KDE CI: Frameworks ktexteditor kf5-qt5 FreeBSDQt5.7 - Build # 34 - Still Unstable!
BUILD UNSTABLE Build URL https://build.kde.org/job/Frameworks%20ktexteditor%20kf5-qt5%20FreeBSDQt5.7/34/ Project: Frameworks ktexteditor kf5-qt5 FreeBSDQt5.7 Date of build: Tue, 25 Jul 2017 02:47:02 + Build duration: 25 min and counting JUnit Tests Name: (root) Failed: 1 test(s), Passed: 65 test(s), Skipped: 0 test(s), Total: 66 test(s)Failed: TestSuite.bug205447 build.log Description: Binary data
KDE CI: Frameworks krunner kf5-qt5 FreeBSDQt5.7 - Build # 18 - Still Unstable!
BUILD UNSTABLE Build URL https://build.kde.org/job/Frameworks%20krunner%20kf5-qt5%20FreeBSDQt5.7/18/ Project: Frameworks krunner kf5-qt5 FreeBSDQt5.7 Date of build: Tue, 25 Jul 2017 02:46:42 + Build duration: 4 min 26 sec and counting JUnit Tests Name: (root) Failed: 1 test(s), Passed: 0 test(s), Skipped: 0 test(s), Total: 1 test(s)Failed: TestSuite.runnercontexttest build.log Description: Binary data
D6877: Add Conan files
This revision was automatically updated to reflect the committed changes. Closed by commit R243:802d106c8e81: Add Conan files (authored by gladhorn). REPOSITORY R243 KArchive CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6877?vs=17127=17153 REVISION DETAIL https://phabricator.kde.org/D6877 AFFECTED FILES conanfile.py test_package/CMakeLists.txt test_package/conanfile.py test_package/example.cpp To: gladhorn, dfaure Cc: dfaure, apol, cfeck, #frameworks
D6856: Reset url in closeUrl()
dfaure requested changes to this revision. dfaure added a comment. This revision now requires changes to proceed. The URL changes twice because you make it change twice ;) But that can lead to the application updating stuff twice, which doesn't seem very optimal. (actions getting disabled and then re-enabled, title bar flickering, stuff like that). Ideally it should change only once, either from oldURL to newURL, or from oldURL to QUrl() if loading failed. This might require a special bool checked by closeUrl(), since we can't just split closeUrl (because it's public *and* virtual - bad design practice - I was young). (if it wasn't virtual, then moving the code into a closeUrlImpl and having the public closeUrl() do closeUrlImpl()+setUrl(QUrl()) would be the best solution, but we can't do that since openUrl has to call the virtual closeUrl, not just closeUrlImpl. So I can't think of another solution than a bool member set by openUrl and checked by closeUrl to skip the setUrl(QUrl()) until after we try to open the new URL). REVISION DETAIL https://phabricator.kde.org/D6856 To: elvisangelaccio, dfaure, #ktexteditor, #kate, #kdevelop Cc: cullmann, #frameworks
D6856: Reset url in closeUrl()
elvisangelaccio edited the summary of this revision. elvisangelaccio edited the test plan for this revision. REVISION DETAIL https://phabricator.kde.org/D6856 To: elvisangelaccio, dfaure, #ktexteditor, #kate, #kdevelop Cc: cullmann, #frameworks
D6889: TestPart: add more test cases
elvisangelaccio edited the test plan for this revision. REPOSITORY R306 KParts REVISION DETAIL https://phabricator.kde.org/D6889 To: elvisangelaccio, #kate, #kdevelop, dfaure Cc: #frameworks
D6889: TestPart: add more test cases
elvisangelaccio added dependencies: D6890: Emit urlChanged() only if we actually open the URL, D6856: Reset url in closeUrl(). REPOSITORY R306 KParts REVISION DETAIL https://phabricator.kde.org/D6889 To: elvisangelaccio, #kate, #kdevelop, dfaure Cc: #frameworks
D6856: Reset url in closeUrl()
elvisangelaccio added a dependent revision: D6889: TestPart: add more test cases. REVISION DETAIL https://phabricator.kde.org/D6856 To: elvisangelaccio, dfaure, #ktexteditor, #kate, #kdevelop Cc: cullmann, #frameworks
D6856: Reset url in closeUrl()
elvisangelaccio added a comment. @cullmann After thinking more about it, the URL does change twice (old url -> QUrl(), QUrl() -> new url), so it should be expected to get two urlChanged() signals. Consider this scenario: - openUrl(foo) - part loads url foo. - openUrl(bar) - openUrl() calls closeUrl() - part fails to load url bar. In this case we only get one urlChanged() signal since we could not open the new url. REVISION DETAIL https://phabricator.kde.org/D6856 To: elvisangelaccio, dfaure, #ktexteditor, #kate, #kdevelop Cc: cullmann, #frameworks
KDE CI: Frameworks ktexteditor kf5-qt5 FreeBSDQt5.7 - Build # 33 - Still Unstable!
