When writing to stdout and stderr from two processes, their output could become
interleaved. To work around that, we write stdout and stderr to different files
and later combine them together in the shell script.
Since tst_deviceshell tests could run for a long time if /usr folder
is too big, added a check that first tests the runtime once.
Since we currently only support linux containers, limit the tests to
only run if the container platform is linux as well.
Change-Id: I4b313596cdf9acc839d54d7cc77c66fd53ac23bf
Reviewed-by: hjk <hjk@qt.io>
A Behavior will be added as a normal ModelNode to the default
property, but we store the property name in behaviorPropertyName.
The value of behaviorPropertyName cannot be changed after the
ModelNode was created, since I do not see any use case and it keeps
things simple.
Change-Id: I69ba1d4d706432cfbbd35b001238f623e6e0b4fd
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Marco Bubke <marco.bubke@qt.io>
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
Some of our auto tests fail, so add an EXCLUDE_FROM_PRECHECK option for
add_qtc_test, which sets a corresponding label on the ctest, and exclude
these when running ctest on COIN.
Disable the Googletest based unittests for now, they are inconsistently
failing on CI, e.g. sometimes timing out after 60 seconds for tests that
usually take only a fraction of a second.
tst_perfdata from the perfparser submodule also fails. Temporarily
disable it explicitly, until the EXCLUDE_FROM_PRECHECK change can be
merged there.
Change-Id: I7e9d7aded75bbe8800f82f7aa125c181271a2a1f
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Jarek Kobus <jaroslaw.kobus@qt.io>
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
Increase the timeout to default 30 seconds when
waiting for finished. Apparently 2 seconds were
sometimes too short. We still ensure that waitForFinished
didn't timeout but finished eariler.
Change-Id: I12ca647b45537174c7c66d271e7c8a477184f2ff
Reviewed-by: Eike Ziller <eike.ziller@qt.io>
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
When outputting a commands stdout / stderr / exitcode,
the shell was free to interlace different outputs.
This could trip the parsing of the output, leading
to the shell missing the exit code of a command.
When that happened, the caller might wait indefinitely
for a result that would never come.
To workaround, the shell script now writes each command output
to a temporary file first, which is then written to stdout
by a single subshell, that way guaranteeing that the output
of multiple commands will not interleave
Change-Id: I9d8e7e788f5922c612ff533e5ba063f61a22aa8c
Reviewed-by: hjk <hjk@qt.io>
The bundled "base64" tool of macos < 10.15 did not support
the "-d" option ( only -D ). We disable the tests running
on the local shell to keep them from failing.
Change-Id: I6af353869033ed15cefedc03e2562621d1ea9904
Reviewed-by: Eike Ziller <eike.ziller@qt.io>
Depending on context, callgrind sees contributions of >8% to
the total cost of project parsing for these functions. The
functional are actualy executed executed out-of-line, often
for a function body of one "payload" instruction only.
Inlining removes the call/endbr64/ret overhead.
Change-Id: I6886f08e322fcaa4e0f54d424279e0a8c24e4718
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
When we reparent to a new node that was created
during the same transaction, then this node has
no position. In this case we have to delete the
reparented nodes and they will be created
as part of the creation of the new node, since they
are children.
Change-Id: Icd1d02f29f529fc0f00809f7ecebf3eabfdc9a5c
Reviewed-by: Miikka Heikkinen <miikka.heikkinen@qt.io>
As on it, correct the handling of the launcher interface.
This still does not work due to wrong call of valgrind-fake.
Change-Id: Id835e84eee29720579ac2947906bb95f84377185
Reviewed-by: hjk <hjk@qt.io>
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
instead of QTEST_MAIN. Reduces the initialization that is done by the Qt
test applications, and can also reduce interference with normal OS
operations like the current window loosing focus.
Change-Id: If88f289281aa1c8703ac7d4dbe0799d067c16588
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
On (debian based) linux the "command -v base64" did fail to find
the base64 exectuable. Instead we now use "which".
