D17085: Fix language load after including keywords from this language in another file

2018-11-23 Thread Nibaldo González 
nibags closed this revision.

REPOSITORY
  R216 Syntax Highlighting

REVISION DETAIL
  https://phabricator.kde.org/D17085

To: nibags, #framework_syntax_highlighting, cullmann, dhaumann, vkrause, jpoelen
Cc: kwrite-devel, kde-frameworks-devel, michaelh, ngraham, bruns, demsking, 
cullmann, sars, dhaumann

D17092: Add syntax highlighting for TypeScript & TypeScript React

2018-11-23 Thread Nibaldo González 
nibags updated this revision to Diff 46105.
nibags added a comment.


  - typescript.xml & typescript-react.xml: add some comments and the operator 
"!" after objects
  
  
https://github.com/Microsoft/TypeScript/wiki/What's-new-in-TypeScript#non-null-assertion-operator

REPOSITORY
  R216 Syntax Highlighting

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17092?vs=45992=46105

BRANCH
  add-typescript

REVISION DETAIL
  https://phabricator.kde.org/D17092

AFFECTED FILES
  autotests/folding/highlight.js.fold
  autotests/folding/test.js.fold
  autotests/folding/test.jsx.fold
  autotests/folding/test.qml.fold
  autotests/folding/test.ts.fold
  autotests/folding/test.tsx.fold
  autotests/html/highlight.js.html
  autotests/html/test.jsx.html
  autotests/html/test.ts.html
  autotests/html/test.tsx.html
  autotests/input/test.jsx
  autotests/input/test.ts
  autotests/input/test.tsx
  autotests/reference/highlight.js.ref
  autotests/reference/test.jsx.ref
  autotests/reference/test.ts.ref
  autotests/reference/test.tsx.ref
  data/syntax/javascript-react.xml
  data/syntax/javascript.xml
  data/syntax/typescript-react.xml
  data/syntax/typescript.xml

To: nibags, #framework_syntax_highlighting, dhaumann, cullmann
Cc: kwrite-devel, kde-frameworks-devel, michaelh, ngraham, bruns, demsking, 
cullmann, sars, dhaumann

D17093: Rust & Yacc/Bison: improve comments

2018-11-23 Thread Nibaldo González 
nibags updated this revision to Diff 46104.
nibags added a comment.


  - rust.xml: fix indentation.
  
  The file uses tabs, but in one part I used spaces as indentation

REPOSITORY
  R216 Syntax Highlighting

CHANGES SINCE LAST UPDATE
  https://phabricator.kde.org/D17093?vs=45993=46104

BRANCH
  improve-comments-rust-yacc

REVISION DETAIL
  https://phabricator.kde.org/D17093

AFFECTED FILES
  autotests/folding/highlight.y.fold
  autotests/folding/test.rs.fold
  autotests/html/highlight.y.html
  autotests/html/test.rs.html
  autotests/input/highlight.y
  autotests/input/test.rs
  autotests/reference/highlight.y.ref
  autotests/reference/test.rs.ref
  data/syntax/rust.xml
  data/syntax/yacc.xml

To: nibags, #framework_syntax_highlighting, dhaumann, cullmann
Cc: kwrite-devel, kde-frameworks-devel, michaelh, ngraham, bruns, demsking, 
cullmann, sars, dhaumann

D16919: Re-add DBus test

2018-11-23 Thread Kurt Hindenburg 
hindenburg added a comment.


  In D16919#364934 , @bcooksley 
wrote:
  
  > Yes, that is basically what the CI does so should be a pretty good method 
of reproducing it's results.
  >  Did you run "make tests" / "ctest" or did you use the run-tests.py script 
to execute the unit tests?
  
  
  The python test script

REPOSITORY
  R319 Konsole

REVISION DETAIL
  https://phabricator.kde.org/D16919

To: davidedmundson, #konsole, #frameworks, hindenburg
Cc: bcooksley, hindenburg, konsole-devel, ngraham, maximilianocuria

D17128: DocumentPrivate: Remove comment mark when joining lines

2018-11-23 Thread loh tar 
loh.tar added a comment.


  This works so far in most cases I have tried.
  