BUILD UNSTABLE Build URL https://build.kde.org/job/Frameworks%20ktexteditor%20kf5-qt5%20FreeBSDQt5.7/33/ Project: Frameworks ktexteditor kf5-qt5 FreeBSDQt5.7 Date of build: Mon, 24 Jul 2017 21:34:02 + Build duration: 9 min 48 sec and counting JUnit Tests Name: (root) Failed: 1 test(s), Passed: 65 test(s), Skipped: 0 test(s), Total: 66 test(s)Failed: TestSuite.bug205447 build.log Description: Binary data
D6856: Reset url in closeUrl()
elvisangelaccio updated this revision to Diff 17149. elvisangelaccio added a comment. - Use setUrl() to reset the URL. CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6856?vs=17070=17149 BRANCH reset-url-in-closeUrl REVISION DETAIL https://phabricator.kde.org/D6856 AFFECTED FILES src/readonlypart.cpp To: elvisangelaccio, dfaure, #ktexteditor, #kate, #kdevelop Cc: cullmann, #frameworks
D6889: TestPart: add more test cases
elvisangelaccio created this revision. Restricted Application added a project: Frameworks. Restricted Application added a subscriber: Frameworks. REVISION SUMMARY TestPart is now able to simulate a part that fails to open an URL. REPOSITORY R306 KParts BRANCH improve-testpart REVISION DETAIL https://phabricator.kde.org/D6889 AFFECTED FILES autotests/parttest.cpp autotests/parttest.h To: elvisangelaccio, #kate, #kdevelop, dfaure Cc: #frameworks
D6829: Add ability to use the new kauth helper in file ioslave
dfaure requested changes to this revision.
dfaure added a comment.
This revision now requires changes to proceed.
Good work, just some small things, and a suggested simplification to the
overall slave<->job protocol.
INLINE COMMENTS
> slavebase.h:941
> + * with elevated privileges. It is similar to data() but has different
> name
> + * so as not to mess up existing connections.
> + */
It's not similar to data(), it doesn't send data from the file being read ;-)
Please remove that sentence from the docu, and rename the signal (see
https://phabricator.kde.org/D6832).
> slaveinterface.cpp:323
> +case MSG_PRIVILEGE_EXEC:
> +emit privilegeExecData(rawdata);
> +break;
Do the bytearray -> enum conversion here, so that everyone else than slavebase
and slaveinterface (i.e. both the job and the slave) only see the enum.
QDataStream can be used to stream an int (the enum value) in and out, no need
to use "strings" in the bytearray. Alternatively, QByteArray::number() can do
the job as long as it's really just the one enum value that's needed here (and
we know it will never need to be more...).
> slaveinterface.h:89
> // add new ones here once a release is done, to avoid breaking binary
> compatibility
> +MSG_PRIVILEGE_EXEC
> };
Move the new value above the "add new ones here" comment, otherwise the next
person will break BC.
> file_unix.cpp:670
> +
> +privilegeExecData("AskConfirmation");
> +readPrivilegeExecData(jobReply);
Wait, why the two step roundtrip here?
Instead of "can you become root?" / "yes I can" / "ok please ask the user"
/ "OK, the user said yes", the protocol here could really be simplified to
"request to elevate privilege" "OK (it's enabled, and the user said yes)".
Which even removes the need for any enum, it's really just one request and one
reply which is true or false.
> file_unix.cpp:691
> +QStringList testData;
> +testData += execAction.name();
> +testData += QString::number(execAction.isValid());
use C++11 initializer to avoid reallocations, i.e. `QStringList
testData{execAction.name(), ..., ..., ...};`
Ah but then you just use a join(","), so the most efficient thing to do here
would be to drop the QStringList and do
const QString metaData = execAction.name() + ',' + QString::number... + ',' +
etc.
