[PATCH 0 of 7] Persistent tag caching

Greg Ward greg at gerg.ca
Fri Jul 3 09:48:24 CDT 2009


At long last, my patches for persistent tag caching are vaguely
approaching a state that might possibly be considered appropriate for
a rough preliminary review.  Maybe.  ;-)  So here they are.

Things to keep in mind:

* I'm not entirely happy with patch #1, which factors _findtags() out
  of tags().  The idea is sound, but the design is still a little
  wonky.  You'll see when you read the comments.  Suggestions
  welcome.

* All the intermediate patches (cleanup, rename variables, factor out
  methods, etc.) are IMHO pretty good.  If you spot problems there,
  please let me know, because that means I screwed up an easy
  refactoring!

* The final patch is, of course, the one that matters.  It's *mostly*
  done, but I just discovered that qpop breaks it.  (Sigh.)  And
  it spews lots of debug output, which trivially breaks (at least) one
  test.  And there's some duplicate-feeling code in tagcache.readtags()
  that I need to fix.  You get the idea.  But the broad strokes are all
  there, and now would be a good time to spot dumb assumptions, deep
  design errors, etc.

Greg


More information about the Mercurial-devel mailing list