Not that I am blaming people for breaking ghc head (since I did it too
recently with other Cabal patches) but I note that this particular bug
was committed by someone who does use validate on a regular basis. :-)
In this case, running validate would not have helped since the compile
error was conditional on #ifdef stuff that triggers on OSX (which I'm
fixing now...).
should i bring up my old alternative suggestion again?-)
in brief, i asked for a process by which anything that might
cause trouble is checked into the main repo in such a way
that only the buildbots use it (eg. via conditional compilation).
the process would be:
- developer X ensures build on at least one platform
- X checks in partially tested code under labelled guard
- Y user builds ignore all guarded code, suffering no instabilities
- buildbots build with guarded code, and include test labels in
build reports
- X fixes bugs as they are reported by buildbots
- when buildbot for platform P reports successful build,
X removes guard for that platform, until all guards for
X's code are removed
that way, the main repo holds code in two stages: guarded code
has been tested on one platform, but hasn't passed the buildbots
yet. unguarded code has passed all buildbots, and builds on all
platforms. also, the buildbots would take care of automated
multi-platform testing, without endangering the build for other
users.
claus
_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc
