The built-in code model also proposes "void_t".
Change-Id: I2462e47d5b1a69065f998244fc8372011e37a0c8
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
The items in the treeView look slightly different than
with Qt 5.10 and thus seem to lose the mouseclicks.
Change-Id: Ibee1abdbc9e9bfe622b45f76ed5e612e81fd1770
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
If we get correct http responses that point to server side
issues handle them appropriate if possible to let the test
continue instead of crash.
Change-Id: I7d2e8848269600762e9c7fe980414c4f7106433c
Reviewed-by: Robert Loehning <robert.loehning@qt.io>
Having an additional MouseArea as child of a ScrollView or a Flickable
is not well defined and leads to inconsistent behavior on different
systems. We can easily catch the relevant events in the FlameGraph item
itself. Also, don't redirect the typeSelected() signals through the
model. They don't belong there.
Change-Id: I77c17977b5a51d57ccd2ef880d3d6c6a604b7f78
Task-number: QTCREATORBUG-20573
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
ClangCodeModel currently does not provide a list of overrides.
Therefore it makes sense to use ClangCodeModel result for
virtual method only if built-in code model does not find anything.
Task-number: QTCREATORBUG-20584
Change-Id: I5b4fac7974f990e741d3438ab61827670a8ce8d8
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@qt.io>
Reproducable with
1. $ ./qtcreator a.cpp b.cpp
2. Switch to a.cpp => no highlighting
Because ClangEditorDocumentProcessor does asynchronous processing, the
backend might receive a DocumentsOpenedMessage where the document is not
the current editor (happens for a.cpp in the example). When switching to
that document, the initial jobs were not processed as the document was
not dirty.
Address this case by also checking for documents that have a revision of
1 and are not dirty. Unify adding the annotations jobs to ensure that
not more than needed are run.
Change-Id: I14030260842f97d58280235e763c8d7490705f8d
Reviewed-by: Ivan Donchevskii <ivan.donchevskii@qt.io>
The warnings from ClangCodeModel confuse our build check.
Change-Id: Ic6b82bb304387f447bdf8eb0b2fd02dd2f250fb2
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
...with runAsync().
The stack size cannot be changed after starting a thread, so specifying
the stack size with a pool does not make sense. However, starting
with Qt 5.10 a stack size can be specified for the whole thread pool, if
needed.
Change-Id: I09eded606321388c779f762b77de6223081609fe
Reviewed-by: Eike Ziller <eike.ziller@qt.io>
This is long overdue since some names were simply wrong and/or
misleading. Also, some of the old names were long enough to almost get
crazy.
The renaming starts from ClangCodeModelServerInterface and
ClangCodeModelClientInterface and affects usages and related functions.
For the ClangCodeModelServerInterface, categorize the messages in
- messages that require a response (request*)
- notification messages (the remaining ones)
Change-Id: I5342ed8e0d87404ee72f3c3766fd8ef7505defb1
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Reviewed-by: Ivan Donchevskii <ivan.donchevskii@qt.io>
When built-in code model fails to follow symbol under cursor
fall back to the clang result even if it only follows
to the decalration.
Change-Id: I22d8c5fee6ab7594b1d1b7ce8104414db28383c7
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@qt.io>
Replace backslashes with normal slashes as Qt handles them
inside paths correctly itself. Otherwise we might end up
using unknown escape sequences which leads to not executing
anything and a passing test.
Escaping the blanks is not necessary as they are handled
correctly automatically.
Change-Id: Ib1a81949c54d41b7864e30a1371de2e148f96bb8
Reviewed-by: Tobias Hunger <tobias.hunger@qt.io>
This should not expire after another year.
Change-Id: I86ed49d491f619509e6bd32fc2624df2a1c24f5f
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
...now that parsing and creating the preamble is faster due to the
skipped function bodies.
As a consequence, we can remove all the extra jobs that were needed to
get an initial AST faster.