REPOSITORY
R241 KIO
REVISION DETAIL
https://phabricator.kde.org/D6829
To: chinmoyr, dfaure, #frameworks
Cc: #frameworks
D6832: Integrate new file ioslave in KIO job
dfaure requested changes to this revision.
dfaure added inline comments.
This revision now requires changes to proceed.
INLINE COMMENTS
> job.cpp:345
> +m_confirmationAsked = true;
> +return status == 5;
> +}
This hardcoded 5 is awful, please use SlaveBase::Continue instead.
> job_base.h:222
> + */
> +bool isPrivilegeExecutionEnabled() const;
> +
This doesn't need to be public, does it? Could be in the private class.
> job_base.h:229
> + */
> +bool wasConfirmationAsked() const;
> +
This doesn't need to be public, does it? Could be in the private class.
> job_p.h:94
> +FileOperationType m_operationType;
> +bool tryAskPrivilegeOpConfirmation();
> +
Please keep all the member variables together. In the orig code it's already
ugly that there are methods after the variables, but this makes it slightly
worse with the enum in the middle and no empty line between vars and methods.
Ideally this should be
[...] private enum, private methods, all private member vars, end of class
> job_p.h:195
> /**
> + * Called when privilegeExecData() is emmited by the slave.
> + *
typo: emitted
> simplejob.cpp:151
>
> +q->connect(slave, SIGNAL(privilegeExecData(QByteArray)),
> + SLOT(slotPrivilegeExecData(QByteArray)));
This is not a very good signal name.
Signals notify a state changed and usually end with "ed".
Would it make sense to call this privilegeOperationRequested() or something?
I need to find more about when this is emitted and what the data is ;)
> simplejob.cpp:346
> +QByteArray reply;
> +if (data == QStringLiteral("CanElevatePrivilege")) {
> +if (m_privilegeExecutionEnabled
use QLatin1String rather than QStringLiteral for comparisons
> simplejob.cpp:354
> +if (confirmed) {
> +reply = QByteArray("ActionConfirmed");
> +}
OK, so the data is two possible strings, and the reply is two possible strings,
why not use enums instead? You can define them in SlaveBase for instance.
You might have to use lambdas to avoid the enum appearing in Q_PRIVATE_SLOT in
public headers (and requiring slavebase.h)
REPOSITORY
R241 KIO
REVISION DETAIL
https://phabricator.kde.org/D6832
To: chinmoyr, dfaure, #frameworks
Cc: #frameworks
D6833: Add support for PrivilegeExecution in KIO jobs
dfaure added inline comments.
INLINE COMMENTS
> chmodjob.cpp:238
> KIO::SimpleJob *job = KIO::chmod(info.url, info.permissions);
> +if (m_privilegeExecutionEnabled) {
> +job->setParentJob(q);
Maybe all these if()s can be removed? I assume setParentJob just stores a
pointer, we could do this in all case --- it's less code, and just in case we
need the parent job for something else later, it'll be available.
> copyjob.cpp:284
> +FileOperationType copyType;
> +if (mode == CopyJob::Copy) {
> +copyType = Copy;
switch() ?
> copyjob.cpp:1674
> } else if (m_mode == CopyJob::Move) { // Moving a file
> -KIO::FileCopyJob *moveJob = KIO::file_move(uSource, uDest,
> permissions, flags | HideProgressInfo/*no GUI*/);
> +KIO::FileCopyJob *moveJob = KIO::file_move(uSource, uDest,
> permissions, flags | HideProgressInfo/*no GUI*/ | PrivilegeExecution);
> +if (m_privilegeExecutionEnabled) {
Should the flag always be set here? Or only if m_privilegeExecutionEnabled is
true?
