Age | Commit message (Collapse) | Author |
|
When traversing the components of composite glyphs, the code was
treating a component pointing to an empty glyph (e.g. space) as an
error, but it is not. An overzealous bounds check.
Regression from:
commit 41007842ed9bb5d6165792a197769f72dae55a2c
Author: Martin Hosken <martin_hosken@sil.org>
Date: Thu Sep 10 10:14:18 2015 +0700
Refactor graphite integration and update graphite
Change-Id: Ie352f17cb08cd5762ff2a339662248a536b4e934
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140897
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
There was a typo in the check, and we also want to fallback to using
blanks if no Kashidas were found, so check for Kashida positions first.
Change-Id: I64cc3bf5c76ecc01764073bb5a2302520157fad8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140896
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
Change-Id: Ib2152968afb92757fa705b03d3e8dd8996440bff
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140851
Tested-by: Jenkins
Reviewed-by: Jim Raykowski <raykowj@gmail.com>
|
|
Change-Id: I257e999ced55d44461e24e30e961648b09ea1087
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140837
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>
|
|
Change-Id: I475a64ac3977bad716ce6f94a88ff02af37a932c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140847
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
part of the process of hiding the internals of SwPosition
Change-Id: I7a1900352a9404b64987d71c9b9f44d07e2a176b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140828
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
part of the process of hiding the internals of SwPosition.
Also, remove code that checks if pCnd is nullptr, because we
have already dereferenced it at that point.
Change-Id: I5362c2540af88bdd35db95472ceed420b90cf1bf
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140827
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
part of the processof hiding the internals of SwPosition
Change-Id: I5789329e9d0a6cbdc03d51b2163fc88ebd9c6293
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140829
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
part of the process of hiding the internals of SwPosition
Change-Id: I4c5cb657fa193ae2efbf0d7e315939e68ca89600
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140830
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
seems a regression from:
commit d7234feac18dbb94fef310e8c241428426b27717 (HEAD, refs/bisect/bad)
Date: Thu Aug 25 18:45:33 2022 +0200
use more SwPosition::Assign
Change-Id: I80602b2cc01b28986b04a97fce5c51673e6fe8a3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140848
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
The chain of fixes for #i119960# tdf#48621 tdf#125440 produced
code that is suboptimal and not robust enough against some further
corner cases, taking quoted field content where there shouldn't
be.
First, in ReadCsvLine() assume that if a generator is broken
enough to start a field quoted followed by containing an unescaped
embedded quote and there is no closing quote (i.e. immediately
before a field delimiter) until the line end then the generator
will not be clever enough to write embedded linefeeds either and
the field starting quote wasn't one but to be taken literally as
all other quotes until the now unquoted field end. In this case do
not read a subsequent source line for the current row.
Then, for individual fields of a row make a similar assumption, a
quote-started field has to end with a quote before a field
separator (or line end) or otherwise all quotes of that field are
literal data up to the next field separator.
This made it necessary to adapt two test cases of the garbage CSV
import test to produce different garbage than before.
Change-Id: I4424b65c87c7f9dcbe717a7e6cf207352cb613f3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140850
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
|
|
Open the edit index dialog when double-clicking an alphabetical index field.
Change-Id: Iaba398e485d2e7e848503189ad1f3b12975a5a71
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140836
Tested-by: Jenkins
Reviewed-by: Andreas Heinisch <andreas.heinisch@yahoo.de>
|
|
Change-Id: Ic43a4e52b5ea40e54b368815e504840cb2c78aab
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140845
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
This patch removes the names of template categories that are not used. None of these template category names are reachable via the Templates Manager nor via the Save as Template function.
Change-Id: Ic2caf485106134de8554b93b5d6e14bdf7f382e6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140610
Reviewed-by: Laurent Balland <laurent.balland@mailo.fr>
Tested-by: Jenkins
|
|
Change-Id: Ie6b2a0cebab375e8b66d5a9658dbb0cc03b8c704
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140844
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
It is not enabled by default, so I don’t see the point of silently
over-riding configure option.
