On Thu, 23 Oct 2003, Guy Harris wrote:
> On Oct 23, 2003, at 7:37 AM, Jason House wrote:
>
> > Does anyone know how to convert the following information into a
> > usable URL? Trying to repend http://www.ethereal.com ends up sending
> > me to http://www.ethereal.com/lists/ instead of viewing the
> > attachment...
> >
> > A non-text attachment was scrubbed...
> > Name: print_dlg.jpg
> > Type: image/jpeg
> > Size: 29949 bytes
> > Desc: not available
> > Url : /pipermail/attachments/20031023/f3e1fba6/print_dlg.jpg
>
> Given that the only usable URL I know of offhand for that image appears
> to be
>
> http://www.ethereal.com/lists/ethereal-dev/200310/jpg00002.jpg
>
> (from searching for Ulf's message in the archives, and doing a "view
> source" on it), I'm not sure how it'd be converted to a usable URL.
>
> Gerald? Is this some issue with the digesting software? (I assume
> Jason gets the digests, given the subject line in his message.)
Mailman's digest module splits out attachments and saves them on the web
server by default. For some reason the URLs it generates are missing
the list name. The URL Jason _should_ have received is
http://www.ethereal.com/pipermail/ethereal-dev/attachments/20031023/f3e1fba6/print_dlg.jpg
The second URL (/lists/...) was generated by MHonArc, which is what's
being used to create the archives on the web site. Its URLs are (of
course!) different than the ones Mailman generates.
I disabled Mailman's digester-gets-to-fool-with-attachments option.
Digest users _should_ recevie attachments unmolested now, just like
everyone else. This bug database entry doesn't inspire confidence,
however:
https://sourceforge.net/tracker/index.php?func=detail&aid=688751&group_id=103&atid=100103