REPOSITORY
R241 KIO
REVISION DETAIL
https://phabricator.kde.org/D6833
To: chinmoyr, dfaure, #frameworks
Cc: #frameworks
D6877: Add Conan files
dfaure accepted this revision. dfaure added a comment. This revision is now accepted and ready to land. I'll use this as basis for generating such files with an ECM macro, similar to what we do for pri files. REPOSITORY R243 KArchive BRANCH master REVISION DETAIL https://phabricator.kde.org/D6877 To: gladhorn, dfaure Cc: dfaure, apol, cfeck, #frameworks
D6877: Add Conan files
gladhorn updated this revision to Diff 17127. gladhorn marked an inline comment as done. gladhorn added a comment. fix url REPOSITORY R243 KArchive CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6877?vs=17126=17127 BRANCH master REVISION DETAIL https://phabricator.kde.org/D6877 AFFECTED FILES conanfile.py test_package/CMakeLists.txt test_package/conanfile.py test_package/example.cpp To: gladhorn Cc: apol, cfeck, #frameworks
D6877: Add Conan files
gladhorn marked 3 inline comments as done. gladhorn added inline comments. INLINE COMMENTS > apol wrote in conanfile.py:5 > Can we re-use the version from the `CMakeLists.txt`? The goal is to generate it. > apol wrote in conanfile.py:7 > Re-use from metainfo.yml? For now we consider this as a proof of concept. > apol wrote in conanfile.py:9 > I don't understand what this means :( building shared libs or static ones. REPOSITORY R243 KArchive REVISION DETAIL https://phabricator.kde.org/D6877 To: gladhorn Cc: apol, cfeck, #frameworks
D6877: Add Conan files
gladhorn updated this revision to Diff 17126. gladhorn added a comment. Updated according to comments REPOSITORY R243 KArchive CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6877?vs=17124=17126 BRANCH master REVISION DETAIL https://phabricator.kde.org/D6877 AFFECTED FILES conanfile.py test_package/CMakeLists.txt test_package/conanfile.py test_package/example.cpp To: gladhorn Cc: apol, cfeck, #frameworks
D6877: Add Conan files
gladhorn updated this revision to Diff 17124. gladhorn added a comment. remove -git from the version REPOSITORY R243 KArchive CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6877?vs=17118=17124 BRANCH master REVISION DETAIL https://phabricator.kde.org/D6877 AFFECTED FILES conanfile.py test_package/CMakeLists.txt test_package/conanfile.py test_package/example.cpp To: gladhorn Cc: apol, cfeck, #frameworks
D6800: Add template for a simple kpart-based application
aacid added a comment. Ok, so i guess that would apply to kparts too. REPOSITORY R306 KParts REVISION DETAIL https://phabricator.kde.org/D6800 To: kossebau, #frameworks, dfaure Cc: lueck, ltoscano, aacid
D6877: Add Conan files
gladhorn updated this revision to Diff 17118. gladhorn added a comment. Fix license REPOSITORY R243 KArchive CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6877?vs=17116=17118 BRANCH master REVISION DETAIL https://phabricator.kde.org/D6877 AFFECTED FILES conanfile.py test_package/CMakeLists.txt test_package/conanfile.py test_package/example.cpp To: gladhorn Cc: apol, cfeck, #frameworks
D6800: Add template for a simple kpart-based application
lueck added a comment.
In https://phabricator.kde.org/D6800#128307, @kossebau wrote:
>
> I cannot see though any catalogs which would match the template-string
catalog names, so seems there is some magic already to prevent that?
l10n-kf5/scripts/extract-messages.sh:
# skip Messages.sh files of KDevelop's app templates
grep '{APPNAMELC}[^ ]*.pot' $subdir/Messages.sh 1>/dev/null && continue
REPOSITORY
R306 KParts
REVISION DETAIL
https://phabricator.kde.org/D6800
To: kossebau, #frameworks, dfaure
Cc: lueck, ltoscano, aacid
D6876: Fix install apk target
apol added a comment. Makes sense to me, I can reproduce the issue, I cannot reproduce the fix because `androiddeployqt` doesn't accept my system's gradle. REPOSITORY R240 Extra CMake Modules REVISION DETAIL https://phabricator.kde.org/D6876 To: aacid, apol, mart, kossebau Cc: #frameworks, #build_system
D6877: Add Conan files
gladhorn updated this revision to Diff 17116. gladhorn added a comment. Remove private test stuff REPOSITORY R243 KArchive CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6877?vs=17115=17116 BRANCH master REVISION DETAIL https://phabricator.kde.org/D6877 AFFECTED FILES conanfile.py test_package/CMakeLists.txt test_package/conanfile.py test_package/example.cpp To: gladhorn Cc: apol, cfeck, #frameworks
D6877: Add Conan files
apol added inline comments.
INLINE COMMENTS
> conanfile.py:3
> +
> +class KarchiveConan(ConanFile):
> +name = "KArchive"
`KArchiveConan`?
