Re: Getting rid of -XImpredicativeTypes

Shot in the dark: Would extensions like QuantifiedConstraints or ImplicationConstraints, if implemented, help with ImpredicativeTypes? 2016-09-30 15:29 GMT+00:00 Simon Peyton Jones via ghc-devs < ghc-devs@haskell.org>: > Alejandro: excellent point. I mis-spoke before. In my proposal we WILL >

Re: How, precisely, can we improve?

On 2016-09-30 19:26, Carter Schonwald wrote: > We all do!. I dont. (Sorry, just had to put that Monty Python joke in there. At least I *think* it was MP?) Obviously, yes, we all *really* *REALLY* like Haskell, otherwise we wouldn't be arguing about it, would we? ;) Regards,

RE: [Diffusion] [Build Failed] rGHC3f27237b0e6d: Make tcrun042 fail

Ben Gamari writes: > [ Unknown signature status ] > Simon Peyton Jones via ghc-devs writes: > >> rn003 is failing apparently (see below) >> But it’s fine on my machine. I don’t know how to see the actual error >> message. >> > It works for me as

RE: [Diffusion] [Build Failed] rGHC3f27237b0e6d: Make tcrun042 fail

Simon Peyton Jones via ghc-devs writes: > rn003 is failing apparently (see below) > But it’s fine on my machine. I don’t know how to see the actual error > message. > It works for me as well. The Harbormaster failure is quite odd: tc042 appears to fail with, Compile

Re: How, precisely, can we improve?

We all do!. And I really appreciate your clear positive suggestions around "what are we trying to understand or ask our selves" no matter what conclusions we come to this day / month / year, its something we will need to revisit (and should) every once in a while. change happens, as does

Re: [Diffusion] [Build Failed] rGHC3f27237b0e6d: Make tcrun042 fail

It works for me. There are two logs because one corresponds to stdout and the other to stderr. I'm not sure if that distinction is present on Windows. It so happens that the test runner uses stdout, so that's the interesting one. Searching for the test name reveals: Timeout happened...killed

RE: [Diffusion] [Build Failed] rGHC3f27237b0e6d: Make tcrun042 fail

Thanks Chris, that’s helpful. I got to your final link, but it didn’t show me what rn003 did. There seem to be two logs for some reason. I tried clicking “Unlimited” lines, but I still didn’t see a clear failure from rn003. Does it fail for anyone else? From: Christopher Allen

Re: [Diffusion] [Build Failed] rGHC3f27237b0e6d: Make tcrun042 fail

It's a bit tricky, you have to follow these steps: Click the commit link in the email: https://phabricator.haskell.org/rGHC3f27237b0e6d Click the build link on the commit page: B11120: rGHC3f27237b0e6d: Make tcrun042 fail ! Click another build link on

RE: [Diffusion] [Build Failed] rGHC3f27237b0e6d: Make tcrun042 fail

rn003 is failing apparently (see below) But it’s fine on my machine. I don’t know how to see the actual error message. Any ideas? Simon From: nore...@phabricator.haskell.org [mailto:nore...@phabricator.haskell.org] Sent: 30 September 2016 17:31 To: Simon Peyton Jones

Re: git question

You can unstage a change using 'git reset', e.g. 'git reset utils/haddock'. As part of an interactive rebase, my approach would be to 'edit' the commit in question, then use 'git reset HEAD^' to undo that commit but keep the changes in your working directory (but unstaged), then either stage the

RE: git question

Ah yes, but it had a lot of OTHER stuff that I wanted. It was only the accidental change to a submodule that I wanted to kill off. From: Harendra Kumar [mailto:harendra.ku...@gmail.com] Sent: 30 September 2016 12:20 To: Simon Peyton Jones Cc: Edward Z. Yang

Re: How, precisely, can we improve?

Dear Moritz, Oh, thank you for teaching me. I'll write to it. Thanks, Takenobu 2016-09-30 20:06 GMT+09:00 Moritz Angermann : > Dear Takenobu, > > may I politely direct you to https://github.com/ghc- > proposals/ghc-proposals/pull/10, > and ask you to add your comments

Re: How, precisely, can we improve?

Hi, Richard said: (1) search engines still find out-of-date documentation (2) the wiki is not discoverable I know trac is treasure house. And I realized old pages are valuable for decision. My concrete suggestion is here: For (1): When we find out-of-date documentation, we directly modify

git question

Friends I have a batch of several commits lined up on my local machine, validated, and ready to push to the master. But one seems to have accidentally got diff --git a/utils/haddock b/utils/haddock index 073d899..a3309e7 16 --- a/utils/haddock +++ b/utils/haddock @@ -1 +1 @@ -Subproject