a285bcba26
QT_SOURCE_TREE is a variable that is set in qtbase/.qmake.conf. In qtbase, it's used throughout various projects to find cpp sources when building standalone tests (among other things). Everything works fine with qmake, because even if qmake is invoked on the tests subfolder, qmake searches up the source directory tree until it finds a .qmake.conf file, and uses that. When building qttools with qmake, the qdoc project expects to have a QT_SOURCE_TREE value, but it's not actually set in the qttools/.qmake.conf file, so the generated include paths that use that value are incorrect. Curiously the build still succeeds. Now in CMake land we replaced QT_SOURCE_TREE with CMAKE_SOURCE_DIR, but that does not work properly when doing a standalone tests build, because the project in that case is the tests one, and not the qtbase one, so configuration fails in a developer build when trying to configure some private tests. So far I've found that only qtbase actively uses this value. A temporary fix is to save the qtbase source directory into a QT_SOURCE_TREE variable inside the generated BuildInternalsExtra.cmake file. The pro2cmake script is changed to handle presence of QT_SOURCE_TREE in a qrc file path. This is handled by finding the location of a .qmake.conf file starting from the project file absolute path. This is needed to stop the script from crashing when handling the mimedatabase test projects for example. The change also regenerates the relevant failing test projects, and thus standalone tests (when doing developer builds aka private_tests enabled) now configure and build successfully. Change-Id: I15adc6f4ab6e3056c43ed850196204e2229c4d98 Reviewed-by: Simon Hausmann <simon.hausmann@qt.io> |
||
|---|---|---|
| .. | ||
| auto | ||
| baselineserver | ||
| benchmarks | ||
| global | ||
| libfuzzer | ||
| manual | ||
| shared | ||
| testserver | ||
| CMakeLists.txt | ||
| README | ||
| tests.pro | ||
This directory contains autotests and benchmarks based on Qt Test. In order
to run the autotests reliably, you need to configure a desktop to match the
test environment that these tests are written for.
Linux X11:
* The user must be logged in to an active desktop; you can't run the
autotests without a valid DISPLAY that allows X11 connections.
* The tests are run against a KDE3 or KDE4 desktop.
* Window manager uses "click to focus", and not "focus follows mouse". Many
tests move the mouse cursor around and expect this to not affect focus
and activation.
* Disable "click to activate", i.e., when a window is opened, the window
manager should automatically activate it (give it input focus) and not
wait for the user to click the window.