[leafnode-list] Re: Requesting "legal opinion" on Newsgroups & Xref headers

[Matthias Andree]

Am 21.11.2009, 02:52 Uhr, schrieb John R. Myers <jmyers at n6wuz.net>:

> On 11/20/2009 at 11:18 PM Whiskers wrote:
>
>> On Fri, 20 Nov 2009 12:00:34 -0800 "John R. Myers" <jmyers at n6wuz.net>
>> wrote:
>>
>>> There is a situation with fetchnews where the "Information Assurance
>>> FAQ" does not get cross-referenced to <news.answers>. A little
>>> investigation shows that the header contains the following line:
>>> Newsgroups:...,news.answers^M
>>
>> I can't find any postings of a FAQ by that name in news.answers - or on
>> Google Groups Advanced Search (not that that is reliable) or at
>> <http://www.faqs.org/faqs/>.  What newsgroups should it be found in?
>>
>
> Sorry, that should be the "Information Research FAQ" posted in:
>    alt.internet.research,sci.research,alt.answers,sci.answers
> together with news.answers^M. Thanks for the cross check.
>
>> If others can verify the presence of an errant ^M, then a polite email  
>> to
>> the maintainer of the FAQ would seem to be in order.
>
> If someone can verify the ^M I suppose a related question would be:
> Does anyone know offhand whether ^M is a legal line delimiter for NNTP?

Greetings,

Some thoughts on this:

- the line delimiter in articles in the network/protocols would be CRLF  
for header and body

- leafnode is fit to work with CRLF

- Could someone verify if the article in question has CRCRLF instead, for  
instance, by telnetting into their upstream server and obtaining a hexdump  
of the article in question? (script or tee to a file and then running xxd  
on the output might also work).

   If that is the case, we could make fetchnews strip CR* LF sequences  
(that is any amount of CR right before LF) from lines read without too  
much of an effort.

HTH

-- 
Matthias Andree