Yes, I made some changes a while back and have more planned.

If there are specific things that you’re missing, please let me know and I’ll make sure they get addressed.

I can elaborate better later but the test output when using match? does not open the diff tool when a mismatch occurs (like when using clojure.test)

Ok, I’ll have to check my notes. I remember looking at this, but IIRC it’s not always clear what the two sides of the diff should be when using match?

you mean which side is the expected data and which is the actual data?

If I remember correctly (not sure, it’s been a while), there aren’t even always two clear sides - I might be remembering that wrong though.

I will try to know more about this and get back to you!

Thanks, like I say I need to check my notes as well, I’m not 100% about this, but I remember it was tricky.

cc @U02G8N0EX44, but I think the biggest issue IIRC is that the error reporting doesn't show up in the REPL pane and only sometimes when mousing over the failed test this was https://github.com/nubank/matcher-combinators/issues/31 on matcher-combinators

Hey, just seeing this (was traveling). While I don't use Cursive and don't know exactly what doesn't work, a few of my colleagues do (cc @U0AD3JSHL), and I help maintain matcher-combinators, so let me know if there is a way I can help.

for context, this is how matcher-combinators extends clojure.test, using the suggested assert-expr multimethod interface: https://github.com/nubank/matcher-combinators/blob/beae388ac30d4079a68619aba36ff7e152b20e42/src/clj/matcher_combinators/clj_test.clj#L38-L71 (and also maybe relevant https://github.com/nubank/matcher-combinators/blob/beae388ac30d4079a68619aba36ff7e152b20e42/src/clj/matcher_combinators/clj_test.clj#L155-L157)

I don't think a diff window will be easy to integrate with the matcher-combinators output because for the actual of the clojure.test result, it provides the datastructure with mismatch information placed in the datastructure iteself

some clojure.test/cursive issues that might be related to also how matcher-combinators is handled by cursive: - https://clojurians.slack.com/archives/C0744GXCJ/p1557445213127100 - https://github.com/cursive-ide/cursive/issues/875 and also someone came up with tweaks for a better failure printing w/ cursive: https://clojurians.slack.com/archives/C0744GXCJ/p1666881823365809