a8e0d9ec32a8fd6ec191c189
Reviewed-on: https://gerrit.libreoffice.org/61253
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Move reading IPC commands and arguments to those commands
to a separate thread and use the signal/slot mechanism
to notify 'FilePickerIpc' whenever a new command and its
arguments have been read.
This allows to handle the events of other commands that have been
received while the dialog is being executed.
This will be needed by a subsequent change that will modify
how IPC is handled on the gtk3 side.
Change-Id: Ia77b21045b0196710cbe164fb640b36a128d5081
Reviewed-on: https://gerrit.libreoffice.org/61252
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: Iee80b40d2c33a0464abfc285360e76774c644b32
Reviewed-on: https://gerrit.libreoffice.org/61372
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
Change-Id: Ia26896cbcd8912f7cad641ea2ecf14c4ddf2d741
Reviewed-on: https://gerrit.libreoffice.org/61489
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I58563735e6cf5004b4ac5e846885414f0c603f4c
Reviewed-on: https://gerrit.libreoffice.org/61491
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I6074035ed8f90e452915e9ecffdbe9363375e126
Reviewed-on: https://gerrit.libreoffice.org/61306
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@cib.de>
|
|
Drop the special / duplicate glyph rect bound cache implementation
for Freetype fonts. The GlyphCache class now looks a lot like the
ImplFontCache class.
The lru handling for Freetype fonts is based on the byte / memory
size, including the glyth bound-rect caching. Using the new common
cache is effectivly removing the bound-rects from this accounting,
so the garbage collection won't free them anymore.
Change-Id: Ie53226596ee2287e1059a74885f52c3d64bbccaa
Reviewed-on: https://gerrit.libreoffice.org/61377
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: Ia9bf718f1158c68971511688fe5af3e335d841ba
Reviewed-on: https://gerrit.libreoffice.org/61386
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
We still do our own cleanup of the LRU map, as we can't drop
any fonts in use.
Change-Id: I8ec5c6ce8f80893635621357e9085950e7010f5b
Reviewed-on: https://gerrit.libreoffice.org/61455
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
The current glyph cache on Windows didn't work for multiple fonts
and had to be manually invalidated on font change. The new glyph
rect cache is coupled to the font cache and will invalidate all
cached glyph rects, if a cached font is released.
So switch to the new cache on Windows.
Change-Id: Ic641f78e2e664dc0ad52e595d5fdfb6ef611eb3f
Reviewed-on: https://gerrit.libreoffice.org/61278
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
This way the font cache can correctly invalidate the cached glyph
rects when a font is dropped from the cache.
Change-Id: I050866099742334f01cac1b872228a017ddb5e9b
Reviewed-on: https://gerrit.libreoffice.org/61371
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
This gets rid of some statics and drops some duplicate code:
- the X11 based GlyphCache => gone
- the svp version of the GlyphCache => gone
- the "normal" GlyphCache
- the PrintFontManager
And while at it move the implementation into its own file
gendata.cxx.
Change-Id: I9063139c9482f5f37285505f389cf5f32c02426b
Reviewed-on: https://gerrit.libreoffice.org/61454
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: I736cc60af6d0c91623a25843bd9ff8fcdd9e8e90
Reviewed-on: https://gerrit.libreoffice.org/61453
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
The Windows-backend based GlyphCache is OpenGL specific, so
reflect that by renaming it.
Change-Id: I1034bfde14792f0b6a807f8e938742556a31fcfb
Reviewed-on: https://gerrit.libreoffice.org/61452
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
only called from one site
Change-Id: I2f8cb927df2f4f389c5d1bee19845aed5535663b
Reviewed-on: https://gerrit.libreoffice.org/61437
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I3928c6e9240ad28a65aeb4fecb07268f789faf67
|
|
Also add an assertion and some verbose debug output (in #if 0).
Change-Id: If4907a682ee7359d71078f72ca2f483c80fc4a58
|
|
Change-Id: Ia555369ae5dd530e5c0fdb4122071eff262059f1
Reviewed-on: https://gerrit.libreoffice.org/61418
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I9d0f21a10708edcc7f5da14f83944da58c33b1d9
Reviewed-on: https://gerrit.libreoffice.org/61408
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
when the menu is too tall to fit and the active entry is sufficiently
far down the menu that the problem gets triggered.
i.e. https://gitlab.gnome.org/GNOME/gtk/issues/94
as a workaround set wrap to 1, which is good enough to make
GtkComboBox take an alternative path to the one which sets
the problematic rect-anchor-dy value and the menu will at least
appear
Change-Id: Ia05887e0db727b7e1045179e36d348d22228a0ce
Reviewed-on: https://gerrit.libreoffice.org/61416
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
We want to store the alpha channel in BitmapColor. To achieve this
we can repurpose bIndex attribute for alpha. Generally we don't
need bIndex at all as we can infer if we use index colors or not
from the context (using palette or not)
Change-Id: I18fe748beeca59e2869368a1c3c2ee9d2062b41e
Reviewed-on: https://gerrit.libreoffice.org/61057
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Change-Id: Ia431650586ec26f5dc321cb162afa632ddb53ab3
Reviewed-on: https://gerrit.libreoffice.org/61361
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
These are all the cases where the warning doesn't suggest a direct replacement,
so the existing code is just wrapped in SAL_WNODEPRECATED_PUSH/POP for now. A
companion commit will cover cases where there is a direct replacement that can
probably be applied (even backwards-compatibly with our current baseline,
hopefully).
