On Fri, Jul 08, 2011 at 12:45:52AM +0900, Stephen J. Turnbull wrote: > Gabriel Kerneis writes: > > but is it really worth the trouble of having IO striking back right > > in the middle of a so beautiful piece of pure code? > > I sincerely hope you are not serious. Darcs is a version control > system used in production, not a PhD thesis.
I was indeed kidding about the purity part. But the point remains that locales and characters encodings are a real PITA, that you have no way to know which one people are using reliably, or to make any decision at all in some cases (for instance, if several commit messages contain characters in various encodings, the resulting "darcs send" message will be a meaningless pile of bits) and that the only reasonable way to sort all this out is to rely on people using utf-8 everywhere. Yes, I firmly believe that users not using utf-8 should be left on their own. But I'll send a follow-up patch as you suggested, to be nice with that part of mankind and warn them that they are using a technology of the past and should be burnt or something like that. Because that's how life is, I guess, and this seems to be the only way to get this patch accepted. Or maybe I'll do as a friend of mine suggested and submit a patch that crashes darcs on start-up if the current locale is not ascii or utf-8, to count the number of utf-8-reluctant darcs users out there ;-) </rant> Best, -- Gabriel _______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
