Age | Commit message (Collapse) | Author |
|
Change-Id: I82405059d900b6605075bf5756f3f0fb99e9002e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129451
Tested-by: Julien Nabet <serval2412@yahoo.fr>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
so instead forget about forcing under/over if it doesn't fit on either
side and just position so its visible
Change-Id: I529bf0d2bb1a3108a8702792ef067372441c2002
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129662
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Entry point to call the ImplLOKHandleMouseEvent function
mainly to process mouse events.
Change-Id: I17513643733bf5990d41ab8cf47cdc322ed526ea
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/118846
Tested-by: Szymon Kłos <szymon.klos@collabora.com>
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/117711
Tested-by: Jenkins
Reviewed-by: Henry Castro <hcastro@collabora.com>
|
|
> vcl/source/app/salplug.cxx(440,6): error: externally available entity 'HasAtHook' is not previously declared in an included file (if it is only used in this translation unit, make it static; otherwise, provide a declaration of it in an included file) [loplugin:external]
> bool HasAtHook()
> ~~~~~^~~~~~~~~~~
Change-Id: I8e293fa53ad0324d2cac4195145dad536e699c47
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129653
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I7ad212dfff8b34d05e3b45107a1ef033a4efc454
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129651
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
In tiled rendering case, each user has a View/Controller
object to interact with a unique document (Model) for
collaborating purposes. However, in the desktop case a
unique user has many View/Controller objects to handle a
unique document, so the user has only one global active focus,
capture and tracking window object.
In order to handle independent drag & drop for each user,
it is created ImplLOKHandleMouseEvent.
Change-Id: I735fae9b9858a75f9fedb603798220ab302d65f6
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/118843
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
Tested-by: Szymon Kłos <szymon.klos@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/117708
Tested-by: Jenkins
Reviewed-by: Henry Castro <hcastro@collabora.com>
|
|
...after ea771e85b2302829394df545bb82c02bff2750c2 "Resolves: tdf#146997 use
sal_Int64 instead of sal_Int32 for spinbutton values" changed the default value
of pcr::ONumericControl::setMaxValue
(extensions/source/propctrlr/standardcontrol.cxx) from int to sal_Int64 max.
For example, UITest_calc_tests9 failed with
> vcl/source/app/salvtables.cxx:5507:12: runtime error: 9.22337e+18 is outside the range of representable values of type 'long'
> #0 in SalInstanceSpinButton::fromField(double) const at vcl/source/app/salvtables.cxx:5507:12
> #1 in SalInstanceSpinButton::get_range(long&, long&) const at vcl/source/app/salvtables.cxx:5541:11
> #2 in weld::MetricSpinButton::update_width_chars() at vcl/source/window/builder.cxx:258:24
> #3 in weld::MetricSpinButton::set_range(long, long, FieldUnit) at include/vcl/weld.hxx:1991:9
> #4 in weld::MetricSpinButton::set_max(long, FieldUnit) at include/vcl/weld.hxx:2012:9
> #5 in pcr::ONumericControl::setMaxValue(com::sun::star::beans::Optional<double> const&) at extensions/source/propctrlr/standardcontrol.cxx:422:38
because std::round(fValue * Power10(get_digits())) = 9223372036854775807 (i.e.,
2^63 - 1) cannot be represented as IEEE 754 double exactly and is represented as
9223372036854775808, which was outside the range of sal_Int64, so treat that max
limit value specially. (The corresponding min limit value -9223372036854775808
doesn't have that issue.)
Change-Id: I9ff7a281278a0c66b0b0d49deb0846bcdb91a8fe
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129639
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Follow-up to commit 50682cea4196819980c8e2d4018f80384097ce6f (PDF
export: improve precision of pdf image sizes, 2022-02-04), sounds like a
good idea to do this in general, not only in that specific case that was
required for the test to pass.
Change-Id: I5b36a5a3be26132e96a4de9d224abcaaf79beadb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129599
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
The position of the dropdwon is absolute to screen so
add this type so we can transform correct position on the client side.
Change-Id: I9e2faeb532e313a1a4043d4044f71ada2e7225a2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/127439
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129459
Tested-by: Jenkins
Reviewed-by: Henry Castro <hcastro@collabora.com>
|
|
Check the URIs that are exported.
Change-Id: I2d8b00f767a61c5abdb9163701fe497bb258c790
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125203
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: Iec23fba24464c6d29ecb63d3d2018ded8d9937f1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/128500
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129495
Tested-by: Jenkins
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
|
|
Change-Id: I45b90c40f062afe950952740901c40fee085ed06
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129552
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>
|
|
Change-Id: I03c1c926441f80c3741bb7d9231363603d0dbffe
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129579
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
so the _ is shown to the user, not used to add an underline
Change-Id: I2de707a2d1da8d842ef03497f9b8f69d37368f12
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129531
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
which for me with french IM enabled on my keyboard is physically
AltGr+[ (crashes here on trying to underline ¨ with a length of 2)
then e.
