Re: djvulibre exception request
Per req, pushed up v -3 in the old source format.
Might help a little, but there are still a lot of false positive
changes listed in the diff due to differential CVS keyword expansion.
This is the horror of CVS keywords I guess.
They're in the upstream tarball, so I can't get rid of the issue by
just not expanding them. It would, at the very least, require some
significantly more complex gyrations that wouldn't play nicely with
either CVS or git.
(Maybe the "right" solution is for debdiff to have an
--ignore-CVS-keywords-in-comments option? If implemented in a general
way, it might be useful for other purposes as well. Perhaps it could
generate two diffs: one of just the changes to be ignored, and the
other of the non-ignored changes. That way it would be
straightforward to scan for false negatives in its what-to-ignore
test. But I digress.)
--Barak.
Reply to: