-
Notifications
You must be signed in to change notification settings - Fork 3.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix IEx tests on OTP27: fix assertions #13352
Conversation
You can fall back to reading specs via Code.Typespec |
This would basically imply reverting #12803 and go back to generating it ourselves rather than relying on |
Btw, OTP 27 has markdown docs. So maybe we should use our own rendering structure for Erlang if the content type is markdown? |
Another alternative is to wait for erlang/otp#8077 to be merged, then you can continue to use shell_docs for rendering erlang docs. |
Good point. So maybe it is best to wait until the next release candidate is out and then we can assess all of our options more throughly? @sabiwara |
Sounds good! |
I think this will be the last blocker to add OTP27 to the CI, once the other PRs are merged. |
@sabiwara we can also stop asserting for specs on tests and add an issue to introduce the specs back once 27 is out (and such an option exists). Your call. :) |
Hi @garazdawi,
I was trying the master branch after 8077 has been merged, but it seems that shell_docs are not displaying specs: I wondered if specs are planned to be added back to |
Yes, it is planned though it won’t make it to 27.0 as a bunch of other things popped up that needed to be done. We would be happy to accept a pr if you feel like helping out! |
I would love to! Will give it a shot. |
Opening this PR for discussion.
We migrated IEx introspection to use
:shell_docs
in #12803, but OTP27 stops returning specs, which would be a regression for us.Before (type)
After (type)
(The
opaque
information got lost as well, not great)Before (spec)
After (spec)
It might be fine given we're aiming at replacing typespecs with set-theoretic types, but I wanted to confirm.
Or should we open an issue upstream?