W3C home > Mailing lists > Public > www-validator@w3.org > November 2002

Re: detailed.html page broken in CVS

From: Terje Bless <link@pobox.com>
Date: Sat, 2 Nov 2002 17:42:09 +0100
To: W3C Validator <www-validator@w3.org>
cc: Frederic Schutz <schutz@mathgen.ch>
Message-ID: <a01060005-1021-07C49AB9EE8211D6B10300039300CF5C@[193.157.66.10]>

Frederic Schutz <schutz@mathgen.ch> wrote:

>I have a problem with the detailed.html page when I use the CVS version
>of the validator: my choices of encoding or type are not transmitted to
>the validator. The culprit seems to be the code that transforms a POST
>request into a GET, but does not copy the doctype or charset
>information. [...] I attach below a possible patch, against the current
>CVS version.

Thanks! I applied it to CVS it will be in the next update.


>It seems to work, but I was just wondering if some more escaping could
>be necessary (eg what happens if the charset variable contains a ";") ?

Nah. A semicolon isn't an allowed character in this parameter (i.e. it
shouldn't appear there) and the failure mode here is simply that it doesn't
set the charset. We can live with that until get around to making the fixes
to the escaping code that Björn has been helping us with recently.


>The code duplication is already in the original script (any simple way
>to fix that ?).

Sure. I just moved the lot of it into a sub-routine. There's one or two
more places where we can use it.

-- 
These are the same customers you are referring to whom Microsoft thought
would need MS Bob and the Talking Paperclip?   One thing is to give them
enough rope to hang themselves,  but a boobytrapped thermonuclear weapon
running on a rand(time) countdown... Is that really wise? - Me to MS rep.
Received on Saturday, 2 November 2002 11:42:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:04 GMT