Bug#339829: potential patch improvements
Hi Paul!
On Wed, Jan 04, 2006 at 01:34:34PM +0800, Paul Wise wrote:
> On Tue, 2006-01-03 at 10:53 -0500, Justin Pryzby wrote:
>
> > I would rather see my original, more aggressive check:
> >
> > $description=~m/homepage/is && $description!~/^ Homepage: [^ ]*$/
>
> Hmm, this misses stuff where people use webpage on the last line, or a
> phrase without web or page in it, which is one thing I wanted to
> specifically detect, since I noticed it a few times.
Good point.
> > Packages for which this is a false-positive (such as slash, gnudip,
> > and bake)
>
> These can be eliminated by checking for a url in the description too.
It reduces some true positives also: abcmidi, achims-guestbook,
airsnort, alsa-base, anjuta, ant, apache, apg, ardour-doc, aspell-bg,
atlc, audacity, etc.
> > Will be orders of magnitudes more rare than packages which
> > are missed by the existing patch which are not missed by check above
> > (such as liblingoteach4, qgo, tcsh, maxdb-sqlcli, tik, mozart,
> > tagcoll, sleuthkit, tdl, xlogmaster, libsqlod75, wmpuzzle,
> > python2.3-dictclient, ...).
>
> How about the attached combination check - does my check and also does
> yours, with the changes that it checks a couple of other words, and
> checks for a url in the description too.
> $description =~ m/(homepage|webpage|website)/is
Good, please also add "|URL|upstream" (see asterisk-sounds-extra).
> && $description =~ m/[a-z]+:\/\//
This is to reduce false positives, right? I don't like it. If this
were an ' || ' condition, it would be great.
> && $last_line!~/^ Homepage: [^ ]*$/ ){
Shouldn't this be: !~/^ {2}Homepage: [^ ]*$/ ?
^^^
Or is the contents of this "description" variable the control field
without the first column of blanks? I guess that is probably it.
Thanks for the patch, again!
--
Clear skies,
Justin
Reply to: