Age | Commit message (Collapse) | Author |
|
Specification: ISO 14289-1:2014, Clause: 7.18.1, Test number: 3
A form field whose hidden flag is not set and whose rectangle is not outside the crop-box shall have a TU key present or all its Widget annotations shall have alternative descriptions (in the form of an Alt entry in the enclosing structure elements)
Form controls are weird because they have an SdrObject with the usual
name/title/description plus a property "HelpText" on the control itself
which is already exported as "/TU" on the /Annot unless it's empty.
Exporting the SdrObject properties via ObjectInfoPrimitive2D doesn't
work as tragically that is only created for form controls when painting
to the screen while PDF export takes a detour that needs special
handling.
Change-Id: Id96f7dd13f190ab439c099cd1f4acb70c1c9fdc9
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149554
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
|
|
The auto-fit algorithm has been tweaked to be more in-line with
the expectations of OOXML. This means a couple of changes to what
properties are scaled by the algorithm have been made:
- most properties that influence the X axis position or size (for
example indent) are not scaled down or changed by scaling.
- properties that influence y axis position and size are scaled
by a separate parameter (like in the OOXML). This is used in the
auto-fit algorithm in a different way.
- if line spacing is proportional, it is now scaled with the
spacing parameter. Fixed line spacing doesn't get scaled.
- the main scaling X,Y parameter only scales the fonts.
- trying hard to scale the fonts to the nearest pt (point) value
With this change the scaling is much more stable than it was
before - for example it doesn't matter what the unscaled font
size is, when it is scaled down to the text box size, it (should)
always look the same (for example scaling from 32pt -> 10pt or
64pt -> 10pt or even 999pt -> 10pt).
The algorithm is also rewritten to be better at finding a fit and
is also better at find a good fit, but it can take more iterations
by doing so (there are ways to improve it however). Previous
algorithm used a linear search to converge to the best fit in less
iterations, but the issue with that was that it could in some cases
miss a solution (especially since change to floating point scaling
parameter). The new algorithm now uses a binary search - always
trying the middle of the search space.
OOXML export and import was also changed to take advantage of the
font scaling and spacing scaling parameters. The additional
scaling at export that was needed to have consistent OOXML support
was removed.
Change-Id: I8f3bb8d43a01931f18bd7ffdf8e0ba40caa73d8b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149207
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
i.e. new Draw document, change the char formatting in the
default drawing style (e.g. font size), insert a table and
fill its cells with text. Save and reload. After reload the
first cell inherits the formatting of the default DS,
instead of its assigned cell style. But dragging the table
around shows the correct formatting in the dragging overlay,
and it's also correct when copy & pasting the table.
This is caused by the order of things at import, as a table
is created initially with a single cell at least, and it
gets its cell style assignment too early. There is also
similar code to what is changed here in SdrTextObj::
BegTextEdit, and it's rightfully overridden by SdrTableObj.
This appears to be a regression from 4-0, but I'm not sure
which commit and what exactly changed there.
Change-Id: Id096c7b5f330d6abd8f5b01c96448a9d8aac8001
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149616
Tested-by: Jenkins
Reviewed-by: Maxim Monastirsky <momonasmon@gmail.com>
|
|
Change-Id: I5d0fc2a4b078e49be684b4a9b8a27f6059d9e5c3
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149475
Tested-by: Jenkins
Reviewed-by: Andreas Heinisch <andreas.heinisch@yahoo.de>
|
|
in favor of sidebar deck
Change-Id: I2daf34edbe71341ebc44ca9b966c97f8e494d583
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149433
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <samuel.mehrbrodt@allotropia.de>
|
|
rather than default to nothing selected, if we switch pages and then
back to page 1 this will be selected anyway, so start off that way
Change-Id: Ia684a5e44b12c984df3920f18b34c99693a2702c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149426
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: Id5cad5f2f4ac21db8675de16f1bc0ceeff39fe6c
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149424
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
* Set minimum size
* Wrap long texts
Change-Id: I4c23ebd47d26fe628187143eeda43efe467b911a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149429
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <samuel.mehrbrodt@allotropia.de>
|
|
because on kf5 they are getting shown, which suggests a wider more
generic bug on that variant, but use remove for this specific case
because that's ok for this usage and works around the problem for now.
