triage: I-nominated
triage: P-low
We discussed this in the @rust-lang/lang meeting some today. General conclusion was that if we can in fact detect a cycle, that is probably OK. It’s worth pointing out there is some code in trait matching that aims to detect similar cycles – it does have some problems around regions though – but also that we can probably ignore regions for the purposes of establishing these cycles. This may take a bit of investigation. However, basically if we DO detect cycles the proper way (not merely by exceeding the recursion limit), then it seems ok to permit cutting off the method call candidate search there.
Classifying as low priority because while this use case would be nice to support, it’s not urgent.
triage: P-medium
Looks like this is already fixed by RFC 1214, as @arielb1 reports.
triage: P-high
Permalink to this digest
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