> conanfile.py:5
> +name = "KArchive"
> +version = "5.37.0-git"
> +license = "lgpl2"
Can we re-use the version from the `CMakeLists.txt`?
> conanfile.py:7
> +license = "lgpl2"
> +url = " package>"
> +settings = "os", "compiler", "build_type", "arch"
Re-use from metainfo.yml?
> conanfile.py:9
> +settings = "os", "compiler", "build_type", "arch"
> +options = {"shared": [True, False]}
> +default_options = "shared=True"
I don't understand what this means :(
> conanfile.py:4
> +
> +class KarchiveTestConan(ConanFile):
> +settings = "os", "compiler", "build_type", "arch"
`KArchiveTestConan`
REPOSITORY
R243 KArchive
REVISION DETAIL
https://phabricator.kde.org/D6877
To: gladhorn
Cc: apol, cfeck, #frameworks
D6877: Add Conan files
cfeck added inline comments.
INLINE COMMENTS
> example.cpp:6
> +KZip a("somefile");
> +FrederikTest ttt;
> +return 0;
:)
REPOSITORY
R243 KArchive
REVISION DETAIL
https://phabricator.kde.org/D6877
To: gladhorn
Cc: cfeck, #frameworks
D6877: Add Conan files
gladhorn created this revision. Restricted Application added a project: Frameworks. Restricted Application added a subscriber: Frameworks. REVISION SUMMARY Conan is a C++ package manager. This is an experiment to make KF5 frameworks available on Conan to make building KDE applications easier. See also https://conan.io/ . TODO: - detect version number automatically REPOSITORY R243 KArchive BRANCH master REVISION DETAIL https://phabricator.kde.org/D6877 AFFECTED FILES conanfile.py test_package/CMakeLists.txt test_package/conanfile.py test_package/example.cpp To: gladhorn Cc: #frameworks
D6800: Add template for a simple kpart-based application
kossebau added a subscriber: ltoscano. kossebau added a comment. In https://phabricator.kde.org/D6800#128256, @aacid wrote: > At least there is a problem l10n wise in which there is a Messages.sh so i guess it will be run and create a "fake" .pot file (and something similar for the .desktop file) > > Though probably this is already happening? Good question, had not thought about before. Indeed the existing template elsewhere also have pure Messages.sh in the tree, e.g. https://cgit.kde.org/kapptemplate.git/tree/src/templates/C++/kde-frameworks5 https://cgit.kde.org/kapptemplate.git/tree/src/templates/C++/kpartapp/src https://cgit.kde.org/kapptemplate.git/tree/src/templates/C++/akonadiresource as well as https://cgit.kde.org/plasma-framework.git/tree/templates/cpp-plasmoid https://cgit.kde.org/plasma-framework.git/tree/templates/plasma-wallpaper Looking at https://websvn.kde.org/trunk/l10n-kf5/templates/messages/kdesdk/ https://websvn.kde.org/trunk/l10n-kf5/templates/messages/frameworks/ I cannot see though any catalogs which would match the template-string catalog names, so seems there is some magic already to prevent that? @ltoscano can you raise the curtain on this matter? :) REPOSITORY R306 KParts REVISION DETAIL https://phabricator.kde.org/D6800 To: kossebau, #frameworks, dfaure Cc: ltoscano, aacid
D6876: Fix install apk target
aacid created this revision. Restricted Application added projects: Frameworks, Build System. Restricted Application added subscribers: Build System, Frameworks. REPOSITORY R240 Extra CMake Modules BRANCH fix_install_apk REVISION DETAIL https://phabricator.kde.org/D6876 AFFECTED FILES toolchain/Android.cmake To: aacid, apol, mart, kossebau Cc: #frameworks, #build_system
D6866: Add Doxyfile highlighting
dhaumann abandoned this revision. dhaumann added a comment. Submitted: https://commits.kde.org/syntax-highlighting/916efa110fffc33d6350f3ca4a7ec73cf5b2b8e3 Just borked the log message, so I'll just abandon this. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6866 To: dhaumann, emaurer, kfunk Cc: #frameworks
D6871: Let the indexer check for keywords with leading/trailing spaces
This revision was automatically updated to reflect the committed changes. Closed by commit R216:9f43abc06b21: indexer checks for leading/trailing whitespaces in XML highlighting definitions (authored by cullmann). REPOSITORY R216 Syntax Highlighting CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6871?vs=17105=17113 REVISION DETAIL https://phabricator.kde.org/D6871 AFFECTED FILES src/indexer/katehighlightingindexer.cpp To: cullmann, vkrause, dhaumann Cc: #frameworks
