Newsgroups: news.software.b
Path: utzoo!henry
From: henry@zoo.toronto.edu (Henry Spencer)
Subject: Re: Changes to CNEWS date parsing - what is now legal?
Message-ID: <1991Jun24.202026.16593@zoo.toronto.edu>
Date: Mon, 24 Jun 1991 20:20:26 GMT
References: <21JUN91230123@carat.arizona.edu> <1991Jun22.213542.2256@twinsun.com>
Organization: U of Toronto Zoology

In article <1991Jun22.213542.2256@twinsun.com> eggert@twinsun.com (Paul Eggert) writes:
>>So: how strictly is C-News going to hold me to RFC-822?  More
>>generally, "what are the new rules?"
>
>The _rules_ are the same as before: RFCs 1036 and 822 as amended by 1123.
>C News is still more permissive than the RFCs,
>but's obviously it's unwise to rely on this.

Exactly.  Any permissiveness beyond 822+1123 should be assumed to be a
historical accident which is subject to being fixed at any time.  I think
about the only looseness we currently allow is that our table of timezones
is bigger than 822's, and also a *single* unrecognized word in the timezone
slot is read as "GMT" on the assumption that it is the local timezone
abbreviation in Swahili or something.  About the only change that is likely
is that if -- as is reportedly likely -- the next revision of 1123 comes
out with a SHOULD almost-mandating use of numeric timezone plus a
parenthesized comment with local abbreviation, we will probably amend the
code to accept the comment (currently rejected because 1036 does not allow
822 comments in headers).
-- 
"We're thinking about upgrading from    | Henry Spencer @ U of Toronto Zoology
SunOS 4.1.1 to SunOS 3.5."              |  henry@zoo.toronto.edu  utzoo!henry
