| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Reviewed-by: Trust Me
(cherry picked from commit ac5c099cc3c5b8c7eec7a49fdeb8a21037230350)
|
|
|
|
|
| |
This test is too slow on our macs too.
It's now disabled by default everywhere but Linux.
|
|
|
|
|
|
| |
nmake and checktest are too slow on Windows for this test to
be enabled by default. Set RUN_SLOW_TESTS=1 to run the test
manually.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When a test runs successfully, it outputs an XML test log, with the
testcase name coming from whatever class is passed to QTest::qExec.
If a test crashes, hangs, fails to compile, or otherwise fails to output
a valid test log, any summary of the test results still has to display
some reasonable testcase name. But we can't tell what the testcase name
_would_ have been had the test run correctly - without, for example,
attempting to parse the C++.
The simplest solution is to make sure that TARGET and the test class
name are always matching. This test will force people to follow this
naming convention.
|
| |
|
|
|
|
|
|
|
| |
Having the tests in separate projects allows for some optimization
strategies when running the tests (e.g. start running corelib tests
while the rest of Qt is still compiling), and allows developers to run
only a subset of tests when appropriate.
|
| |
|
|
|
|
| |
The generated `tests/auto/tmp' directory was causing this test to fail.
|
|
|
|
| |
Reviewed-by: Trust Me
|
|
This selftest enforces that tests/auto/auto.pro is properly maintained.
It may be extended to check other elements of the environment necessary
for autotests to run correctly.
|