D5338: Add syntax highlighting file for Doxyfile
This revision was automatically updated to reflect the committed changes. Closed by commit R216:916efa110fff: Doxygen: Add Doxyfile highlighting (authored by dhaumann). CHANGED PRIOR TO COMMIT https://phabricator.kde.org/D5338?vs=13205=17112#toc REPOSITORY R216 Syntax Highlighting CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D5338?vs=13205=17112 REVISION DETAIL https://phabricator.kde.org/D5338 AFFECTED FILES autotests/folding/Doxyfile.example.fold autotests/html/Doxyfile.example.html autotests/input/Doxyfile.example autotests/reference/Doxyfile.example.ref data/syntax/doxyfile.xml To: kfunk, dhaumann, kwrite-devel, vkrause Cc: turbov, emaurer, #frameworks
D6875: Add --gradle to androiddeployqt
aacid created this revision. Restricted Application added projects: Frameworks, Build System. Restricted Application added subscribers: Build System, Frameworks. REVISION SUMMARY Otherwise when trying to build with the newest Qt/android toolchains it fails REPOSITORY R240 Extra CMake Modules BRANCH add_gradle REVISION DETAIL https://phabricator.kde.org/D6875 AFFECTED FILES toolchain/Android.cmake To: aacid, apol, kossebau, mart Cc: #frameworks, #build_system
D6871: Let the indexer check for keywords with leading/trailing spaces
dhaumann accepted this revision. dhaumann added a comment. This revision is now accepted and ready to land. Looks good to me... REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6871 To: cullmann, vkrause, dhaumann Cc: #frameworks
D6870: add missing standard types to C highlighting and update to C11
This revision was automatically updated to reflect the committed changes. Closed by commit R216:c40ad3513f91: add missing standard types to C highlighting and update to C11 (authored by brauch). CHANGED PRIOR TO COMMIT https://phabricator.kde.org/D6870?vs=17104=17109#toc REPOSITORY R216 Syntax Highlighting CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6870?vs=17104=17109 REVISION DETAIL https://phabricator.kde.org/D6870 AFFECTED FILES data/syntax/c.xml data/syntax/isocpp.xml To: brauch, vkrause, dhaumann Cc: vkrause, dhaumann, #frameworks
D6870: add missing standard types to C highlighting and update to C11
dhaumann accepted this revision. dhaumann added a comment. This revision is now accepted and ready to land. Please inc the version number. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6870 To: brauch, vkrause, dhaumann Cc: vkrause, dhaumann, #frameworks
D6800: Add template for a simple kpart-based application
aacid added a comment. At least there is a problem l10n wise in which there is a Messages.sh so i guess it will be run and create a "fake" .pot file (and something similar for the .desktop file) Though probably this is already happening? Any idea on how we would make that not happen? REPOSITORY R306 KParts REVISION DETAIL https://phabricator.kde.org/D6800 To: kossebau, #frameworks, dfaure Cc: aacid
D6871: Let the indexer check for keywords with leading/trailing spaces
cullmann added a comment. It is no error ATM, as first one needs to cleanup the xml files with something like: perl -pi -e 's@\s*(\S+)\s*@\1@gi' *.xml REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6871 To: cullmann, vkrause, dhaumann Cc: #frameworks
D6871: Let the indexer check for keywords with leading/trailing spaces
cullmann created this revision. Restricted Application added a project: Frameworks. Restricted Application added a subscriber: Frameworks. REVISION SUMMARY Let the indexer check for keywords with leading/trailing spaces, they lead to not needed allocs for trimmed() after loading. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6871 AFFECTED FILES src/indexer/katehighlightingindexer.cpp To: cullmann, vkrause, dhaumann Cc: #frameworks
D6870: add missing standard types to C highlighting and update to C11
brauch added reviewers: vkrause, dhaumann. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6870 To: brauch, vkrause, dhaumann Cc: vkrause, dhaumann, #frameworks
D6870: add missing standard types to C highlighting and update to C11
