TMI: Tying up loose ends
Oct. 12th, 2012 08:54 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
timToday:
* Got my massive pull request for issue 2633 (removing last-use analysis) ready to submit. Yes, I said "pull request", even though normally I'd be committing this directly. This week we decided to switch to a code-review model, even for core Rust contributors. So all of us will have to file a pull request and request a code review for any non-trivial change (we're being trusted as to deciding what's "trivial"). I think that's a good thing -- should foster more of us understanding more of the code in the compiler.
* Worked on issue 3521 and related. I had a minor "aha!" of realizing that there was already a convenient place in resolve to attach the check I needed to attach (that constants don't refer to names that aren't constant, such as local vars or arguments): in upvarify. A bit of resolve hacking later, I got the test case that tests that a constant declaration referring to a local variable is rejected working. Still working on getting discriminant exprs in enums to be treated the same way.
* Got my massive pull request for issue 2633 (removing last-use analysis) ready to submit. Yes, I said "pull request", even though normally I'd be committing this directly. This week we decided to switch to a code-review model, even for core Rust contributors. So all of us will have to file a pull request and request a code review for any non-trivial change (we're being trusted as to deciding what's "trivial"). I think that's a good thing -- should foster more of us understanding more of the code in the compiler.
* Worked on issue 3521 and related. I had a minor "aha!" of realizing that there was already a convenient place in resolve to attach the check I needed to attach (that constants don't refer to names that aren't constant, such as local vars or arguments): in upvarify. A bit of resolve hacking later, I got the test case that tests that a constant declaration referring to a local variable is rejected working. Still working on getting discriminant exprs in enums to be treated the same way.