On Fri, Nov 19, 2021 at 06:43:20AM +0100, Yadd wrote: > >> uscan 2.21.5 now fails when filenamemangle regex doesn't match upstream > >> filename. This avoids some conflicts with GitHub downloads (which names > >> all files "v<version>"): a previously downloaded file can be modified > >> when more than one download is done. > > > > I'm sorry but this is a bug, uscan must not crash.
It's not a crash: it's an error log that clearly says "filenamemangle
failed" (even if perl's die() write down the failure location)
And your watchfile is buggy: that kind of filenamemangle would have
never ever matched (since you provided a regex that matches on a whole
URI when the regexp is then used against a single filename), so in your
specific case was mostly a nop. Yadd is concentrating on the fact that
there might be cases that could lead to file overwrites, but here I'm
focusing on the fact that you have a uscan rule that is just wrong.
> > Here I've 485 sources to scan and uscan leaves 300 sources not
> > scanned.
Does that mean that you copied the same filenamemangle option across 300
packages without ever verifying that the options were correct?
--
regards,
Mattia Rizzolo
GPG Key: 66AE 2B4A FCCF 3F52 DA18 4D18 4B04 3FCD B944 4540 .''`.
More about me: https://mapreri.org : :' :
Launchpad user: https://launchpad.net/~mapreri `. `'`
Debian QA page: https://qa.debian.org/developer.php?login=mattia `-
signature.asc
Description: PGP signature