Change-Id: Ib2bc316cc33894253a0b6335ae8643d0c8d78794
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149325
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Also simplified using the test cases, these now depend on
an ENV VAR called MCGR_TEST. Fallback is no test. For
seeing a multi-color gradient use 1, for Case16 use 16.
If active, all gradients are replaced with the one active
for the test, 2D and 3D. This is temporary but also for
pro build to check for speed there.
Change-Id: I90f3c7e59d9d0a3e070a849af3f9ea1c9e5462a0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149316
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
|
|
veraPDF complains:
Specification: ISO 14289-1:2014, Clause: 7.18.1, Test number: 1
An annotation, excluding annotations of subtype Widget, Popup or
Link, shall be nested within an Annot tag
This is very similar to Link annotations, that is to say, extremely
complicated to get it thought the convoluted PDF export code, with
additional complication that the StructElem is produced by drawinglayer
and the page annotation by sw.
Put another map into PDFExtOutDevData where sw code puts stuff for the
SdrObject that drawinglayer can find.
The test had the problem that PDFObjectParser::parse() could not handle:
<</Nums[
0 [ 6 0 R ]
1 6 0 R
]>>
Fix dropping the "1".
Change-Id: If5bf7c552e26ebb7e631030b8aaecd4281e77acc
(cherry picked from commit c78e90bd28cc4d6d3bde473535107784b12d9c0d)
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149008
Tested-by: Jenkins
Reviewed-by: Michael Stahl <michael.stahl@allotropia.de>
|
|
Project URLs are extracted to the Common.xcu files
and read in the source files.
Change-Id: I0da72ff033af6c155195b2f91177e7bf7d554daa
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149115
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Jenkins
|
|
The biggest change here is to allow multiple ColorStops
with the same Offset. That allows to define gradients
with 'hard' color changes at any place using two
ColorStops with the same Offset and different Colors.
This required quite some adaptions, but works well.
Also removed in this context checking for all Colors
being the same to not mix up things. Also works well.
Also changed the need for having Start/EndColors AKA
ColorStops for 0.0 and 1.0 in place, instead 'imply'
the 1st ColorStop to also define the StartColor and
the last one the EndColor. This allows e.g. Gradient
definitions with two GradientStops at the same
Offset e.g. 0.5 with different colors to already
define a full Gradient.
Also added a tooling method to reverse ColorSteps,
which changes the order and mirrors the Offsets
(what even keeps an existing sort valid).
This is useful e.g. for GradientAxial which is the
only one where for decomposition the Gradient had
to be interpreted 'reverse' since it's defined
from center to edge, but for creating correct filled
polygons to represent this the inverse order had to
be used, creating polygons from edge to center.
This led to 'wild' code for this one of six cases
and prevented unifications with the other cases
(also made your brain flip).
Thus I adapted this now to use the reversed
ColorSteps consequently, and the same principle
loops than the other implementations to make things
easier for the future and to use common tooling.
Change-Id: If2943348d17d5b9cd165f4d78f22638a1dff5237
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149208
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
|
|
Commit e97e62a4707e09739f9c00f8fc4a943f0123a9cf had wrongly changed
'Rectangular' to 'Square' (API RECT, ODF rectangle) and 'Square' to
'Quadratic' (API SQUARE, ODF square).
This patch reverts 'Square' to 'Rectangular' and 'Quadratic' to
'Square (Quadratic)'. It uses 'Square (Quadratic)' instead of simple
'Square' to help users to find the correct type.
In addition, use of the wrong type names in comments is fixed.