Change-Id: Id59bd84af2ab4d8403b11b1a13efb2bbdb608199
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129538
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
which is true for the one bundled with LibreOffice but is not
true for the fedora one (kacst-book-fonts-2.0-26.fc35), but
neither of them have Cyrillic glyphs
Change-Id: Ide8b84502886c7b9e0bc4fcef4685b4b89ec2bc3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129494
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
fedora KacstBook ver is less than bundled KacstBook ver so bundled is
preferred, bundled KacstBook has no Latin glyphs, but glyph fallback
finds system one and sees it has the missing glyphs, determines that the
whole thing can be done in the fallback font, so the toplevel is unused
and so "invalid" and this test fails on checking that.
The test is supposed to ensure that glyph caching gives the same result
as uncached, which it does, so just assert if there is a difference
between cached and uncached
Change-Id: I3c424a17f3e544aa64f4af98e0c151fc00b2e6e5
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129499
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Not needed in the end.
This reverts commit 688051edf59a311371b84ea34db4991d3a164d52.
Change-Id: If2474ccff0ada5fd0ca5206f10ca2893df0fe1f8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129491
Tested-by: Jenkins
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
|
|
the uses of new are wholly unnecessary
Change-Id: Id7238be6e73567add8a0d3ed6dc9f3262e15c043
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129493
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
This helps exporting full-page PDF images from Draw more accurately. In
case the page size was A4, then the page height is 841.8897637795276
i.e. 842 points. Full-page PDF images are scaled to this size, so the
referred PDF form XObject has the reciprocal of that scaling. We used
to just write 0.00118, doubling the precision leads to
0.0011878840.
In practice the old precision resulted in e.g. hyperlink rectangles to
get out of sync with link text after 2 roundtrips, while the new
precision doesn't result in any noticeable modification to the link text
position after a roundtrip.
Change-Id: I72cc68696b9b9bcc1cbfde8df331c2b9c5f9eb29
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129482
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
When we open dialog with tab pages, help should open
website for corresponding tab.
Use focused window - but not global which can be modified
by other view in lok case.
followup for:
commit 70f2fc7a10f41204e353bd3487377983a1018b53
lok: don't use focus window to request help
Change-Id: Ib7544a91696ce2f2c8ab01e62a47d279d2c5c655
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/128863
Reviewed-by: Mert Tumer <mert.tumer@collabora.com>
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129167
Tested-by: Jenkins
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
|
|
Change-Id: I70f34ac5e9b5d2f2d6c0375e823908eaa2e540b2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129487
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I292d36ad9bb900881662898345f8e1f00ebf28b0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129484
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
for these cases where draw wants to massively scale the units
the underlying "metric conversion" are already using sal_Int64 anyway
Change-Id: I94e120d72644319548f75b2f68cfe60d4829a2e8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129356
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Which is useful to see, as a 2 page Draw document has 2 normal pages and
a master page (3 sizes), but in practice there can be only a single
size, so it's useful to see what size is coming from where when they
don't match.
Change-Id: I505653029ae67ea0a57c8f8bb61cf475d77aaccb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129425
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
Tested-by: Jenkins
|
|
Change-Id: Ifd25bf75c5e173a7db4778713f1d371e9ee24ac1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129381
Tested-by: Jenkins
Reviewed-by: Xisco Fauli <xiscofauli@libreoffice.org>
|
|
When inserting a PDF file as an image or pasting PDF data from the
clipboard on a Retina iMac the resulting rendered image did not look
sharp.
Using a surprisingly large extra scaling factor helps. The exact
reasons for this are unknown. It isn't enough to use a scaling factor
of just 2 (which is the HiDI ("Retina") scale factor on my iMac).
Possibly the fuzziness is related to what Pdfium uses to render text.
Also, look at CountDPIScaleFactor() in vcl/source/window/window.cxx.
The GetDPIScaleFactor() function lies on macOS even more than it does
on other platforms. It claims that the DPI scale percentage is always
100. But in fact most Macs nowadays have a Retina display so it would
make more sense, in theory, to at least always return 200 instead.
That wouldn't be any more wrong. But that causes regressions in the UI
rendering, like needlessly large icons in the toolbars.
Change-Id: Idc694f742c4ac32a5a134f8d206cf4eee467c39a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129369
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
Change-Id: Ic551b34069496c683a065997f051037ecabe3b7f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129398
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I2202ca853edea29e987e920b1f8bf65ca622291e
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129397
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
The backend test unittest already needs some special handling (such
as not smooth-scaling in Skia backend when in HiDPI mode), but
graphics test were failing because the name of the unit test was
used for detecting whether to do special handling.
