I've always found the 50 or so executables that Open Babel currently creates as part of the test build to be annoying. They slow the build, hog screen space on the MSVC IDE, and yet are still not fine-grained enough to indicate which test inside the executable actually failed.
CMake provides a test runner for just this purpose. If you take a look at my combinetests branch (https://github.com/baoilleach/openbabel/tree/combinetests) I have modified two of the tests (automorphism and builder) to suit this framework. These are combined into a single executable test_runner and when you run ctest you get: Test project C:/Tools/openbabel/mySmilesValence/windows-vc2008/build Start 1: test_automorphism_1 1/53 Test #1: test_automorphism_1 .............. Passed 0.10 sec Start 2: test_automorphism_2 2/53 Test #2: test_automorphism_2 .............. Passed 0.10 sec Start 3: test_automorphism_3 3/53 Test #3: test_automorphism_3 .............. Passed 0.09 sec Start 4: test_automorphism_4 4/53 Test #4: test_automorphism_4 .............. Passed 0.08 sec Start 5: test_automorphism_5 5/53 Test #5: test_automorphism_5 .............. Passed 0.11 sec Start 6: test_automorphism_6 6/53 Test #6: test_automorphism_6 .............. Passed 0.09 sec Start 7: test_automorphism_7 7/53 Test #7: test_automorphism_7 .............. Passed 0.10 sec Start 8: test_automorphism_8 8/53 Test #8: test_automorphism_8 .............. Passed 0.09 sec Start 9: test_automorphism_9 9/53 Test #9: test_automorphism_9 .............. Passed 0.09 sec Start 10: test_automorphism_10 10/53 Test #10: test_automorphism_10 ............. Passed 0.08 sec Start 11: test_builder_1 11/53 Test #11: test_builder_1 ................... Passed 14.49 sec Start 12: test_builder_2 12/53 Test #12: test_builder_2 ................... Passed 1.12 sec Start 13: test_builder_3 13/53 Test #13: test_builder_3 ................... Passed 0.73 sec Start 14: test_builder_4 14/53 Test #14: test_builder_4 ................... Passed 0.54 sec Start 15: test_builder_5 15/53 Test #15: test_builder_5 ................... Passed 0.52 sec If you just want to run a particular test, you can still run the tests individually at the commandline (or in a debug session) with "test_runner buildertest 5" (this is exactly how the test suite runs the tests). So there you have it: more finely grained test failures, a single executable, and still the ability to run a single test. What's not to like? I'm happy to modify the tests to suit the proposed framework, but it's somewhat tedious so now would be a good time to say whether you think this is a bad idea. Regards, Noel ------------------------------------------------------------------------------ AlienVault Unified Security Management (USM) platform delivers complete security visibility with the essential security capabilities. Easily and efficiently configure, manage, and operate all of your security controls from a single console and one unified framework. Download a free trial. http://p.sf.net/sfu/alienvault_d2d _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
