infrastructure: keep QTest's output visible when ctest runs on Windows (#9751)

#### Brief overview of PR changes/additions

- Appends `QT_ASSUME_STDERR_HAS_CONSOLE=1` to the `ENVIRONMENT` test
property of every registered test - 95 of 95 confirmed with `ctest
--show-only=json-v1` - leaving each test's existing `ASAN_OPTIONS`,
`QT_QPA_PLATFORM` and `ENVIRONMENT_MODIFICATION` untouched.
- One `cmake_language(DEFER CALL)` per directory that registers tests,
rather than the variable copy-pasted into a dozen strings, so a test
added later cannot miss it wherever in the file it lands.
- Drops the 8 `QT_FORCE_STDERR_LOGGING` entries from the four workflows.
`shouldLogToStderr()` is `forceStderrLogging() ||
stderrHasConsoleAttached()`, so the test property now covers what CI was
setting by hand, and every one of those steps runs nothing but `ctest`.

#### Motivation for adding to Mudlet

Qt on Windows diverts QTest's output to `OutputDebugString` unless it
believes stderr has a console attached, and an MSYS2 shell gives it
none, so a failing test reported an exit code with no `FAIL!` lines, no
compared values and no totals. Setting it as a test property fixes local
runs and CI from one place instead of two.

#### Other info (issues closed, discussion etc)

**Test case:** full Linux `ctest` suite 95/95 pass; `ctest
--show-only=json-v1` shows all 95 tests carrying the variable with no
other property changed; `ctest -V` shows it in the test process
environment. The Windows behaviour itself is not reproducible on Linux -
it rests on the reporter's 219 vs 12233 byte A/B and on
`QPlainTestLogger::outputMessage`, which only calls `OutputDebugStringA`
when `!QtPrivate::shouldLogToStderr()`.

Closes #9747

Assisted-by: Claude:claude-opus-5

.github/workflows/build-mudlet-pr.yml

@@ -417,7 +417,6 @@ jobs:
       run: ctest --output-on-failure
       env:
         QT_QPA_PLATFORM: offscreen
-        QT_FORCE_STDERR_LOGGING: 1
         # This runner has Qt's FFmpeg backend and can decode, so TMediaLoopTest must
         # demonstrate every media behaviour rather than skip any of them. Without a floor
         # somewhere, a lost codec or a changed default backend would turn the whole file
@@ -428,15 +427,11 @@ jobs:
       if: runner.os == 'macOS'
       working-directory: '${{runner.workspace}}/b/ninja'
       run: ctest --output-on-failure
-      env:
-        QT_FORCE_STDERR_LOGGING: 1
 
     # the full ctest run above already covers every functional-labelled test
     - name: Run QTest
       if: matrix.run_tests != 'true'
       run: ctest --test-dir ${{runner.workspace}}/b/ninja -L functional --output-on-failure
-      env:
-        QT_FORCE_STDERR_LOGGING: 1
 
     - name: add ssh-agent for release uploads
       if: (runner.os == 'Linux' || runner.os == 'macOS') && matrix.deploy == 'deploy' && startsWith(github.ref, 'refs/tags/Mudlet-')

.github/workflows/build-mudlet-win-pr.yml

@@ -139,7 +139,6 @@ jobs:
 
         ctest --test-dir $GITHUB_WORKSPACE/build-$MSYSTEM/test --output-on-failure
       env:
-        QT_FORCE_STDERR_LOGGING: 1
         # This runner has Qt's FFmpeg backend and can decode, so TMediaLoopTest must
         # demonstrate every media behaviour rather than skip any of them. Without a floor
         # somewhere, a lost codec or a changed default backend would turn the whole file

.github/workflows/build-mudlet-win.yml

@@ -115,8 +115,6 @@ jobs:
         export LUA_CPATH
 
         ctest --test-dir $GITHUB_WORKSPACE/build-$MSYSTEM/test --output-on-failure
-      env:
-        QT_FORCE_STDERR_LOGGING: 1
 
     - name: (Windows) Run Lua tests
       timeout-minutes: 2

.github/workflows/build-mudlet.yml

@@ -403,21 +403,16 @@ jobs:
       run: ctest --output-on-failure
       env:
         QT_QPA_PLATFORM: offscreen
-        QT_FORCE_STDERR_LOGGING: 1
 
     - name: (macOS) Run C++ tests
       if: runner.os == 'macOS'
       working-directory: '${{runner.workspace}}/b/ninja'
       run: ctest --output-on-failure
-      env:
-        QT_FORCE_STDERR_LOGGING: 1
 
     # the full ctest run above already covers every functional-labelled test
     - name: Run QTest
       if: matrix.run_tests != 'true'
       run: ctest --test-dir ${{runner.workspace}}/b/ninja -L functional --output-on-failure
-      env:
-        QT_FORCE_STDERR_LOGGING: 1
 
     - name: add ssh-agent for release uploads
       if: (runner.os == 'Linux' || runner.os == 'macOS') && matrix.deploy == 'deploy' && startsWith(github.ref, 'refs/tags/Mudlet-')

test/CMakeLists.txt

@@ -9,6 +9,19 @@ endif()
 
 find_package(Qt6 6.8.2 REQUIRED COMPONENTS Test Network Widgets)
 
+# On Windows Qt diverts QTest's stdout to OutputDebugString unless it believes
+# stderr has a console attached, which an MSYS2 shell or a CI runner does not give
+# it, so a failing test reports an exit code and nothing else (#9747). Deferred so
+# that tests registered further down the file are covered too; TESTS does not
+# descend into subdirectories, hence the second call in functional_tests.
+function(restore_windows_test_output)
+    get_property(registeredTests DIRECTORY PROPERTY TESTS)
+    foreach(testName ${registeredTests})
+        set_property(TEST ${testName} APPEND PROPERTY ENVIRONMENT "QT_ASSUME_STDERR_HAS_CONSOLE=1")
+    endforeach()
+endfunction()
+cmake_language(DEFER CALL restore_windows_test_output)
+
 set(UNIT_TESTS
     TEntityResolverTest
     TEntityHandlerTest

test/functional_tests/CMakeLists.txt

@@ -2,6 +2,9 @@ if(NOT WIN32)
     include(${CMAKE_SOURCE_DIR}/src/cmake/EnableSanitizers.cmake)
 endif()
 
+# Defined in the parent test/CMakeLists.txt; TESTS is per-directory, so call it here too.
+cmake_language(DEFER CALL restore_windows_test_output)
+
 set(FUNCTIONAL_TEST_SOURCES
     ConfigDirOverrideTest.cpp
     TelnetTextDisplayedTest.cpp