Hi. I've implemented CPPUNIT_TEST_XFAIL to add test cases to a suite, which are expected to fail. See
https://gerrit.libreoffice.org/#/c/19213/ https://gerrit.libreoffice.org/#/c/19214/ https://gerrit.libreoffice.org/#/c/19215/ The current approach unfortunately breaks the ABI and API. The API break can be prevented, but I wanted to get some review of the approach. One minor problem I'm struggling with is the accounting of the test in src/cppunit/TestResultCollector.cpp and the resulting summary output of src/cppunit/CompilerOutputter.cpp and src/cppunit/TextOutputter.cpp So errors are a part of the failures, but for XFAIL a pass is actually a failure. Currently XFAIL pass and failure are both handled as errors. << "Run: " << m_result->runTests() << " " << "Failure total: " << m_result->testFailuresTotal() << " " << "Failures: " << m_result->testFailures() << " " << "Errors: " << m_result->testErrors() << " " << "Expected: " << m_result->testFailuresExpected() << " " << "Unexpected passed: " << m_result->testUnexpectedPasses() accounted as if ( failure.isError() ) ++m_testErrors; if ( failure.isExpected() ) { ++m_expectedFailures; if ( failure.isUnexpectedPass() ) ++m_unexpectedPasses; } Suggestions welcome. Thanks for reviewing. Jan-Marek _______________________________________________ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