brauch created this revision. Restricted Application added a project: Frameworks. Restricted Application added a subscriber: Frameworks. REVISION SUMMARY See bug: https://bugs.kde.org/show_bug.cgi?id=367798 My suggestion is to add the new C11 keywords to the C highlighting, add the missing standard types (like time_t) to both C and iso-C++, and ignore the POSIX stuff. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6870 AFFECTED FILES data/syntax/c.xml data/syntax/isocpp.xml To: brauch Cc: vkrause, dhaumann, #frameworks
D6867: Add PowerShell highlighting
vkrause closed this revision. vkrause added a comment. Pushed to master. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6867 To: vkrause, dhaumann, vonreth Cc: #frameworks
D6867: Add PowerShell highlighting
dhaumann accepted this revision. dhaumann added a comment. This revision is now accepted and ready to land. Looks good to me. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6867 To: vkrause, dhaumann, vonreth Cc: #frameworks
D6867: Add PowerShell highlighting
vkrause updated this revision to Diff 17103. vkrause added a comment. Right, license is due to the original file. REPOSITORY R216 Syntax Highlighting CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6867?vs=17094=17103 REVISION DETAIL https://phabricator.kde.org/D6867 AFFECTED FILES autotests/folding/craftenv.ps1.fold autotests/html/craftenv.ps1.html autotests/input/craftenv.ps1 autotests/reference/craftenv.ps1.ref data/syntax/powershell.xml To: vkrause, dhaumann, vonreth Cc: #frameworks
D6867: Add PowerShell highlighting
dhaumann added a comment. Ah, and we need spellChecking="0" almost everywhere except for HereString and String, and comments right? REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6867 To: vkrause, dhaumann, vonreth Cc: #frameworks
D6866: Add Doxyfile highlighting
dhaumann updated this revision to Diff 17097. dhaumann added a comment. Update again to reflect unit testing changes. REPOSITORY R216 Syntax Highlighting CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6866?vs=17096=17097 REVISION DETAIL https://phabricator.kde.org/D6866 AFFECTED FILES autotests/folding/Doxyfile.example.fold autotests/html/Doxyfile.example.html autotests/input/Doxyfile.example autotests/reference/Doxyfile.example.ref data/syntax/doxyfile.xml To: dhaumann, emaurer, kfunk Cc: #frameworks
D6866: Add Doxyfile highlighting
dhaumann updated this revision to Diff 17096. dhaumann added a comment. Remove section header, since it did not work as expected at all. REPOSITORY R216 Syntax Highlighting CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D6866?vs=17092=17096 REVISION DETAIL https://phabricator.kde.org/D6866 AFFECTED FILES autotests/folding/Doxyfile.example.fold autotests/html/Doxyfile.example.html autotests/input/Doxyfile.example autotests/reference/Doxyfile.example.ref data/syntax/doxyfile.xml To: dhaumann, emaurer, kfunk Cc: #frameworks
D5338: Add syntax highlighting file for Doxyfile
emaurer added a comment. In https://phabricator.kde.org/D5338#128106, @dhaumann wrote: > @emaurer We just agreed that your version is better and would like to use this. Are you also OK with using the MIT license? Currently, we are trying to unify the licensing to MIT for as many of our xml highlighting files as possible. yes, confirmed, you can use the code with MIT license. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D5338 To: kfunk, dhaumann, kwrite-devel, vkrause Cc: turbov, emaurer, #frameworks
D6867: Add PowerShell highlighting
dhaumann requested changes to this revision. dhaumann added a comment. This revision now requires changes to proceed. Looks almost good, but I think another iteration would be nice. I assume we cannot relicense to MIT? INLINE COMMENTS > powershell.xml:5 > + version="1" > + kateversion="3.4" > + extensions="*.ps1;*.ps1m;*.ps1d" If I remember correctly, the default style dsBuiltIn was added with the first release of KDE Frameworks 5, see all bold entries here: https://kate-editor.org/2014/03/07/kate-part-kf5-new-default-styles-for-better-color-schemes/ So kateversion="5.0" is required here. > powershell.xml:918 > + > + > + Since we require kateversion 5.0 anyways, we could also use dsVerbatimString instead of dsString. > powershell.xml:922 > + > + selColor="#ff" bold="0" italic="0" spellChecking="false" /> > + color="#FF" selColor="#ff" bold="1" italic="0" spellChecking="false" > /> Suggestion: use dsVariable for variables, and remove all hardcoded colors and bold and italic. > powershell.xml:923 > + selColor="#ff" bold="0" italic="0" spellChecking="false" /> > + color="#FF" selColor="#ff" bold="1" italic="0" spellChecking="false" > /> > + dsConstant would be better. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6867 To: vkrause, dhaumann, vonreth Cc: #frameworks
