Stream: t-compiler/wg-nll

Topic: nikos-prs


nikomatsakis (Jun 16 2018 at 21:07, on Zulip):

Hey @pnkfelix, question for you:

So I have these two PRs:

#51538 has a flaw, but I think I have the fix locally. My question is, would you rather just look at #51460 directly, even though it's long? It also has some "extra commits" that did refactorings I didn't always wind up using later — I could plausibly pull those out, but I'm not sure I want to invest the time (and I do expect to use the refactorings later on)

pnkfelix (Jun 18 2018 at 05:04, on Zulip):

I already looked at the prefix (and commented on it)

pnkfelix (Jun 18 2018 at 05:08, on Zulip):

In general I prefer to attempt to land such factored-out prefixes on their own, in order to aid future bisection attempts. But in this case, I’ll review whichever PR you want; I don’t expect people to be doing much bisection of this portion of NLL’s dev history

nikomatsakis (Jun 18 2018 at 13:02, on Zulip):

one complication is that — in fixing the problems with #51538 — I encountered some further problems that would be hard to fix without some of the refactorings I did later

nikomatsakis (Jun 18 2018 at 13:03, on Zulip):

but I'm leaning now towards "land the immediate fix" and then land the other stuff as a FIXME for later

nikomatsakis (Jun 18 2018 at 13:12, on Zulip):

bah, in writing up my comment, I realize that this change could theoretically affect working code... but you'd have to work pretty hard

nikomatsakis (Jun 18 2018 at 21:12, on Zulip):

@pnkfelix you still around? I'm still contemplating the best way to fix this

nikomatsakis (Jun 18 2018 at 21:12, on Zulip):

if we do manage to sync tomorrow, it might be helpful for me to try and catch you up on on all the madness i've been doing here

Last update: Nov 21 2019 at 14:45UTC