| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Reviewed-by: Trust Me
|
|
|
|
| |
Reviewed-by: Trust Me
|
|
|
|
| |
Reviewed-by: Trust Me
|
|
|
|
|
|
|
|
|
|
|
| |
Added WA_StyledBackground to QMenuBar when using style sheets. This also implies
that CE_PanelMenuBar (drawing only the menubar border) no longer needs to
be drawn.
Tested in uiloader/baselne/css_borderimage_allwidgets.ui.
Task-number: 230363
Reviewed-by: olivier
|
|
|
|
|
|
|
| |
The menu bar is currently broken with border-image,
fix is about to be commited
Task-number: 230363
|
|\ |
|
| |
| |
| |
| | |
Reviewed-by: Oswald Buddenhagen
|
| |
| |
| |
| |
| |
| |
| | |
string is a multi-length one
Task-number: QT-10
Reviewed-by: Oswald Buddenhagen
|
| |
| |
| |
| |
| | |
Reviewed-by: Oswald Buddenhagen
Task-number: QT-10
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
of the biggest string
In case the strings contains multiple strings separated by \x9c
Reviewed-by: Oswald Buddenhagen
Task-number: QT-10
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Patch originally from Oswald on Jira QT-10, with few a modifications.
If a string contains multiple variants sorted by decreasing length,
separated by \x9c, it will try to paint the longest variant which fits
into the bounding box.
Reviewed-by: Oswald Buddenhagen
Task-Number: QT-10
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
remove unused includes;
tabs -> whitespaces;
clean extra whitespaces
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| | |
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
un-inline isDirPath() since it too large for this (reduce
size of QtCore binary in a few kilobytes)
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
-in most cases GetFullPathName returns string with at least path.size() chars;
-". " isn't valid path; ". " isn't valid path too...should we to pay more?
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
isUncPath() is always called with native separator-ed paths, so we can avoid
needless comparisons;
don't declare isUncPath() under CE since it never used
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| | |
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| | |
this commit just moves closing bracket to the function end
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| | |
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
LocalDiskFlag actually means "Local File Engine" and can be effectively
used for testing file path for target storage type (local/network/virtual and so on)
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
_unix code always sets HiddenFlag for special dirs which is wrong;
also there is some inconsistence under win:
* FindFirstFile sets FILE_ATTRIBUTE_HIDDEN flag for ".." of hidden dir
*even* if parent dir is not hidden;
* GetFileAttributes sets FILE_ATTRIBUTE_HIDDEN flag for ".."
*only* if parent dir is hidden.
so, _win part sets HiddenFlag wrong too;
finally, we never test parent dir's flags;
futhermore hidden special dirs (dotAndDotDot) makes dir iterator's filtering a bit more complex
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| | |
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
tried_stat, could_stat, need_lstat, and is_link are members marked as mutable;
prefer mutable over const_cast
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| | |
which is a bit faster since it doesn't creates new file engine instance
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| | |
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| | |
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This function now returns early if a non-directory is
met in the path. Something like /foo/bar/&&/ will
bail out early.
Merge-request: 1176
Reviewed-by: Joerg Bornemann <joerg.bornemann@trolltech.com>
|
| |
| |
| |
| |
| |
| | |
This test used too much memory for Windows CE <= 5.
Reviewed-by: thartman
|
| |
| |
| |
| |
| |
| |
| |
| | |
This function was too strict. It returned 0 if the property wasn't of
type QVariant::Double. Now it tests for QMetaType::Float too.
Reviewed-by: kh1
Reviewed-by: mauricek
|
| |
| |
| |
| |
| |
| |
| |
| | |
qFuzzyCompare doesn't support 0 as parameter.
So this function is pretty useless for QPoint, where coordinates can
be zero.
Reviewed-by: Harald Fernengel
|
| |
| |
| |
| | |
Reviewed-by: mauricek
|
|/
|
|
| |
Reviewed-by: Trust Me
|
|\ |
|
| |
| |
| |
| | |
Reviewed-by: Kavindra
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
has been set.
The hasFrame() or frame property of the QStyleOptionSpinBox was not being copiedthrough to the QStyleOptionComboBox inside the paintEvent method if a
calendarPopup had been enabled.
Task-number:259510
Reviewed-by:Jens Bache-Wiig
|
| |
| |
| |
| |
| |
| |
| | |
This feature on Windows was added in Windows 2000, but its not so easy
to utilize until at least Vista was released.
Reviewed-by: Marius Storm-Olsen
|
| |\ |
|
| | |
| | |
| | |
| | | |
Reviewed-by: Maurice
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| |\ \
| | |/
| | |
| | |
| | | |
Conflicts:
src/gui/dialogs/qfiledialog_win.cpp
|
| | |
| | |
| | |
| | | |
Reviewed-by: Joerg
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Upload the glyph one scanline at a time to work around what probably is
a driver bug. This replaces a previous fix
6d0290b2202d4fc084595ba678c2a2d984392e72.
Reviewed-by: Tom
|
| | |
| | |
| | |
| | |
| | | |
The problem was that when switching source, the previous one was still
running. So we now explicitly stop it.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Instead of hard coding "make", use "$MAKE". If $MAKE is not set,
configure will export it for us when searching for "make".
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
small patch to make sure even the native events of the pending ones
won't trigger anything in the paintevent
Task-number: 251776
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Zero timers don't fire on the first pass in the GLib event
dispatcher. Ideally I should fix the bug of course, but time doesn't
permit at the moment. Submitting this test instead as a "reminder".
The UNIX event dispatcher passes the test, and it also passes if
moved to the end of the slots.
Task: 259505
|
| | |
| | |
| | |
| | | |
Reviewed-by: Trust Me
|