Change-Id: Ibbd90be2cceb65c9857e473cb74c39df1b9530f1
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149004
Tested-by: Jenkins
Reviewed-by: Regina Henschel <rb.henschel@t-online.de>
|
|
Change-Id: I6d9d6dca0118567f68db11b7c0cce60643211ab2
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/149000
Tested-by: Jenkins
Reviewed-by: Samuel Mehrbrodt <samuel.mehrbrodt@allotropia.de>
|
|
Added tooling replaceStart/EndColor to allow simple
transition for code that does not immediately adapt
to multi color gradients. Also added
createColorStepsFromStartEndColor for the same
purpose.
Adapted XGradient to no longer have Start/EndColor
at all, but only use ColorSteps.
Adapted all usages of XGradient to no longer use
Get/Set/Start/EndColor, but access the ColorSteps
instead.
Replaced quite some XGradient constructors that
used XGradient() as Start/EndColor since this is
already the default.
Adapted ColorBlending to black AKA Start/EndIntens
in XGradient to work now on all ColorSteps in the
required linearly-scaled manner.
UNO API changes:
Added com::sun::star::awt::ColorStep as basic data
element that holds a pair of Offset and Color.
Added com::sun::star::awt::ColorStepSequence to
handle an array of sorted entries.
Added com::sun::star::awt::Gradient2 derived from
com::sun::star::awt::Gradient, extended by the
needed com::sun::star::awt::ColorStepSequence.
Added MID_GRADIENT_COLORSTEPSEQUENCE to UNO API
to provide access to ColorSteps directly.
Adapted XFillGradientItem::QueryValue/PutValue to
make use of new UNO API data structures. To do so,
added tooling methods for data transition:
- fillColorStepSequenceFromColorSteps
- fillGradient2FromXGradient
- fillColorStepsFromAny
- fillXGradientFromAny
and adapted
- case '0' (all data)
- MID_FILLGRADIENT
- MID_GRADIENT_COLORSTEPSEQUENCE
- MID_GRADIENT_START/ENDCOLOR
to make use of these.
Tested usage of these in the office.
Renamed from GradientStep to GradientStop after
discussions with members on the list to make this
closer related to other norms/definitions.
Also renamed classes and class members to better
reflect to GradientStop, so grepping/finding will
be easier (e.g. 'Color' just exists pretty often,
but 'StopColor' is more precise).
Changed the used UNO API class for reprsenting the
Color to better reflect to ranges [0.0 .. 1.0] and
usage of RGB.
Change-Id: I1eeb3e97e81d6785967615d1ff256551fc3b882d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148849
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
|
|
Refactory
Cause
Currency must be added in the MRU list in SvxCurrencyToolBoxControl::execute
method.
Unfortunately in this point the only information available is the number format.
It is not possibile to infer currency from number format (e.g: both USD and
AUD use symbol $).
Encapsulating currency data
In order to solve the problem currency data is enacpsulated in a SvxCurrencyData
struct. Added a vector of SvxCurrencyData to SvxCurrencyToolBoxControl.
Static and non static call
Former static SvxCurrencyToolBoxControl::GetCurrencySymbols has been
preserved, as it is called by SvxNumberFormatShell.
A new GetCurrencySymbols is used by the control.
To avoid code duplication, both functions call a static private function
inner_GetCurrencySymbols.
MRU Currencies
MRU currencies are stored in another member vector m_mru_curencies:
During currencies vector population (inner_GetCurrencySymbols) a space is reserved
on top of vector for mru currencies.
A new method addMruCurrency updates m_currencies. This method is called in
SvxCurrencyToolBoxControl::execute
Change-Id: I6f86179efd1839e31b089e0086b6ddbdda358f31
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148896
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.com>
|
|
If there's a paragraph scoped tab default distance defined,
use that as the default tab distance in the SvxRuler instead
of the document wide setting.
