- User Since
- Jun 9 2014, 9:34 AM (146 w, 1 d)
Let's rebase and commit asap -- fixes a good bug
Thu, Mar 23
Exactly right! A few commenting/refactoring suggestions but you have nailed the cause
Fri, Mar 17
Thu, Mar 16
In general I'm ok with this though it seems a bit heavyweight.
Tue, Mar 14
Fri, Mar 10
Fine modulo comments
Wed, Mar 8
Great! Richard could you look?
I'm ok with deferring beyond 8.2 (which is ready to go), but I do think we should commit this after the fork, for the reasons Eric says. It'd be worth checking why allocations do sometimes go down -- but more inlining is a plausible cause.
Tue, Mar 7
I'd prefer to always to the "float-in-the-desugarer" (rather than leave a hack in the code) and open a ticket for the unexpected bump in cases like T1969. It's very much a corner case, but it needs a ticket of its own.
Mon, Mar 6
Wed, Mar 1
If you look at the rules that change here, you'll see, for example, that we currently have
More comments; but otherwise looks good to me.
and, or, all, any, elem, and notElem, at least.
See Trac Trac #11715 comment:67. Richard has agreed to have a go.
Looks good. Obviously check perf carefully
Looks great thanks!
Tue, Feb 28
Can we land this?
This is great. But could be better.
do you recall under what circumstances you saw uniques creep into the SPEC rule names? I'm not seeing the issue with this test.
Let's finish this! It's not a big deal!
Mon, Feb 27
some final coments
I've added comments. You are spot on; just needs careful documentation.
I'm pretty sure that D3218 works only because the bad cases don't arise in the testsuite.
Feb 24 2017
Great, modulo the Note
I've committed a patch to HEAD that makes unpackCString CONLIKE (and documents why). And improves exprIsConApp_maybe.
As I say on the ticket, I would prefer not to make catch and catchException distinct, unless we need to. The difference is very subtle, and we got it wrong here.
Looks like a good improvement. What's "not quite correct"?
Feb 22 2017
Eric I'm keen to nail this one. It's dragged on for too long.
Feb 21 2017
Feb 17 2017
Feb 16 2017
I have an upcoming patch for this, as I mentioned
Feb 10 2017
Great -- go for it.
I'm OK with this. Let's add a Note as I suggest above, and land it
Feb 9 2017
OK well, we probably get better error messages if we report the errors from simplifyDeriv. So I've done that -- see the patch I've put in email to you. Works for all of tests/deriving, except that two error message improve
you should also delete
OK I see.
By all means add the ASSERT (with an explanation). Then good.
Feb 8 2017
Is there a way to grab the insoluble constraints in approximateWC?
It seems that... works