On 18 Feb 2008 02:14, Roman Kyrylych wrote:
> ...
> Then we can forget about this corner case, it can be worked
> around by cache cleanup scripts based on filename parsing
> (available on BBS and ML).
Along these lines this could be useful for someone searching
through the list archives. It just removes old duplicates. I
need it to clean up my local custom repo when I move $PKGDEST
contents into it before rebuilding the repo database.
This construct has so far caught every package name I've thrown
at it...