In theory I have nothing against this. In practice though, I see a few problems: * as @codebrainz pointed out, with the current code classic unit tests will likely not be able to do test much, as we have way too many state everywhere, and that reproducing it basically means starting Geany. * I have nothing against Criterion, apart that it's apparently not currently in any distribution, which means it'll have to be manually installed, and thus 1. we can't hard-depend on it 2. very few people will actually run the tests which is kind of sad for testing purposes.
So IMO it's not a great idea to add an "uncommon" test framework, no matter how great it is, if we won't have much test to add to it and won't have much people running them. So I'd rather add tests using something we already have available (maybe GLib's thing) even if it's slightly less fun -- at least until we actually have a strong use-case, as again I don't expect much useful tests to be runnable that way. Just my ยข2 -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/geany/geany/pull/1427#issuecomment-287548353
