We discussed the current regressions in compile-time errors, as well as some older cases. Mostly these have to do with attempting to recover from earlier errors, leading to erroneous follow-on reports. We’ll be making some efforts to tamp these down until we can make the overall error recovery mechanism more reliable.
Seeing as many of us were on vacation, we decided to hold a special meeting this week discussing possible designs for int generics, along with special guest @eddyb. The goal is to design a “max min” proposal that can have us making progress on the basic issues while skirting some of the more complex topics (notably: how far to go in determining constant equality.) Expect to see some concrete write-ups soon(ish).
This week we reached decisions on a number of PRs affecting
In addition, we are moving several RFCs into final comment period:
- FCP PR #1432: String/Vec::splice(RangeArgument, IntoIterator)
- FCP PR #1434: ‘Contains’ method for ranges
- FCP PR #1479: Unix socket support in the standard library
- FCP PR #1498: Add octet-oriented interface to std::net::Ipv6Addr
Next week we will be choosing APIs to consider for stabilization/deprecation in 1.9. Please take a look at the list and leave comments on any APIs you think should be considered!