Change-Id: I79a66b8a0e8a180850af6daf353d9a679089bbb1
Reviewed-by: Ivan Donchevskii <ivan.donchevskii@qt.io>
Failing tests:
TokenProcessor.TemplateFunctionCall
TokenProcessor.TemplateClassDeclaration
Do not introduce fatal error in test source file.
Otherwise template function is not recognized anymore.
Such template calls were also broken before but could
provide a valid Cursor kind. So it's a minor regression
in Clang but does not change anything for Qt Creator user.
Bug is reported: https://bugs.llvm.org/show_bug.cgi?id=37550
Change-Id: I788e8d9f88141cfefc6bbde77f36c459d91aff0b
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@qt.io>
Do not counstruct NativeFilePathView from temporary object.
Change-Id: Ifcd6bc4878f6949e98de44089a2c2b3feca4795a
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@qt.io>
Adapt versions and tests, remove code assuming clang < 6.0.
Switch also to our custom repositories instead of dealing with patch
files.
LLVM/Clang 6 was released on 09 Mar 2018.
Task-number: QTCREATORBUG-18535
Task-number: QTCREATORBUG-18552
Change-Id: I0ec2c2f56265e161ae7cbb5b03e7b8a182ba6cc6
Reviewed-by: Ivan Donchevskii <ivan.donchevskii@qt.io>
...since it's superseded by the tidy integration.
Change-Id: Idafa5e1fb5129b1af8e42231a664684d4b90821f
Reviewed-by: Ivan Donchevskii <ivan.donchevskii@qt.io>
Calling it several times inside the same run results in
unexpected behavior. The list of items inside the Locator
gets filled asynchronously, so give it a chance to populate
before trying to click something.
Change-Id: I542a9fda37ae09e98e496d12a40d507a040bb247
Reviewed-by: Robert Loehning <robert.loehning@qt.io>
gtest-creator-printing.cpp:493:12: warning: enumeration value
'OverloadedOperator' not handled in switch [-Wswitch]
Change-Id: Ic28dcbcc402a8f5c0f2bd284cc9b6fd560221208
Reviewed-by: Ivan Donchevskii <ivan.donchevskii@qt.io>
This allows us to share code between the two, in particular the QML code
for the Details window, and the theme code. This way we can potentially
deduplicate some code.
Change-Id: I3a0d26b18488bd2a46b5b077b5b5d79ac2dfc5ce
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Reviewed-by: Eike Ziller <eike.ziller@qt.io>
This change limits the set of tokens that fall under
Token::isOperator(). That allows cpphighlighter.cpp to
distinguish operator tokens from punctuator tokens
(without changing any logic in cpphighlighter.cpp).
This change moves punctuators from "Operator"
to the "Text" style category where they belong.
Punctuators are not operators. Punctuators are
dumb text tokens.
Why don't we let the clang backend alone separate
these tokens for us?
1. Clang is slow on big files. Sometimes the
highlighting dictated by clang is painted _seconds_
after cpphighlighter.cpp runs. CppHighlighter is way
faster so we use it to "prepaint" code while clang is
busy in the background.
2. Secondly, clang cannot yet handle all operator types.
In particular, none if its "operator cursors"
CXCursor_UnaryOperator:
CXCursor_BinaryOperator:
CXCursor_CompoundAssignOperator:
CXCursor_ConditionalOperator:
includes the -> and . operators.
We still need CppHighlighter to paint those tokens.
However, once clang has finished processing the file some
operator tokens will be repainted. We need clang to get
all operators' semantics. In particular, we need clang to
tell us if < is a "smaller than"-operator or part of a
template parameter like set<int>.
Task-number: QTCREATORBUG-19659
Change-Id: I952cb58f7c79134b3281e2a8221425cc1d0ad263
Reviewed-by: Ivan Donchevskii <ivan.donchevskii@qt.io>
Color types and enable Ctrl+click for the functions
and types inside SIGNAL/SLOT macros.
Change-Id: Ic1c0b7372fe9a73c5607b1973d75a6656c75ef0e
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@qt.io>
