Niklas Larsson
Niklas Larsson
Oh, of course the `**`syntax is required to be followed by `*.`. That is completely asinine. That means we can't use that to grab `Makefile`, `expected` and `run` files.
The paths are superfluous, but harmless in that there are paths on the command leading to the correct libdirs afterwards. I think they're there because both the lib paths from...
Yes, the thing that is causing the 1.1.1 failure is a trailing -I followed by nothing, so the filename gets misinterpreted. I haven’t managed to reproduce that when building. I...
So now we are back where we started with a huge list of files :-) If that's the only way to keep it working on all cabal versions I guess...
I guess we should put in 8.6 too, but one thing at a time
I saw that Travis has Xenial now. So let's test that.
Thanks for the PR! I think that's the right decision, we should keep snprintf for the platforms that supports it. I'll get around to reviewing it soon.
This might have been caused by the updates in the cabal file to work with newer cabal-install, it certainly wasn't intentional.
Probably the fix is to use `pprintDelabTy` instead of `pprintDelab` on line 517 in Idris/REPL.hs like it is on line 1421 (where the CLIs version of browse is).
Hmmm, doesn't that change the meaning of it, even if the end result is the same?