Change-Id: I9b57527d93952b08bcb9abc7484a7e6891aa664c
Reviewed-by: Christian Stenger <christian.stenger@qt.io>
The quitBlockingProcess() test examines different types
of process quitting and illustrates differences in
behavior.
Change-Id: I9209f00576e03eef66fbdf5665351138ed437ac9
Reviewed-by: hjk <hjk@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Previously the runInShell and outputForRunInShell methods were
exclusively processed single threaded, meaning all calls
were processed sequentially. With the multiplexed
helper script we can now run multiple processes simultaneously.
( see tst_manual_deviceshell )
Additionally the new script allows us to capture both
stdout and stderr from commands which was not possible previously.
Change-Id: I52f4fb46d872dc274edb9c11872d2f6543741b34
Reviewed-by: David Schulz <david.schulz@qt.io>
Reviewed-by: Jarek Kobus <jaroslaw.kobus@qt.io>
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
Round 1 - focus on headers.
For classes with initial in range [T-Z].
Replace QT_FORWARD_DECLARE_CLASS with QT_BEGIN_NAMESPACE and
QT_END_NAMESPACE and forward declare inside.
Change-Id: I4ac3a8391e6167aa2db3973a9f94a45ac3c8ebd1
Reviewed-by: hjk <hjk@qt.io>
Round 1 - focus on headers.
For classes with initial in range [D-G].
Try to keep the same separators between different kind of headers.
Change-Id: Id42cd7b3743a816f75cecb6b576c0e37cb057473
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
Reviewed-by: hjk <hjk@qt.io>
Move it to its own header.
Get rid of SshConnection and SshConnectionManager,
as they are not used anymore.
Change-Id: I52fe20d7816ea57e7a7158ab2ae9565d50a76e21
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
Remove also sftpdefs.h. They are not used anymore.
Change-Id: I6c45a70cec2e01afdd1a668068e090f5d4abde9b
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
... to the client interface. JsonRpcMessages are the only messages used
so far and no other types of messages are currently used by any of the
supported Language Servers. If a client is going to need special message
parsing it can still implement a specialized client interface and
overwrite parseCurrentMessage. This is the preparation to move receiving
and parsing data passed to and from the language server out of the GUI
thread.
Change-Id: Ibd4cd95daab7efff947273ca9e7d457de0286f47
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
FileTransferErrorHandling::Ignore wasn't used anywhere.
Change-Id: I5826fc9f7871b06cd8f423d7e3b1d01a5b6cbdff
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Currently, as a result of many refactorings, the QtcProcess
is designed so that if you want to customize the behavior
you should implement custom ProcessInterface subclass.
Adding virtual methods directly to QtcProcess caused tons of
unpredicted side effects and nasty bugs, which were hard to
track and hard to fix, as provided fixes were usually fixing
particular case while introducing regressions in not related
code paths.
Consider also aggregating QtcProcess object instead of
deriving from it when some additional methods are required.
This patch removes the last virtual methods from QtcProcess API
and makes the class final in order to prevent from adding
any new virtual methods to this class in the future.
This commit message should make it clear that having subclasses
of QtcProcess is not a desired design. It's a post-mortem
conclusion. So: don't derive from QtcProcess - we mean it!
Change-Id: I1e43ed45be326b366422fd7db6e05ba48ea5fb98
Reviewed-by: hjk <hjk@qt.io>
It's being replaced by QtcProcess with path on device.
Change-Id: I29eb038d1b17151683f86855eb547e47f7f7dea5
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: hjk <hjk@qt.io>
It's being replaced by QtcProcess with path on device.
Change-Id: I89abacd686e00e3f42f6ab90e9439f015d73ce80
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: hjk <hjk@qt.io>
qmljsscanner did interpret ?? and ?. as a single ? and thus as the
start of a ternary operator, breaking reformatting.
Fixes: QTCREATORBUG-27344
Change-Id: I0429d20aed0196743f1c20e6ceac11351d5a7a3b
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
Reviewed-by: <github-actions-qt-creator@cristianadam.eu>