Change-Id: Ib2c14b2ce7519775ed5a3365c34ade3c96b66532
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140843
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
Change-Id: I6bec0af1c7fe109b81b5392c06f795eed03df061
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140716
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
Reviewed-by: Eike Rathke <erack@redhat.com>
|
|
The “system” font list is used for both actual system fonts and fonts we
add with AddTempDevFont(), so the caching makes loading embedded fonts
broken the first time the file is opened, requiring a next open without
closing the application.
The callers, I think, are careful not to call GetDevFontList() unless
they really want to refresh the font list, so the caching is probably
superfluous if not wrong.
Change-Id: Ie14cc473a64425fd5270bc65ffd61dea2d873cfe
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140840
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
This reverts commit ce54ba96f38b4af3aab1a7064078ee406eb021c6 and the
followup commit 88c511981e31c73dced95b5dc3c200fdf2a4e932.
Both effectively disabled enabling libeot support.
See also: https://lists.freedesktop.org/archives/libreoffice/2022-August/089205.html
Change-Id: I54780c69ca83b230b0c9b3b18065782fb5626da6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140838
Tested-by: René Engelhard <rene@debian.org>
Reviewed-by: René Engelhard <rene@debian.org>
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
Change-Id: Iefe265584d710bd6cfc9e7ed0a656dd83d91c84c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140811
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
With the previous commit, when we want the full font data we pass 0 for
table tag. This does not work with CoreText’s CTFontCopyTable(), we we
emulate it by using hb_face_builder to construct a full font from
individual font tables.
Change-Id: I0edb10982434872221466e9ec9ef9cd39087967a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140831
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
Change-Id: I6f7c4508f7cef022eaf65a998cb242078d3771c6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140826
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
It was used for subsetting, but we switched subsetting to a more generic
code path a while ago.
Change-Id: I6066b7697268e9444587d73810f9465488aa4740
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140825
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
We don’t need full access to PhysicalFontFace. This will be needed when
we use hb-subset API for instantiating variable fonts.
Change-Id: I2578525cd54167d01cd2748d5ac1900c607867a1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140823
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
In the content of Equals/notEquals, Eike said,
"This should not use collation (which should only be used
in sorting context) but ignore case transliteration instead."
"I'm undecided about replacing this (and the Between|NotBetween)
collator with transliteration. Locale sensitive collators have a
different comparison than transliteration
and order characters differently."
I also added a few code clean-ups.
Change-Id: I94f73d99b36f8995c4e80fd2f317f1b5bca4f165
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140676
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
Reviewed-by: Eike Rathke <erack@redhat.com>
|
|
Change-Id: Ice3068f3b20c0f8a20e8b3c05ddf358b971f66ee
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140810
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: Ib547189e0f87f3e3a3096dd587127706474983cc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140809
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I97292df52c50e73b3f0d4ee9b5f8134959f3b7e5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/137653
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ifd4d67fa8211cf703c5d5d6ea99a0f41bd427e1c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140612
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
I've decided this is not an API change because there exists
offapi/com/sun/star/script/vba/VBATextEventProcessor.idl
This patch allows New, Open, and Close events to be
sent to VBA macro event handlers.
I tested with DOCUMENT_OPEN.
Amazing - all this stuff is here in place,
but isn't properly wired up. How can so much
ground-work be done without it ever being functional?
I don't think this has ever worked,
and the same problem/code is seen in OpenOffice.
The service names were already mismatched in the initial patch:
commit fc135411fa926f4d89e98378d113f597af79e2a2
Author: Daniel Rentz on Tue Jul 6 19:34:53 2010 +0200
mib17: #i112634# add VBA sheet event handling,
based on a patch from Noel Power
https://bz.apache.org/ooo/show_bug.cgi?id=112634
says "Additionally I added the hooks for vba for word
( but they are compiled out for now )"
Enabling this will open up the doors to a lot of
untested code I think.
