W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2009

Re: Meaning of invalid but well-formed dates

From: Julian Reschke <julian.reschke@gmx.de>
Date: Sun, 03 May 2009 19:30:24 +0200
Message-ID: <49FDD4B0.3040607@gmx.de>
To: Geoffrey Sneddon <foolistbar@googlemail.com>
CC: HTTP Working Group <ietf-http-wg@w3.org>
Geoffrey Sneddon wrote:
> Hi,
> 
> "Wed, 32 March 2009 00:00:00 GMT" matches HTTP-date, and as far as I can 
> tell is completely valid to produce and send. However, what it actually 
> means is completely undefined.

Nope. It may be valid according to the ABNF grammar, but that doesn't 
make March, 32 a valid date.

> Equally, "Mon, 03 May 2009 00:00:00 GMT" matches HTTP-date, but the 
> wkday and and date1 disagree. This too is valid, but what it means is 
> undefined.

See above.

> Actually, I can't find anything that even defines what "Sun, 03 May 2009 
> 00:00:00 GMT" a valid and logically correct date means.

I think it's totally obvious what it means. The day is "03", the month 
is "May", and the year is "2009". What am I missing here?

Maybe you could clarify what you're looking for? Do you have any 
evidence of implementations interpreting valid dates incorrectly?

BR, Julian
Received on Sunday, 3 May 2009 17:31:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:51:02 GMT