On Thu, Jun 07, 2007 at 03:41:59PM -0700, Stephen Lau wrote: > Mike Kupfer wrote: > >>>>>>"Matt" == Matt Mackall <mpm at selenic.com> writes: > > > >Matt> It'd be helpful to prioritize these. Some of them are easy, some > >Matt> are hard, some are of questionable merit. > > > >We want this one very soon... well, *I* want this very soon, because the > >issue looks like it will complicate our internal beta testing. > > > >> 222: [extensions] in a repo's .hg/hgrc fi... > > > > I just saw Matt paste the following in #mercurial: > > (15:18:03) lisppaste5: mpm pasted "enable extensions in .hg/hgrc" at > http://paste.lisp.org/display/42429 ... > So it looks like that one is coming soon :) Nice work Matt!
Unfortunately, my original 10-line fix (which didn't actually work!) turned into 18 separate cleanup patches. The original problem was that you can't parse the command line until you know what the command set is. And you don't know that until you've loaded the extensions. And you don't know what extensions to load until you've read all the hgrc files. And you don't know which hgrc file to read until you've parsed the command line. And there were a few other circular issues to deal with too. But the result is that the whole command processing system is a bit cleaner and moved into a separate file from the commands themselves. And, of course, you can now specify extensions in .hg/hgrc files. I'll give my thoughts on the rest of the bug list in a bit. I'm going to be at Google for the rest of the week at the Linux Collaboration Summit. -- Mathematics is the supreme nostalgia of our time.