I had to move initialization of mxVbaEvents
because it was being set without a VBA project,
but it didn't work to put it in VbaGlobals like Calc does.
Change-Id: I08cc0e33444dc7dfaac82cf6cebb1a89dbb69faa
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140801
Tested-by: Jenkins
Reviewed-by: Justin Luth <jluth@mail.com>
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Change-Id: I063fda1f578ced27111d34c98e8bf25f412800c1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140805
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>
|
|
No need to make it module public. Easier to modify it.
Change-Id: I119381d463c8f176b355fc7df111011ae518b49d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140821
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
Change-Id: I5f79fddd4e514c8224446e2327528002e9daa566
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140808
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
since the installer fails because it cannot find it.
(was removed in 5f1eb2a12b6e6ee646b5e9536822531ad0bd3259)
Change-Id: Id7d8cccc43b461bc10a447d8d5d48009bc48f6d3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140807
Reviewed-by: خالد حسني <khaled@aliftype.com>
Tested-by: Jenkins
|
|
It makes no since to have overline of different sizes than underlines,
and in fonts with big ascenders (like Amiri), we end up calculating much
thicker overline than underline.
Change-Id: I97e9818c94e815b2eb57813ab22329bc833fd34c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140804
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
This respects the values set by font designer, which are usually better
than the values we calculate especially for fonts with deep descenders
(like Amiri).
Old code is kept as fallback in case the font does not provide such
values.
Change-Id: I51a5147e4c6e006d1dcd13817ed21f0f62b47e97
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140803
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
Use HarfBuzz API instead of reading the raw font table directly, since
HarfBuzz will apply the font variations as needed. For non-variable
fonts we still also use HarfBuzz font metrics API, but in a more
fine-grained way to maintain backward-compatibility.
Change-Id: If6b12a11ecb63356be92ef4f0714355ae77378f2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140799
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
Asturian (ast), Paraguayan Guaraní (gug), Sidamo (sid)
and Upper-Sorbian (hsb) localizations did't work despite
their existing localization stored in the language fallback
dictionary of LibreLogo.py, because their 3-letter language
identifiers with their country codes trimmed to 2-letter,
for example: "hsb-DE" -> "hs" instead of the correct "hsb".
Missing Upper-Sorbian support was reported by Michael Wolf.
Change-Id: I57aa6752d05452d1f8bb95ff91f4d4456dffcd80
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140796
Tested-by: László Németh <nemeth@numbertext.org>
Reviewed-by: László Németh <nemeth@numbertext.org>
|
|
Change-Id: I5d4795fcaf20ae89c18ddb1d75ca93f078e0e3fb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140798
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
seen in format, cells in calc
try an explicit depth
Change-Id: I395563d2b412b2ca8839292296f9dad6db704620
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140797
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I5a7b16fb5de83118bb2d0b3bbfd30fce1b9c8d82
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140795
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I55a469f30f837b7c55b6c7d9672c7e909abe6a77
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140791
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I0e6b4bf7a6420e555ef72b58995055a116afd43d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140790
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
part of the process of hiding the internals of SwPosition
Change-Id: I3dde04bf78e1412f731b3cd63c658adfb1b2fea5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140793
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
as part of the process of hiding the internals of SwPosition
Change-Id: Idc7bc4cd22f13cf7bfb7982d445d13970a9e70a8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140792
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I6552f3daee7b18dbd311ac7e80ebf1e84e36704d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140787
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I60d777b27ffee095807490f4b181b1b08ed88462
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140683
Tested-by: Jenkins
Reviewed-by: Maxim Monastirsky <momonasmon@gmail.com>
|
|
We can take it from the font instance.
Change-Id: I54c507d498c58e271c479b0a615b165345a1c33b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140788
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
noticed in gtk darkmode on switching from initial application
to start center
Change-Id: I1cffab97ecd69ca6043531a6b2b5fc34b1ca84f1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140789
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I3569d8dd41bf1c2db6a1fd379ca5596342922e45
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/140786
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|