Maxim Cournoyer <ma...@guixotic.coop> writes: > Hi Rob, > > Rob Browning <r...@defaultvalue.org> writes: > >> Rob Browning <r...@defaultvalue.org> writes: >> >>> So for now I've attempted to just allow us to mix and match SRFI-64 and >>> (test-suite lib) based tests by integrating SRFI-64 into the existing >>> automake/check-guile arrangement via a suitable test driver. >> >> I've pushed this preliminary support to main along with SRFI-197. So if >> you add a file that uses SRFI-64 and ends with .sr64 to >> test-suite/tests/ and then add it to SCM_TESTS, it should be fully >> incorporated into "make check", and should also be available via say >> ./check-guile srfi-197.sr64. > > I've never heard of the .sr64 file extension; was there a reason to not > stick to .scm? sr64 probably requires a prop variable to have Emacs > understand the source as Scheme? > > Otherwise, it sounds good! Thanks for your efforts!
I would assume the reason here is Automake. --8<---------------cut here---------------start------------->8--- -TEST_EXTENSIONS = .scm .test +TEST_EXTENSIONS = .scm .sr64 .test --8<---------------cut here---------------end--------------->8--- The dispatch to different test runners is done via the extension, using the `<EXTENSION>_LOG_RUNNER' (`SR64_LOG_DRIVER' here). While single runner for .scm that would dispatched based on some marker in the file *could* be used, this is probably way less magic. Tomas -- There are only two hard things in Computer Science: cache invalidation, naming things and off-by-one errors.