  - C/C++, // and /* */ works, but sadly not Oxygen style comments. Guess the 
single star is not as special marked. Could that be fixed in the highlight 
stuff?
  - Bash, # works
  - Ruby, # works
  - Python, # works
  - PHP, // works, # works NOT which surprised me because they are as comment 
colored
  - Tcl, # works, backslash comment continuation works not but it's not 
surprising

REPOSITORY
  R39 KTextEditor

REVISION DETAIL
  https://phabricator.kde.org/D17128

To: loh.tar, #ktexteditor
Cc: kwrite-devel, kde-frameworks-devel, #ktexteditor, michaelh, ngraham, bruns, 
demsking, head7, cullmann, kfunk, sars, dhaumann

D17128: DocumentPrivate: Remove comment mark when joining lines

2018-11-23 Thread loh tar 
loh.tar created this revision.
loh.tar added a reviewer: KTextEditor.
loh.tar added a project: KTextEditor.
Herald added projects: Kate, Frameworks.
Herald added subscribers: kde-frameworks-devel, kwrite-devel.
loh.tar requested review of this revision.

REVISION SUMMARY
  ...from the next line when current line is comment.

REPOSITORY
  R39 KTextEditor

REVISION DETAIL
  https://phabricator.kde.org/D17128

AFFECTED FILES
  src/document/katedocument.cpp

To: loh.tar, #ktexteditor
Cc: kwrite-devel, kde-frameworks-devel, #ktexteditor, michaelh, ngraham, bruns, 
demsking, head7, cullmann, kfunk, sars, dhaumann

D16919: Re-add DBus test

2018-11-23 Thread Ben Cooksley 
bcooksley added a comment.


  Yes, that is basically what the CI does so should be a pretty good method of 
reproducing it's results.
  Did you run "make tests" / "ctest" or did you use the run-tests.py script to 
execute the unit tests?

REPOSITORY
  R319 Konsole

REVISION DETAIL
  https://phabricator.kde.org/D16919

To: davidedmundson, #konsole, #frameworks, hindenburg
Cc: bcooksley, hindenburg, konsole-devel, ngraham, maximilianocuria

D17123: KTextEditor::ViewPrivate: Avoid warning "Text requested for invalid range"

2018-11-23 Thread Christoph Cullmann 
cullmann accepted this revision.
cullmann added a comment.
This revision is now accepted and ready to land.


  Looks good, thanks.

REPOSITORY
  R39 KTextEditor

REVISION DETAIL
  https://phabricator.kde.org/D17123

To: loh.tar, #ktexteditor, cullmann
Cc: cullmann, kwrite-devel, kde-frameworks-devel, #ktexteditor, michaelh, 
ngraham, bruns, demsking, sars, dhaumann

D17123: KTextEditor::ViewPrivate: Avoid warning "Text requested for invalid range"

2018-11-23 Thread loh tar 
loh.tar created this revision.
loh.tar added a reviewer: KTextEditor.
loh.tar added a project: Frameworks.
Herald added a project: Kate.
Herald added subscribers: kde-frameworks-devel, kwrite-devel.
loh.tar requested review of this revision.

REVISION SUMMARY
  ...when cut or copy a line without selection

REPOSITORY
  R39 KTextEditor

REVISION DETAIL
  https://phabricator.kde.org/D17123

AFFECTED FILES
  src/view/kateview.cpp

To: loh.tar, #ktexteditor
Cc: kwrite-devel, kde-frameworks-devel, #ktexteditor, michaelh, ngraham, bruns, 
demsking, cullmann, sars, dhaumann

T10112: KDEInstallDirs on Mac

2018-11-23 Thread René J . V . Bertin 
rjvbb created this task.
rjvbb added a project: Build System.

TASK DESCRIPTION
  (I hope this is the better place to discuss something that should be done 
than a WIP diff review...)
  
  There is a discrepancy under the Mac's OS (and unpatched Qt) between the 
install locations expected in the code and the install locations used by the 
build system. The former depend on QStandardPaths for the most part while the 
latter are hardcoded and depend on the install prefix, assuming an 
XDG-compatible installation layout.
  
