[Pan-users] Re: pre-downloading articles

[Duncan]

Brad Rogers <address@hidden> posted
address@hidden, excerpted below, on  Fri,
04 May 2007 08:04:01 +0100:

> On Fri, 4 May 2007 06:53:33 +0000 (UTC) Thufir
> <address@hidden> wrote:
> 
> Hello Thufir,
> 
>> >> headers, but the body and attachments?
>> > Currently, no.  If you absolutely must have that feature, you'll have
>> > to install "old-style" Pan 0.14.x
>> Oh, and it's *so* close to supplanting leafnode for me :(
> 
> Even with an always on connection, I prefer to d/l the message bodies at
> the same time as the headers.  I'm fairly sure it's coming, but IIRC,
> it's on the post 1.x release to-do list.

Agreed, it's on the active todo list, but for post-1.0.

What I do for binaries (for text I download on demand) is download 
overviews (what everybody calls headers, even tho that's not really 
correct), go thru and delete what I know I don't want right away or with 
just a sample download or two, then set everything else to download to 
cache (which I've set by hand in preferences.xml to 12 gigs, so pan 
doesn't start erasing stuff before I've had a chance to look at it!), 
while I go off and do something else, visit the text groups or sort thru 
a previously downloaded group or whatever.  Then when everything's in 
cache, I come back and decide what I want to keep and save it off 
accordingly, deleting what I don't want to keep, as well as stuff I've 
already saved.

That works well here, but using a local news server such as leafnode is 
the other alternative, used by many.

The auto-download feature is currently slated to be added along with 
other automatic actions for scored articles.  The options will probably 
include auto-delete and auto-mark-read checkboxes for each of ignored and 
negative scored articles, and auto-download checkboxes for each of zero 
scored (normal, no score applies, or +/- scores cancel out), medium 
(1-4999), high (5000-9998), and watched (9999+) scored articles.  This 
was it seems the main thing old-pan's rules were used for, but setting it 
up was rather more complex than Charles would have liked, and I agree, 
the suggested new interface checkboxes sound simpler.  I'd just be nice 
to actually have them, but I suppose as they say, patience is a virtue, 
and I do understand the need to draw a line and limit new features so 
what's there can be stabilized.  Meanwhile, one of the above solutions, 
either using the two-stage download to cache and process from there, or 
using leafnode or another external but local news server, should work.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman