Re: trn and yEnc

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Aug 20, 2003 at 11:54:31AM -0700, Wayne Davison wrote:
> On Tue, Aug 19, 2003 at 08:29:13PM -0400, William Denton wrote:
> > What's the best way to get trn decoding [yEnc] files?
> 
> Trn doesn't support yEnc directly because it was a poorly designed
> format and I was kinda hoping it would go away (however, that doesn't
> look to be very likely these days).

Poorly designed compared to what?
- uuencoding doesn't contain crc checksums and/or "end of article" marker 
  and has to blindly rely on the Subject: line to stitch multi-part binaries 
  back together in the right order.
- MIME/base64 I'd describe as a fscking mess where multi-part binaries
  are concerned.   It looks like about every author of a newsreader
  has read a _different_ MIME spec...   Fortunately multi-part MIME/base64
  seems to have disappeared off the face of usenet... :)

I'm not saying yEnc is good [1], but IMHO it's the least bad of the 3...

[1] good would be a redesigned usenet where encoding and splitting
of binaries would no longer be needed and where binaries would only
be cached to/'in the direction of' where they are actually being
downloaded instead of to all newsservers.  Usenet is totally out of controll
at 600+Gbyte/day... 
 
> Since you have an external yEnc decoder installed, you could try one of
> these commands and see if it works on the selected articles:
> 
>     :e|ydecode
>     :s|ydecode

That will only work if ydecode has a 'dumb' "blindly decode and append an
 article" mode...
If you've got a 4-article multi-part binary, the above does:
ydecode article1
ydecode article2
ydecode article3
ydecode article4

instead of ydecode article1 article2 article3 article4

IMHO, that part of trn isn't very compatible with external multi-part
binary decoders, which greatly prefer the latter...

> Personally, I've switched over to using "ubh" for extracting binaries --
> since that's it's sole purpose in life, it does a good job of it.  I
> haven't tried the 3.x version (it puts the article data into a mySQL
> database), but the 2.x version works well for me (grab the ubh script
> from the web-browsable CVS to get the latest fixes).  I combine it with
> uudeview (a fast decoder) for extra speed.
> 
>     http://ubh.sourceforge.net/
> 
> One of its advantages over trn is that it caches the news headers, so if
> the group is large, there is less delay "re-entering" the group.

I use a mix of trn4 + later decoding with uudeview and
'nget -K' plus uudeview, but that's partly because I want to retain some info
from the article headers for later processing and partly because
uudeview gets confused by some yEnc files if the "-s" option isn't used
and nget uses uudeview's uulib for decoding, resulting in me not trusting
uulib to function 100% OK on yEnc files... ;) 


Marco van Loon (slightly paranoid w.r.t. these things...)


-------------------------------------------------------
This SF.net email is sponsored by: VM Ware
With VMware you can run multiple operating systems on a single machine.
WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines
at the same time. Free trial click here:http://www.vmware.com/wl/offer/358/0

[Index of Archives]     [Photo]     [Yosemite]     [Epson Inkjet]     [Mhonarc]     [Nntpcache]

  Powered by Linux