| Age | Commit message (Collapse) | Author |
|---|
|
ksharedconfig_in_global_object.cpp is now in kdelibs4 too
(where it works) and reproduces Albert's KgDifficulty testcase.
CHANGELOG: fix assert when using KSharedConfig in a global object destructor.
REVIEW: 122232
|
|
|
|
QByteArrayLiteral
|
|
Add new variable to specify it in *.kcfgc : "CategoryLoggingName"
CHANGELOG: Allow to generate file with qloggingcategories support.
REVIEW: 122931
|
|
|
|
The local config file gets locked on line 415, if we return early we
need to unlock it.
REVIEW: 122549
BUG: 344032
|
|
which was always TRUE
REVIEW: 122445
|
|
This reverts commit 9eee15917e01a89d937d1cba2eebbe9d65daeb72. As agreed
with David Faure, we are going to leave this change in place for 5.7 and
see what happens.
CCMAIL: faure@kde.org
CHANGELOG: Reverts previous commit, both can be left out of the changelog.
|
|
On further reflection, introducing a SIC to fix the bug is not ideal. I'm
working on something that will address the underlying bug, but I didn't have
it in time for 5.7.
The versioning behaviour is left, as I do want it to exist for the future.
This only reverts the mandatory versioning of files to avoid the compatibility
break.
CHANGELOG: Undo 915976c123, which never saw a release, enough to avoid the SIC.
|
|
With -Winconsistent-missing-override (Clang), headers from KConfig are
throwing a lot of warnings.
REVIEW: 122539
|
|
when parsing the default value for string lists, skip empty parts
also, an empty string means an empty list (not a list containing
an empty string)
|
|
Fixes build with Qt 5.5
|
|
|
|
REVIEW: 121838
Acked by Matthew Dawson
|
|
This patch is necessary because:
When we use kf5 + kde4 application, kconf_update which launchs at the start when we launch kde.
But it will migrate some config file, for example it will show that we need to migrate konversation
so it will create a konversationrc in .config/
But when we launch konversation there is a kdelibs4migrator which wants to migrate settings and config in .config
but it shows a konversationrc in .config so it will never migrate and we will lose all settings.
So we can force to remove all .upd in kf5 but it will not fix problem during migration or when we have kde4 application
install in same directory as kf5.
So now I force for each upd file to have a "Version=5" so kconf_update (kf5) will migrate just kf5 upd file and it will fix my bugs.
REVIEW: 121797
CHANGELOG: Now kconf_update doesn't process upd file from kde4. We need to add "Version=5" in top of the upd file otherwise it will be skipped.
|
|
|
|
|
|
|
|
KSharedConfig has been ported to QStandardPaths in KF5.
REVIEW: 121272
|
|
REVIEW:120947
|
|
|
|
this way we actually get the same shared pointer and thus does not
require a sync() call on the config object to get the right values
REVIEW:120283
|
|
|
|
gcc 4.5 doesn't support "#pragma GCC diagnostic" inside functions.
Apparently it also doesn't support "#pragma GCC diagnostic push/pop",
but it doesn't fail for that, so meh.
|
|
|
|
Add a class description to KConfigBase so that KConfig::sync can link to
KConfigBase::sync. Also makes KConfigBase easier to find and understand.
|
|
This is needed so doxygen includes the far more useful documentation on
the contained methods.
|
|
Switch to using QT_TRANSLATE_NOOP3 so that context information can be
given for actions, allowing the translators to do better translations.
BUG: 337979
FIXED-IN: 5.2.0
REVIEW: 119577
|
|
KConfig fails building because QBasicAtomicInt is being treated as an int.
REVIEW: 119417
|
|
using Qt-standard QT_TRANSLATE_NOOP instead of a fake I18N_NOOP2
|
|
like all the other Qt-based modules do
|
|
|
|
Previously the warning "Invalid escape sequence "\;"." would appear and
"\;" was replaced with just the backslash as is done for all
unrecognized escape sequences. Keep both characters so that
readXdgListEntry() works with values containing semicolons
REVIEW: 119074
|
|
|
|
This enables the mainConfig optimization in all threads,
and ensures the user warning only happens in the main thread.