Change-Id: Ieb23915f1eeafc13c921de06475cb761dc4c485f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129292
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: Ica74dd5e55d30605ee03affa4b724ffa4ec65b5a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129357
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
under gtk. When the formatter sets a value during focus out then gtk
won't emit a value-changed because it wasn't changed by gtk, move that
emission to the formatter itself
Change-Id: I6e2681b77d6effcd3abaa028e3e5383aa218863a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129306
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I657cc2b99e84c0846891da0402b84a4ae668553f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129389
Tested-by: Julien Nabet <serval2412@yahoo.fr>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
It passed "make check"
Change-Id: I02c46c42c3e705849562d465a086801425e32828
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129312
Tested-by: Julien Nabet <serval2412@yahoo.fr>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
(*) use more cache-dense temporary arrays
(*) use float instead of double - I cannot see a difference in results
(*) tools::Long->sal_Int32, we don't need more than 32-bits for x/y
coordinates
Change-Id: I81a1da573735f183a7d2a117f6170cf373433e4d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129324
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ifac9aedb4972323da8b70d0a8f6f663513d2fbdb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129310
Tested-by: Julien Nabet <serval2412@yahoo.fr>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
In 534fc16a9f8cd01bb88485e2cf6b9cd33aed3745 I got this backwards,
if the value is not supposed to be 0, then it should be at least
the small value, not at most.
Change-Id: Ic55d2b2402985d26e77be0d1605b9ba5985b5ac7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129277
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
The conversion from Metafile to Bitmap in GraphicExporter
is based on metafiles which does just not deliver enough
precision when getting the bounds to do the correct size
calculation for the target Bitmap.
So I changed that to use Primitive functionality what
produces better data. That old fix/correction itself was
based on hairlines only which is anyways no longer
sufficient since LO uses less hairlines nowadays, what
is good in general (They are a relic of non-AA times
when it was not possible to paint/work with lines taller
than one pixel).
Thus, cases need to be solved more generic based on
better data. It would also be good to change this
to primitives completely, but that is too much for now.
Change-Id: I71bd136b106ef8ff3ba51458c46df08269773c01
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129235
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
|
|
Change-Id: I0f4f6d4dec543780a2421193785d099c0b998f95
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129307
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: Ie948064ea09b6fe415b28ec4165883bc78136619
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129279
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Changing the orientation value to anything else than 'Automatic'
didn't do anything by default. This was because by default
neither isPaperSizeFromUser() nor getPapersizeFromSetup()
were set, so PrintDialog::setPaperOrientation() did nothing.
It looks to me like 8cbdc6a068ad88fc43a98bd0f88 that introduced it
was rather broken (not just this bug, but also e.g. the ugly
modifying of the paper sizes by non-const reference from a const
function). In fact this whole stuff still looks broken to me, why
does it change paper size instead of just setting the orientation?
It seems like the orientation gets reset, or maybe the setting
was just a band-aid. I don't know how to fix that all though.
Change-Id: If5fdf4c47e06f2b0797d27126d21b3451b8334cf
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129239
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
- enable subcontrol .ui for jsdialogs in annimation panel
- optimize to not recreate widget on every sidebar refresh
- don't sent close message for whole sidebar when called from
subcontrol
This fixes widgets in annimation panel like: Direction listbox,
or other replacements for selecting color, font etc. for font effect
Change-Id: I5683ca9cefe384ed0d2a34d46936ddf4a9b45bce
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125757
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Mert Tumer <mert.tumer@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129170
Tested-by: Jenkins
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
|
|
- remember WindowId for statically created message dialogs
- remember _DIALOG_ handle for message dialogs so response
action can be done
Change-Id: I1f6c9877e20dcbed4855b32d1e89d9ce3ff12111
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125686
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129169
Tested-by: Szymon Kłos <szymon.klos@collabora.com>
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
|
|
- enable validity listbox in online
- flush when sending close popup to be sure it is sent
Signed-off-by: Szymon Kłos <szymon.klos@collabora.com>
Change-Id: Icf33e26c48b78ab60030a7395e8c7ee5462fbf7c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/125647
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Jan Holesovsky <kendy@collabora.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129168
|
|
Change-Id: Ic116d0207df6d37fed850357ad41d7778bf08df9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129233
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
This commit implements a WebP reader and writer for both lossless
and lossy WebP, export dialog options for selecting lossless/lossy
and quality for lossy, and various internal support for the format.
Since writing WebP to e.g. ODT documents would make those images
unreadable by previous versions with no WebP support, support
for that is explicitly disabled in GraphicFilter, to be enabled
somewhen later.
Change-Id: I9b10f6da6faa78a0bb74415a92e9f163c14685f7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/128920
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Change-Id: Ia73317a79de66778d4b2d0face07ee4c41e328d7
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129157
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I66b6c0157759caf20df9af1642e0f5488d47dc22
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129151
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I7c99ead4a7a3107039d5ec096ad4a96463a8a432
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129137
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
not the parent allocation
Change-Id: I1801fca14e864e4e18c0b42d7f3c3956e71f8e23
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/129115
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|