Enhance switch on non-exhaustive enums - #24381
Conversation
c97a209 to
057d8f6
Compare
|
I will try to rebase on top of writergate and look into the CI failures until tomorrow. |
057d8f6 to
34abbdf
Compare
There was a problem hiding this comment.
Amazing work wrangling Sema's switch logic; it's borderline impenetrable. Just one small nitpick about your behavior test; once you fix that and rebase, I'll be happy to hit merge.
By the way, as mentioned on another PR of yours, these lines in test cases are redundant:
// backend=stage2
// target=nativeIt's be good if you could drop them from your newly-added cases. Thanks!
Mainly affects ZIR representation of switch_block[_ref] and special prong (detection) logic for switch. Adds a new SpecialProng tag 'absorbing_under' that allows specifying additional explicit tags in a '_' prong which are respected when checking that every value is handled during semantic analysis but are not transformed into AIR and instead 'absorbed' by the '_' branch.
… statements If both are used, 'else' handles named members and '_' handles unnamed members. In this case the 'else' prong will be unrolled to an explicit case containing all remaining named values.
34abbdf to
0ecbd5a
Compare
|
Thank you! |
andrewrk
added
the
release notes
|
Great work, thank you! |
| else => {}, | ||
| _ => return error.TestFailed, |
This comment was marked as resolved.
This comment was marked as resolved.
Sorry, something went wrong.
There was a problem hiding this comment.
It's completely order independent, just like you could already put else or _ at any position. The only limitation is that the else prong cannot have any additional explicit cases specified, this isn't a new limitation though.
alexrp
removed
the
release notes
5 participants
Closes #19438
Closes #12250
More details in commit messages
The commits are mostly separate but the second one is still based on the first one. If only one gets accepted I am happy to rebase accordingly.