The test-mode-enabled logic is only really useful in the main thread,
but it's simpler to just do it in all threads.
REVIEW: 118985
|
|
|
|
|
|
... by having a different list of shareable objects per thread.
REVIEW: 118739
|
|
This fixes a regression introduced in
988f09bb051dca0437ecec431ee44ed5b4a560d8.
The mentioned commit ensures that if the locale is e.g. "de_DE" the
entry "de" will be used. But this breaks if there is a translation
for another country. E.g. for "de_CH" it would also pick the "de"
entry.
This change now operates on both just the language code and the locale.
If an entry with the language code is present it will be picked. If
another entry with the exact locale is found it will be overwritten.
Modifiers are not supported as this is currently missing in QLocale.
REVIEW: 118692
|
|
Yet another awesome application of the Qt implicit sharing trick.
Since config files often contain only few different keys and even
value strings, we can share them. This reduces memory consumption
and also speeds up parsing, as we do not have to allocate the
duplicated strings, but can simply reuse the previous values.
The most extreme case for this of my knowledge, is KatePart:
katesyntaxhighlightingrc has more than 20k lines which triggered
nearly 30k allocations on startup. With this patch applied, this
value goes down dramatically. I added a simple static counter for
the cache hit/miss ratio, which resulted in 5442 cache misses compared
to 43624 cache hits across all KConfig files parsed by kwrite.
REVIEW: 118587
This is a forward-port of b8aaeff128233cfaecf67899168887572589dde8.
|
|
Before, these kind of read-only operations did a lot of allocations:
1) allocate a list of all sub groups
2) for the above, also allocate a sub-group match key
3) iterate over sub groups, allocate a list of all keys in there
and then finally check whether that list is non-empty
All of the above is now done without a single allocation, by simply
iterating over the list of entries.
Note: The whole list was iterated even before in allSubGroups. Now
we still do that, but check for non-empty keys in the group or
sub group directly. Much more efficient.
Note2: While at it, allSubGroups is also optimized to not require the
allocation of the subgroup match key.
REVIEW: 118586
forward-port of commit eaffd50adfd7fcbeafadb0248904176808b4499d
|
|
forward-port of commit 416559cdb86f9cacf4ea4ed7530cbbbdb427302f
|
|
The underlying KConfig used QLocale::name() for getting the locale
aware part. But this returns "de_DE" while the desktop files store
"de".
In addition it constructs a QLocale object instead of using the
system locale. This has the advantage that the usage of
QLocale::setDafault() gets honored by KConfig.
REVIEW: 118564
|
|
- use Q_NULLPTR instead of 0 or NULL
- simplify some foreach loops
- use QStringLiteral in a few places
- added a few consts
REVIEW: 118666
|
|
kreadconfig only needs to print help when a key is not given or when extra
text is on the end of the command line.
BUG: 335806
|
|
This should have been part of 8b88eaee.
REVIEW: 118339
|
|
According to http://qt-project.org/doc/qt-5/qvariant.html#type the
return value of QVariant::type() should be interpreted as a
QMetaType::Type.
This also silences the compile-time warnings about 38 not being a member
of QVariant::Type.
REVIEW: 118332
|
|
Simplify KConfigGroup::readEntry/writeEntry.
KConfigGroup::readCheck/writeCheck caused some compilers to complain
since they were private functions that were being reimplemented outside
the class definition.
Updated KCONFIGGROUP_DECLARE_ENUM_QOBJECT/
deprecated global functions created in KDE4 times.
|
|
Previous code said if we found an executable in kconf_update_bin
try to find a different executable in the normal path
Reviewed-by: Rohan Garg
|
|
/home/scripty/prod/git-unstable/frameworks_kconfig/src/core/bufferfragment_p.h:37: Ignoring definition of undeclared qualified class
/home/scripty/prod/git-unstable/frameworks_kconfig/src/core/bufferfragment_p.h:37: Ignoring definition of undeclared qualified class
|