Change-Id: I899537cb36dd69e08b5442ad4af7ce69ce2ddaec
Reviewed-on: https://gerrit.libreoffice.org/61354
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
"static_cast from 'CGContextRef' (aka 'CGContext *') prvalue to 'CGContextRef'
(aka 'CGContext *') prvalue is redundant" with
--with-macosx-version-min-required=10.14 at least. (Not exactly sure why this
warning only starts to appear now. Presumably something changed in the system
headers. Lets see whether older baseline builds are happy with the dropped
casts, too.)
Change-Id: Id52d1891e5e33f7c37d095440339980b4000ca4e
Reviewed-on: https://gerrit.libreoffice.org/61353
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: Ia45e4ce13b92fd73fcc8ad52233aa5eccc3cfb6d
Reviewed-on: https://gerrit.libreoffice.org/61365
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: Id9e3127a7ac53dc9e50caa4d4c4c4d7c6e7008fb
|
|
Change-Id: I1a6f0a2b0de82fd0e8c30818c2ddf6e0a2d3f1b0
|
|
Change-Id: I85e310a4e2cf8e9f726d8d7c6ac8bbb399f0cd7c
|
|
Change-Id: I5804467c97758641718a6d6314cae015a0442d1b
Reviewed-on: https://gerrit.libreoffice.org/61233
Tested-by: Jenkins
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
|
|
Change-Id: I44053d445098efc619ad9255d8f21962531a0654
Reviewed-on: https://gerrit.libreoffice.org/60674
Tested-by: Jenkins
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
|
|
this adds just enough of basic UNO ifaces to have SID_PASTE enabled.
It can't actually paste anything yet.
Change-Id: I82e8ef37e11e9df21d27f08d4d199f3e5924ed84
Reviewed-on: https://gerrit.libreoffice.org/60494
Tested-by: Jenkins
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
|
|
So breeze-dark turns into breeze_dark
Change-Id: Id5d06f73a79f099048b31fd16f9f9611236af392
Reviewed-on: https://gerrit.libreoffice.org/61321
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I965d4b042cf1510d5286c85043792a140116497f
Reviewed-on: https://gerrit.libreoffice.org/61331
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I89693a01bd3e89f6d4000901c01accf88b453691
Reviewed-on: https://gerrit.libreoffice.org/61232
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|
|
instead of returning a mutable ref, which just reads weird
Change-Id: Ia2e2a7222f6ccd73e1d7f69dd7c2523f3a05ae24
Reviewed-on: https://gerrit.libreoffice.org/61309
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ie6818586e9dbbcbf79f0bb8e1031fc0a0b13763a
Reviewed-on: https://gerrit.libreoffice.org/61308
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
the magic smoke connecting the dialog to the viewshell was in
SfxModalDialog, so not inheriting from that drops the connection
add a SfxDialogController to enable setting up the connection
Change-Id: I29c45cd43e05295309fc647b039db0c6565ca0d1
Reviewed-on: https://gerrit.libreoffice.org/61304
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
It was unresolved external symbol CrashReporter::AddKeyValue()
in SalAbort()
Change-Id: I21c48e40a9b9b73e43f6b5f58079b7877a1deac6
Reviewed-on: https://gerrit.libreoffice.org/61293
Tested-by: Jenkins
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
|
|
found with
git grep -A2 -nP '!.*\.empty()' | grep -B1 -w clear
Change-Id: I87013eab9c6988048b891d20577e1f7efbce1b6c
Reviewed-on: https://gerrit.libreoffice.org/61295
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I9e583df920657ab401162db79decb192c4f69f90
Reviewed-on: https://gerrit.libreoffice.org/61283
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I9670407cd9e848d208f844f6d0879a5a19b86f4c
|
|
so 1-10 in format numbering etc isn't slightly clipped
with a scrollbar on top of it
Change-Id: I55909993d4b4c846ca0aeff2f2cd458dfa59d860
Reviewed-on: https://gerrit.libreoffice.org/61262
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I8ea74dc15bba04c0615e61c069407160a7b95c4c
Reviewed-on: https://gerrit.libreoffice.org/61242
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
I thought just border windows can act as proxies, but this bug
proves me wrong; also floating windows set mpClientWindow and may
have an additional nested proxy border window.
Interestingly proxies can have multiple children, which I tested
with a temporary assertion I added!?
So just ignore all proxy windows until we find a real one, And
drop the assertion as it doesn't make sense any more.
I'm just wondering why I always read Child instead of Client...
Change-Id: I83a1a5743b434399aae361cb14058993ba60c83c
Reviewed-on: https://gerrit.libreoffice.org/61196
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>
|
|
Change-Id: I3e72909ca1aa9f97721d982f16624f496152ae2d
|
|
To get that with CoreGraphics on iOS we need to use also
kCGImageByteOrder32Little in the CGBitmapContextCreate() call,
otherwise the bytes will be in ARGB order in memory.
(Not touching the macOS code here.)
Change-Id: I3c2dd94feb1c6bf46c5b335f5901b29e5fe1e7fb
|
|
Change-Id: Ie2ae18b1722e91996887162b4a8ff02d2d812341
Reviewed-on: https://gerrit.libreoffice.org/61228
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I373886d3613607b8ee8e96d8c0314865f3991a92
Reviewed-on: https://gerrit.libreoffice.org/61241
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I1859d612e13b2f0d837aa65012552c441346cb1a
Reviewed-on: https://gerrit.libreoffice.org/61236
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
|