Change-Id: I9b0e7d0db0b25aee08bd27948b2e462b4a4ee496
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148060
Tested-by: Jenkins
Reviewed-by: Sarper Akdemir <sarper.akdemir@collabora.com>
|
|
Change-Id: Id6381d77cdf05c24a1b557206f530f4e004e8e52
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148986
Tested-by: Jenkins
Reviewed-by: خالد حسني <khaled@aliftype.com>
|
|
Essentially reapply
commit 66291d467be4bf0e841814111d57d1af76117f8e
Date: Mon Jan 23 12:15:50 2023 -0600
tdf#153167 ImageMap Editor dialog had no title
after it has been reverted in
commit a90ef17809071b808a8c703e7e47e045ddabcd4a
Date: Tue Jan 24 08:36:48 2023 +0000
Revert "tdf#153167 ImageMap Editor dialog had no title"
This reverts commit 66291d467be4bf0e841814111d57d1af76117f8e.
Reason for revert: didn't wait for jenkins and broke the build
translatable strings need context attribute.
, but this time add the required context attribute
as well, as it used to be before
commit 8a35ae3e6c5c45aa1426cfd81472b309d02a1b48
Date: Thu May 16 15:17:38 2019 +0100
weld SvxIMapDlg
Co-authored-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
Change-Id: Id777e457fe144b0f0746cfc19e7e13e6db7a89a4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148914
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
|
|
Change-Id: I7833b6ab19334ea17f5210a0381109aa55d6efd0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148925
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I37b5056026f29d2619a4f9d3da9251924c7f5fdd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148898
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Stephan Bergmann <sbergman@redhat.com>
|
|
where we can avoid constructing temporary OUStrings
Change-Id: I0eacd68a8d1b450894c2ea769055f16886b78ad0
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148780
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Iaf5f20c8952b15f3dcccb65277dadb171a705605
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148606
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
to find more locking we can remove
Change-Id: Ief7bc5ec2a1ff31f22a0ad366910b7fcc4725818
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148599
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Add a map between ID and title and use that in a newly
introduced tooltip query handler to return the item's title
so it is shown as a tooltip.
This also implies that the title is used as the accessible
description for the items for the non-gtk case, see
`SalInstanceIconView::EntryAccessibleDescriptionHdl`,
introduced in
commit 2a28ebeef5ea3e2b01d836a7233d2316b765bf38
Date: Wed Jun 1 11:18:26 2022 +0300
Accessibility for IconView
Therefore, the NVDA screen reader on Windows now
announces the items using their title when they
receive focus.
Change-Id: Ic77cf485ed4b2b413d3d3368c15b788d693111cd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148523
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
Change-Id: I794e528a08a5a76cef1955f5c4f3e594f1e90f4a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148537
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Iecdeeda5c10b2322bdf71e84864ed8104581589f
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148424
Tested-by: Jenkins
Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
|
|
Calling `rRenderContext.GetFontCharMap(mxFontCharMap)`
in `SvxShowCharSet::RecalculateFont` means that a new
`FontCharMapRef` gets assigned to `mxFontCharMap`.
Therefore, clear the old items based on the previous
one from the map, so that new ones will be created
based on the new font char map in `SvxShowCharSet::ImplGetItem`
as needed instead of still returning the old ones.
Adapt the a11y UI test to check again that the "!" character
now has the proper accessible name. The incorrect
accessible name in the special characters dialog had
actually been uncovered by the initial version of the
test that failed on Windows due to this, and the test was adapted
to workaround that, s. tdf#153918 and the discussion in the
Gerrit change adding the test [1] for more details.
[1] https://gerrit.libreoffice.org/c/core/+/142260/24..26/sw/qa/extras/accessibility/dialogs.cxx#b85
Change-Id: I0ac578750b1fa0a7e63e62a6937a125c7ceab510
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148423
Reviewed-by: Colomban Wendling <cwendling@hypra.fr>
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
The `nCount` member counts the number of elements
in `m_aItemList`, but `std::unordered_map::size()`
also returns that, so just use that instead.
Use `SvxSearchCharSet::getMaxCharCount` in most places
and simplify to use `std::min` in `SvxSearchCharSet::LastInView`.
