Uh oh!
There was an error while loading. Please reload this page.
[WIP] Implement RFC 1560. - #32213
Conversation
e333048 to
c759b3eCompareretep998
commented
Mar 12, 2016
Also if you could test it on Windows with |
4ebba6d to
5c3ab23Comparebors
commented
Mar 13, 2016
☔ The latest upstream changes (presumably #32112) made this pull request unmergeable. Please resolve the merge conflicts. |
5c3ab23 to
1e48cd1Comparejseyfried
commented
Mar 13, 2016
alexcrichton
commented
Mar 13, 2016
1e48cd1 to
6dbe2adComparebors
commented
Mar 13, 2016
☔ The latest upstream changes (presumably #32227) made this pull request unmergeable. Please resolve the merge conflicts. |
6dbe2ad to
349958cComparenrc
commented
Mar 14, 2016
Just a note that the proposed changes are a draft RFC that hasn't been PR'ed yet, we should wait for the RFC to be approved (and of course to be a PR first) before landing this. I would be curious to see a crater run in the meantime though. |
jseyfried
commented
Mar 14, 2016
I know; this is a ways off from being ready to land regardless. |
jseyfried
commented
Mar 14, 2016
@alexcrichton I had to rebase anyway -- this should be ready for a crater run now! |
alexcrichton
commented
Mar 14, 2016
r? @nikomatsakis (or @nrc, ... or others!) |
alexcrichton
commented
Mar 14, 2016
(also started a crater run) |
jseyfried
commented
Mar 14, 2016
Thanks! |
alexcrichton
commented
Mar 14, 2016
Oh dear, looks like quite a few regressions :( |
jonas-schievink
commented
Mar 14, 2016
Most of them seem to be caused by the use of |
jseyfried
commented
Mar 15, 2016
All of the root regressions appear to be due to a bug that I fixed in the above commit. I had generalized the last bullet point ("The visibility of a glob-imported name is the minimum visibility of the glob itself and the name in the imported module") to apply to single imports as well. This won't make a difference once the warning cycle from #31362 is complete, but until then a |
alexcrichton
commented
Mar 15, 2016
Scheduled another crater run |
alexcrichton
commented
Mar 15, 2016
Hurray, zero regressions! |
jseyfried
commented
Mar 16, 2016
Excellent. It also looks like these changes fix two crates that were broken by new names in upstream crates. |
bors
commented
Mar 17, 2016
☔ The latest upstream changes (presumably #32284) made this pull request unmergeable. Please resolve the merge conflicts. |
bors
commented
Apr 6, 2016
☔ The latest upstream changes (presumably #32767) made this pull request unmergeable. Please resolve the merge conflicts. |
9ae2401 to
698caddComparebors
commented
Apr 13, 2016
☔ The latest upstream changes (presumably #32814) made this pull request unmergeable. Please resolve the merge conflicts. |
4c2626e to
0d86604Compare0d86604 to
1402963Comparebors
commented
Apr 17, 2016
☔ The latest upstream changes (presumably #32875) made this pull request unmergeable. Please resolve the merge conflicts. |
alexcrichton
commented
Jul 19, 2016
@jseyfried this has been inactive for awhile now, should we close or try to rebase + land? |
@alexcrichton I'm planning on rebasing, finishing this up, and |
alexcrichton
commented
Jul 19, 2016
Ok, awesome! |
nikomatsakis
commented
Aug 12, 2016
@jseyfried going to close this PR due to inactivity. Feel free to re-open though. :) |
jseyfried
commented
Aug 22, 2016
I finished this in #35894. |
This prototypes rust-lang/rfcs#1560 (cc #35120).
More specifically,
pub use super::*is equivalent touseing the private names andpub useing the public names fromsuper(modulo shadowing). See this comment from resolve: Privacy rules for re-exports can be too restrictive #31783.This implementation still needs documentation, tests, and more precise diagnostics (most importantly, an error for when an ambiguous name is used), but other than that it is complete.