D6867: Add PowerShell highlighting
vkrause created this revision. Restricted Application added a project: Frameworks. Restricted Application added a subscriber: Frameworks. REVISION SUMMARY Based on https://github.com/jgm/highlighting-kate/pull/80/ with some fixes and improvements. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6867 AFFECTED FILES autotests/folding/craftenv.ps1.fold autotests/html/craftenv.ps1.html autotests/input/craftenv.ps1 autotests/reference/craftenv.ps1.ref data/syntax/powershell.xml To: vkrause, dhaumann, vonreth Cc: #frameworks
D6866: Add Doxyfile highlighting
dhaumann created this revision. Restricted Application added a project: Frameworks. Restricted Application added a subscriber: Frameworks. REVISION SUMMARY This adds Doxyfile highlighting support, initially taken from https://github.com/KDE/syntax-highlighting/pull/5 and slightly modified. Also, changed license to MIT. TEST PLAN Added unit test created with doxygen -g. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D6866 AFFECTED FILES autotests/folding/Doxyfile.example.fold autotests/html/Doxyfile.example.html autotests/input/Doxyfile.example autotests/reference/Doxyfile.example.ref data/syntax/doxyfile.xml To: dhaumann, emaurer, kfunk Cc: #frameworks
KDE CI: Frameworks ktexteditor kf5-qt5 FreeBSDQt5.7 - Build # 30 - Fixed!
BUILD SUCCESS Build URL https://build.kde.org/job/Frameworks%20ktexteditor%20kf5-qt5%20FreeBSDQt5.7/30/ Project: Frameworks ktexteditor kf5-qt5 FreeBSDQt5.7 Date of build: Mon, 24 Jul 2017 09:13:09 + Build duration: 24 min and counting JUnit Tests Name: (root) Failed: 0 test(s), Passed: 66 test(s), Skipped: 0 test(s), Total: 66 test(s) build.log Description: Binary data
D5338: Add syntax highlighting file for Doxyfile
kfunk abandoned this revision. kfunk added a comment. Note: Abandoning this Diff in favor of the one from @emaurer REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D5338 To: kfunk, dhaumann, kwrite-devel, vkrause Cc: turbov, emaurer, #frameworks
D5338: Add syntax highlighting file for Doxyfile
dhaumann added a comment. @emaurer We just agreed that your version is better and would like to use this. Are you also OK with using the MIT license? Currently, we are trying to unify the licensing to MIT for as many of our xml highlighting files as possible. REPOSITORY R216 Syntax Highlighting REVISION DETAIL https://phabricator.kde.org/D5338 To: kfunk, dhaumann, kwrite-devel, vkrause Cc: turbov, emaurer, #frameworks
D6856: Reset url in closeUrl()
elvisangelaccio planned changes to this revision. elvisangelaccio added a comment. In https://phabricator.kde.org/D6856#128029, @cullmann wrote: > Hi, I see the issue that the url is changed without calling setUrl. That doesn't emit urlChanged() that way. > On the other side, if setUrl would be used, you will get the urlChanged twice in openUrl which might lead to not-wanted side-effects. Good catch, this patch is wrong but the fact that the part has an URL set even if it failed to open that URL is a bug that should be fixed, imho. I'll try another approach. REPOSITORY R306 KParts REVISION DETAIL https://phabricator.kde.org/D6856 To: elvisangelaccio, dfaure, #ktexteditor, #kate, #kdevelop Cc: cullmann, #frameworks
KDE CI: Frameworks ktexteditor kf5-qt5 FreeBSDQt5.7 - Build # 29 - Unstable!
BUILD UNSTABLE Build URL https://build.kde.org/job/Frameworks%20ktexteditor%20kf5-qt5%20FreeBSDQt5.7/29/ Project: Frameworks ktexteditor kf5-qt5 FreeBSDQt5.7 Date of build: Mon, 24 Jul 2017 08:31:06 + Build duration: 10 min and counting JUnit Tests Name: (root) Failed: 1 test(s), Passed: 65 test(s), Skipped: 0 test(s), Total: 66 test(s)Failed: TestSuite.katesyntaxtest build.log Description: Binary data
