On Tuesday 25 October 2011, 12:55:47, Manuel M T Chakravarty wrote: > I am all for it. I hate tabs in Haskell programs.
I support that motion. > Manuel > > Max Bolingbroke: > > If we are going to make whitespace changes, we should probably have a > > check to ensure that tabs don't get added back in by later commits. > > > > I've created a pre-receive hook that verifies the following property: > > Taken *as a whole*, the series of commits you are trying to push.. > > ..for all file *modified* (i.e. I'm ignoring renames) by the > > commits.. ..that do not contain tabs *before* the push.. > > ..your commits do not add a *new* line containing a tab > > > > Your push is rejected with a list of all violations if this property > > is violated. At this point you can either write a new patch that fixes > > the validation problems, or just rebase to edit the commit introducing > > the problem. > > > > I've also written a pre-commit hook that GHC developers could copy > > into their own git repos to ensure that such bad commits never get > > created in the first place. > > > > Is this something we want to check? Should we use this pre-receive > > hook on darcs.haskell.org? > > > > Max > > _______________________________________________ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