  As a result several developers have come up with workarounds. I myself have 
developed a patch for QStandardPaths allowing to make it return XDG-compatible 
locations while still supporting native locations (based on a compile-time 
decision using preprocessor tokens). This is IMHO the preferred solution for 
package managers that aim to install everything into a dedicated prefix, like 
MacPorts, Fink and HomeBrew.
  Others have designed solutions to build standalone app bundles in order to 
mimick the simple installer-less way of distributing software that is possible 
on Mac.
  
  I've been thinking of better ways to get things installed where the running 
software will be looking for it. Initially I thought that a simple wrapper 
around QStandardPaths could be written so that cmake can simply look up the 
actual locations. After going over `KDEInstallDirs.cmake` I'm no longer certain 
if that's feasible, and there might be bootstrap issues I'm not aware of.
  
  I notice though that almost all locations in KDEInstallDirs are based on the 
`DATAROOTDIR` path. Indeed, a quick test with the patched DrKonqi (from D4929 
) and
  
if (WIN32)
_define_relative(DATAROOTDIR BINDIR "data"
"read-only architecture-independent data root"
SHARE_INSTALL_PREFIX)
elseif(APPLE AND NOT APPLE_FORCE_INSTALL_PREFIX)
_define_absolute(DATAROOTDIR "/Library/Application Support"
"read-only architecture-independent data root")
else()
_define_absolute(DATAROOTDIR "share"
"read-only architecture-independent data root"
SHARE_INSTALL_PREFIX)
endif()
  
  gave an install that worked just as well as with my patched Qt and 
`APPLE_FORCE_INSTALL_PREFIX=ON` or `APPLE_FORCE_INSTALL_PREFIX=OFF`. 
(APPLE_FORCE_INSTALL_PREFIX could also be called APPLE_FORCE_UNIX_DIRS or even 
APPLE_FORCE_UNIX_STANDARDS as a generic switch usable elsewhere too).
  
  There are a few variables though for which I don't know if this is the right 
thing (and I might be missing others):
  
  - KDE_INSTALL_APPDIR: this is one where querying QStandardPaths would give an 
undesirable result (/Applications). But in general this might have to be kept 
at $SHARE_INSTALL_PREFIX/applications so that any non-KDE5 XDG-compliant 
software the user might have installed in $PREFIX will be able to interact as 
expected with KDE software
  - KDE_INSTALL_MANDIR: Apple manpages are installed in standard Unix locations
  - KDE_INSTALL_DBUS* : these are presumable used by KDE AND the D-Bus daemon 
which is typically installed through MacPorts, Fink or HomeBrew (with $PREFIX 
one of /opt/local, /sw or /usr/local).
  
  Additional interrogation:
  
  - Can we set `DATAROOTDIR=/Library/Application Support/KDE` so that 
everything remains nicely bundled? I notice that Qt tend to use 
`/Library/Application Support/QtProject` for its stuff but can we be certain 
that the extra path component won't break any lookups?
  
  I'd like to address this so that it becomes possible to do quick builds and 
installs KDE projects against a stock Qt on Mac in a way that's as similar as 
possible to how it's done on other Unix versions. It could be simpler than I 
thought but I'd need some feedback to avoid introducing regressions or cause 
too much breakage of the standalone-appbundle approaches designed by others 
(setting `APPLE_FORCE_INSTALL_PREFIX=ON` should ensure that?).

TASK DETAIL
  https://phabricator.kde.org/T10112

To: rjvbb
Cc: #kde_applications, #frameworks, #build_system, rjvbb

D14000: qqc2-desktop-style: basic support for QGuiApplication-based apps

2018-11-23 Thread Marco Martin 
mart accepted this revision.
This revision is now accepted and ready to land.

REPOSITORY
  R858 Qt Quick Controls 2: Desktop Style

REVISION DETAIL
  https://phabricator.kde.org/D14000

To: rjvbb, #frameworks, mart
Cc: alexeymin, davidedmundson, mart, broulik, plasma-devel, ragreen, Pitel, 
ZrenBot, lesliezhai, ali-mohamed, jensreuterberg, abetts, sebas, apol

D16919: Re-add DBus test

2018-11-23 Thread Kurt Hindenburg 
hindenburg added a comment.


  In D16919#364587 , @bcooksley 
wrote:
  
  > When you did your reproduction attempt @hindenburg did you use the binaries 
the CI system itself uses, or did you build Frameworks, etc yourself?
  
