On Tue, Dec 12, 2006 at 10:02:42PM +0100, ann wrote: > On Tue, 12 Dec 2006, strk wrote: > > >Oops, I'm afraid I'll need to review this. > >Our "test output parser" doesn't parse 'PASS'/'FAIL' but > >rather 'PASSED', 'FAILED'. > >We're actually just "mimicing" Dejangu... Rob can you also > >help some here ? > > Not that I'm against you changing/rolling back my commit, but if > you want to mimic Dejagnu, wouldn't it make more sense to use 'PASS' > and 'FAIL'? Or is there a compelling reason to differ from Dejagnu > in this regard? (You can simply say 'yes', you don't need to explain > why.)
IIRC Rob initially told me to use PASSED/FAILED as output. I'm not a Dejagnu user, just started it's use with Gnash, so dunno details. What I can tell is that the script that parses PASSED/FAILED prints PASS/FAIL back ... that's why I asked Rob for help :) A possibility is that you used the wrong section of Dejagnu manual, listing labels output from Dejagnu, rather then labels expected by Dejagnu in input. But I don't really know what I'm talking about. What I'm sure about is that the "Test Runners" of gnash will need to print (X)FAILED/(X)PASSED or our simple.exp parser script will not understand. You can also try verifying this by following the instructions in the manual (you can use any programming language for a test runner, just have it output some labels and modify the Makefile.am so that it runs your testrunner). --strk; _______________________________________________ Gnash-commit mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnash-commit