Change-Id: I8b66b1d83c0327026b13147f201995c7d1df1238
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148422
Tested-by: Jenkins
Reviewed-by: Michael Weghorn <m.weghorn@posteo.de>
|
|
What we got
- Most controls rendered on Writer and Impress (on Calc already
implemented by Tomaž Vajngerl)
- Text labels rendered correctly
- Mouse events forwarded to controls
- Control state changed on click for Writer and Calc
- Control invalidation for all apps
- Fixed broken LOK_CALLBACK_MOUSE_POINTER msg
- Correct pointer style when mouse is hovering over a control
Need to be improved
- in impress click method for a control is not executed even if the
mouse event is forwarded correctly
- avoid not needed control invalidations (as the one occurring on
document autosaving)
Change-Id: I4d5012af7f90a2c726b6b6b5b068e2be1ed5568a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/146569
Tested-by: Jenkins CollaboraOffice <jenkinscollaboraoffice@gmail.com>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/147320
Tested-by: Jenkins
Reviewed-by: Marco Cecchetti <marco.cecchetti@collabora.com>
|
|
until
commit eec42f0dbcc79a4c9f456ce97fa1066b8031ea28
Author: Noel Grandin <noelgrandin@gmail.com>
Date: Sun Aug 15 17:35:58 2021 +0200
pass OutlinerParaObject around by value
it served two purposes, now it serves only one, so rename it
to reflect that.
Change-Id: Ifc4f9361fd562cc6c572cdc3ad90151d87f41501
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148382
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
they are all one or two words in size
Change-Id: Iebbe17f114b098337305d4f8f5c3835be4744c30
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148376
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
was closing the whole dialog on tab switch in
Format -> Characters -> Highlight when switching to
"None"
also avoid sending multiple "full updates" by using the same
builder for sending message
Change-Id: I4540c9c857acf2b9f40482678a30335edc24e7cb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148314
Tested-by: Jenkins
Reviewed-by: Szymon Kłos <szymon.klos@collabora.com>
|
|
similar to FontFeaturesDialog.
FWIW gtk's viewport implementation automatically overwrites the step
size to be 10% of the visible height (and page size of 90% of the
visible height)
Change-Id: If356724a47b0a10069f129bed4a40cfdee57bfa8
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148138
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
During testing the NbcRotate function on SdrObject it turned out
the the function is essentially a noop, because it uses normal
equality to test the values of a double, which are a product of
sin and cos functions (to determine the 90 degree angles). So
because of this the input rectangle was never modified - noop.
Because of this we can just remove the impl. of the function and
declare it abstract, so that the actual implementations define
a valid function to rotate.
There were some subclasses that didn't override the NbcRotate so
they used the one implementation in SdrObject. These subclasses
now override the function and in the implementation we call
assert(false), which is never called during a test run. It seems
we never rotate those objects.
Change-Id: I1b1a45a8e96ed2d061f9b9f80c5fdaa5a84d4c05
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148266
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
This change makes SdrObject an abstract class that can't be as
we don't want it to be instantiated. With this the ClonseSdrObject
became a pure virtual function.
With this change also the SwFlyDrawObj needs a implementation of
the CloneSdrFunction, which was previously provided by SdrObject.
Change-Id: I2efb8c1e6fac12b17ce497285067029d7da1c1fd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148239
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Change-Id: I62182fa02843d428d1b745c55ab695450ec4940a
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148235
Tested-by: Tomaž Vajngerl <quikee@gmail.com>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Change-Id: I448f82d87be34c58d0abf4f6a0c1e6192567b807
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148234
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Change-Id: If430eace7e82ff52bb10c83b5f66148e2d9ade13
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148248
Tested-by: Tomaž Vajngerl <quikee@gmail.com>
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Due to the fact that handling of a vector GradientSteps
will get used more often with ongoing changes I moved
some handling to tooling, see sortAndCorrectColorSteps.