  
  I used the script in the comments by knauss in 
https://phabricator.kde.org/T3689 which I thought is what the CI uses.  If 
there's other instructions, let me know.

REPOSITORY
  R319 Konsole

REVISION DETAIL
  https://phabricator.kde.org/D16919

To: davidedmundson, #konsole, #frameworks, hindenburg
Cc: bcooksley, hindenburg, konsole-devel, ngraham, maximilianocuria

D17122: RFC: Use wl_display_add_socket_auto by default

2018-11-23 Thread Fabian Vogt 
fvogt created this revision.
fvogt added reviewers: KWin, Plasma.
Herald added a project: Frameworks.
Herald added a subscriber: kde-frameworks-devel.
fvogt requested review of this revision.

REVISION SUMMARY
  If there is no socketName specified explicitly, it will use
  wl_display_add_socket_auto to allocate the next free socket.
  The resulting name can be retrieved using socketName after a successful
  start afterwards.
  
  This is a RFC for now because this changes the default value of socketName
  (and causes the respective autotest to fail).
  
  If that's ok, I'll adjust the test accordingly. If not, please suggest how 
this
  should be implemented, a new autoSocketName bool property?

TEST PLAN
  Ran the new autotest, passes. kwin_wayland picks the next free
  socket without modifications and WAYLAND_DISPLAY is set properly.

REPOSITORY
  R127 KWayland

BRANCH
  master

REVISION DETAIL
  https://phabricator.kde.org/D17122

AFFECTED FILES
  autotests/server/test_display.cpp
  src/server/display.cpp
  src/server/display.h

To: fvogt, #kwin, #plasma
Cc: kde-frameworks-devel, michaelh, ngraham, bruns

T10112: KDEInstallDirs on Mac

2018-11-23 Thread René J . V . Bertin 
rjvbb added a comment.


  > Can we set DATAROOTDIR=/Library/Application Support/KDE so that everything 
remains nicely bundled?
  
  Sadly not without patching any code, apparently.
  
  We'll also need something like this I think:
  
_define_relative(DATADIR DATAROOTDIR ""
"read-only architecture-independent data"
DATA_INSTALL_DIR)
if(APPLE AND NOT APPLE_FORCE_INSTALL_PREFIX)
_define_non_cache(DATADIR_KF5 "/Library/Application Support/kf5")
else()
_define_non_cache(DATADIR_KF5 "${CMAKE_INSTALL_DATADIR}/kf5")
endif()
if(NOT KDE_INSTALL_DIRS_NO_DEPRECATED)
set(KF5_DATA_INSTALL_DIR "${CMAKE_INSTALL_DATADIR_KF5}")
endif()

TASK DETAIL
  https://phabricator.kde.org/T10112

To: rjvbb
Cc: kde-mac, #kde_applications, #frameworks, #build_system

D12674: Mark `Phonon4Qt5` dependency as optional in CMakeLists file

2018-11-23 Thread Alexander Schlarb 
ntninja added inline comments.

INLINE COMMENTS

> davidedmundson wrote in CMakeLists.txt:79
> What about this comment?

Well, it isn't actually required and it compiles just fine with both canberra 
and phonon missing.

REPOSITORY
  R289 KNotifications

REVISION DETAIL
  https://phabricator.kde.org/D12674

To: ntninja, kde-frameworks-devel, apol
Cc: davidedmundson, kde-frameworks-devel, apol, ltoscano, cgiboudeaux, 
michaelh, ngraham, bruns

D12674: Mark `Phonon4Qt5` dependency as optional in CMakeLists file

2018-11-23 Thread David Edmundson 
davidedmundson added inline comments.

INLINE COMMENTS

> CMakeLists.txt:79
>  DESCRIPTION "Qt-based audio library"
> -# This is REQUIRED since you cannot tell CMake "either one of those 
> two optional ones are required"
> -TYPE REQUIRED

What about this comment?

REPOSITORY
  R289 KNotifications

REVISION DETAIL
  https://phabricator.kde.org/D12674

To: ntninja, kde-frameworks-devel, apol
Cc: davidedmundson, kde-frameworks-devel, apol, ltoscano, cgiboudeaux, 
michaelh, ngraham, bruns