It seems I've got some spurious _darcs/patches/pending files,
leading to conflict warnings on pull. Those pending files might
also be related to the earlier hang in darcs pull, according to
darcs issue911. Whenever I've seen such file appear out of
nowhere in the past, there was a pattern to their contents, also
noted in issue911 <http://bugs.darcs.net/issue911>:
move A B
addfile A
rmfile B
Same pattern in these current cases as well, together with
the equally odd:
addfile A
hunk A
rmfile A
namely:
$ grep '^[^-+]' _darcs/patches/pending
{
move ./libraries/cabal-bin.hs ./libraries/cabal.hs
addfile ./libraries/cabal-bin.hs
hunk ./libraries/cabal.hs 1
rmfile ./libraries/cabal.hs
addfile ./rts/sm/Evac.c-inc
hunk ./rts/sm/Evac.c-inc 1
rmfile ./rts/sm/Evac.c-inc
addfile ./rts/sm/Scav.c-inc
hunk ./rts/sm/Scav.c-inc 1
rmfile ./rts/sm/Scav.c-inc
}
$ grep '^[^-+]' libraries/Cabal/_darcs/patches/pending
{
move ./doc/Cabal.css ./doc/fptools.css
addfile ./doc/Cabal.css
hunk ./doc/fptools.css 1
rmfile ./doc/fptools.css
}
It would seem that there is something about moving/renaming
files that tends to cause darcs hickups (is it because darcs
keeps track of what a file used to be called before a move?).
Removing the spurious pending changes gets rid of the
conflict warnings in pull, just as darcs revert did get rid
of the hang.
Just wanted to record this here,
Claus
_______________________________________________
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
