Regression report stable-2015-05-15 vs. nightly-2015-06-12

brson · 2015-06-15T01:05:43.399Z

Nothing interesting here. The regressions I looked at were intermittent failures. This run reports a suspiciously low number of working crates (599), probably networking failures, sigh.

Still trying to get a report against beta done.

gist.github.com

https://gist.github.com/brson/2fbfc1c20d5a2ad9b8a7

gistfile1.md

# Regression report stable-2015-05-15 vs. nightly-2015-06-12

* From: stable-2015-05-15
* To: nightly-2015-06-12

## Coverage

* 2324 crates tested: 599 working / 1062 broken / 7 regressed / 594 fixed / 62 unknown.

## Regressions

This file has been truncated. show original

llogiq · 2015-06-15T06:46:21.691Z

One data point: I know that clippy was broken because we use some unstable rustc::middly::ty API that was changed from under us. No big deal, but I’ll have to look at rustc changes to find out how to regain that functionality

Perhaps other nightly-dependent crates also share the same fate?

oli-obk · 2015-06-15T10:27:56.642Z

all the type enum variants were changed to camel case. Maybe that’s what you are noticing?

llogiq · 2015-06-15T10:53:18.035Z

Yep. Edit: …and fixed.

bluss · 2015-06-16T12:58:20.259Z

Either you chose the wrong day to run this (crates.io had a bad day), or the crater run was itself part of the problem that bogged down crates.io. Maybe it should use a local crates.io mirror somehow, it would reduce the download spam.

bluss · 2015-06-17T00:04:06.342Z

Since I was hit by a regression from stable, a new run might be interesting.

I guess this case is really uncommon, so my crate may be the only one, and I’m working on updating it already.

system · 2019-03-25T08:24:42.774Z

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.