That method sorts and corrects a vector of ColorSteps
so that only valid entreis remain in a sorted order,
for details please refer to the docu at function
definition. I took the occasion to rearrange that to
work on the single provided vector which is better for
speed & ressources.
Also changed the ColorStep constructor to not
automatically correct constructed entries: While that
is formally correct, it moves an invalid entry to 0.0
or 1.0, thus creating additional wrong Start/EndColor
entries. Those may then 'overlay' the correct entry
when corrections are applied to the vector of entries
which leads to getting the wanted Start/EndColor to be
factically deleted, what is an error.
Also rearranged FillGradientAttribute to now work
initially with ColorSteps, no longer requires the
Start/EndColor, also adapted all usages. This securely
allows start/end and in-between single-color sections
in gradients.
Also needed to re-formulate GradientRect &
GradientElliptical ::appendTransformationsAndColors
method for correct 2D mapping(s) - always incrementing
from the start to end conflicts with the adaption of
the start value for the inner loop mainly because
increment is at start of inner loop (pre-increment).
Corrected errors from clang plugin, but also some
wrong initialiations for basegfx::ColorSteps.
Change-Id: I292592ed9abcfa591f68a680479f4fcdda46cbeb
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148196
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@me.com>
|
|
Change-Id: I123aef9f9763b2ae88c4b4f6748b69f9d9c394af
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148051
Tested-by: Jenkins
Reviewed-by: Andreas Heinisch <andreas.heinisch@yahoo.de>
|
|
There is no need to pass the rectangle by reference and to change
it inside the function, better to return a new instance.
Also clean-up the code of Poly2Rect and rename the function to
svx::polygonToRectangle.
Change-Id: I25e77c8abd12e2075939f55e06f40343ac23ca97
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148211
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Change-Id: I0a416fa2ac470650c2ef430dbb91bf8d5a8013cd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148210
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Also removed the "_10x22" from the icons
Change-Id: Iccdd5b784f43f1588aa1b5e72d01dd220209cc21
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148120
Tested-by: Jenkins
Reviewed-by: Rizal Muttaqin <rizmut@libreoffice.org>
|
|
This reverts commit d1046e7c3f66e5f3384ee1ef534ef28346702fc6.
and all its associated follow up fixes. Doesn't actually have any effect
on the issue which isolated it as a source of the problem which has
mutated in the meantime
Revert "tdf#126184 Use max paper dimensions to calculate clip region"
This reverts commit 1dccd6814f1fe7a06f3168d01d18d347269cd3c1.
and:
Revert "Resolves: tdf#101187 colossal page width wraps around if doubled"
This reverts commit 2fd88ab1cbb4690a770ca2ca5d66157ec4906a2e.
and:
Revert "Resolves: tdf#98366 paste document has 10x10 sized SdPages..."
This reverts commit 93efd7ebbad293d3729b8ea4b9726aff498f607f.
and:
Revert "Resolves: tdf#97276 don't clip out line if the underlying page size is unknown"
This reverts commit ad99c633908f7c70d06812ebfb4e0696666f0158.
Change-Id: I22d3e1592c172f137fcec8085ff3240428f0d248
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148170
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
|
|
Also removed the "_14" from the yes/no icons
Change-Id: Ic43219fe0e4555416b78831d45f9edc8060f73cc
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148063
Tested-by: Jenkins
Reviewed-by: Heiko Tietze <heiko.tietze@documentfoundation.org>
|
|
Change-Id: I267f3655dcc41613db9de4952beceb34dffefbac
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148061
Tested-by: Jenkins
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
|
|
Change-Id: I288f8fb3375e152b5ee746fab2c05d08150d6c99
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/146817
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
|
|
Remove visual artifacts around the corners by using horizontal and
vertical round.
Change-Id: I9821103c43dfbd93e6f295034b05da8b74e802e4
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/148040
Tested-by: Jenkins
Reviewed-by: Andreas Heinisch <andreas.heinisch@yahoo.de>
|