Re: [Pan-devel] Generating NZB files

[Konrad Karl]

Hi Charles,

On Fri, Feb 02, 2007 at 01:48:58PM -0600, Charles Kerr wrote:
> Darren Albers wrote:
> >On Thu, 01 Feb 2007 00:22:36 +0100, Konrad Karl wrote:
> >>Even better: create an option to make an nzb out of selected
> >>articles, ignoring the cached status.
> >>
> >
> >This actually wouldn't be that hard to put together since Pan has all the
> >basics available to do this.  I /think/ what you would want to do is
> >something very similar to do_download_selected_Article () and instead
> >of adding it to the queue call a function similar to nzb_to_xml to write
> >out the tasks. There are some additions you would need to make such as
> >adding a dialog box to get the path for saving the file etc....

I thought the same but are not at the necessary level in gtk programming to
quickly hack something together.

> >
> >I haven't looked deeply into this part of pan to know for sure but I think
> >that should be all you want to do, Charles can step in and smack me if I
> >missed something  ;-)
> 
> Duncan, you've got it just right. :)  That's how I'd go about it.
> 
> Generating nzb files for other programs raises a bigger question
> for me though.  If another program is more useful at handling nzb,
> then Pan generating nzb only treats a symptom, not the problem.
> The real solution is to make Pan handle nzb better than anyone else.

In my situation it is just very useful to throw the .nzb onto 
something which (typically) runs on a remote screen session on some
machine accessed over ssh (low bandwidth between office and
"download machine"). Thus I can even avoid the traffic which the
progress meters of nzbperl are generating by simply detaching from
screen.

Another advantage of nzbperl is that it just needs a text console
and very little resources (a few perl modules). Pan needs the
whole gtk stuff which is not strictly necessary for the download
process itself. 

Being able to limit bandwidth is also quite nice.

> Konrad, what does nzbperl.pl do better than Pan?

Pan by itself downloads quite nicely but see my special
situation above.

And please think also about the bytes/lines ratio filtering
(else I will have to think about a perl script which operates
on the group file.... quite a bit of work to make it right and
in Pan itself it will be most likely just a couple of lines
of code)

Greetings,
Konrad 

> 
> Duncan, what did you have in mind?
> 
> cheers,
> Charles
> 
> 
> _______________________________________________
> Pan-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/pan-devel