On Friday 01 February 2008 14:00:52 [EMAIL PROTECTED] wrote: > I think that runs into the wish to not get into the internals of > svn/svk/git in doing these tests. Not sure I agree, but that was a comment > I saw somewhere. It does seem svk could have a "are you setup?" query or > not start setup up on a bare 'info' ... but there seems something wrong w/ > the Parrot::Revision doing 3 consecutive svn/git/svk trys at info and just > living w/ the dieing/missing exes as a brute-force sort of test.
Heck, we could have a pre-commit hook which updates a REVISION file in our repository with the canonical revision number for the repository, and avoid all of this fragile mucking around. That doesn't solve the problem where we have tests that check Subversion properties on files (and I don't even want to think about how Git handles those), but given the amount of time I've personally spent fixing those failing tests versus fixing real errors found by those tests, I'm not sure they're giving us anything but busy work either. Again, a pre-commit hook could set SVN metadata properties on added files correctly if they don't already have the appropriate properties, and that's if the repository doesn't already have a notion of default properties. ... or we could continue wasting lots of CPU cycles running redundant tests that people don't run before checkin because they're slow and not very useful, that cause false negative failures that novice Parroters don't understand, and that don't really measure code quality at all. -- c
