- From: Terje Bless <link@pobox.com>
- Date: Mon, 6 Aug 2001 05:52:37 +0200
- To: "Jukka K. Korpela" <jkorpela@cs.tut.fi>
- cc: www validator mailing list <www-validator@w3.org>
[ I don't know where the origial context was, but feel free to forward ] On 03.08.01 at 04:49, Jukka K. Korpela <jkorpela@cs.tut.fi> wrote: >So apparently the validator does not recognize the information in the ><meta> tag when there is a line break. This is very strange since a line >break should be equivalent to a space. I hope this is just because the >validator processes such <meta> tags in some ad hoc way, since the >alternative would be that it has some flaw in its fundamental parsing >routines. Yes, this is a result of the ad hockery that plagues META on all sides. We use a Perl regex to pick out the META tag from the document and the regex is a tradeoff between "getting it right" and "this is a kludge so lets not worry about it too much". The alternative was to reimplement a stupid SGML parser in Perl regex or do two stage parsing and we went the easy road. It's been somewhat alleviated in recent development versions -- we do two stage parsing in that we preparse it with a lightweight parser before handing it off to the real parser -- and are working on a "real" fix.
Received on Monday, 6 August 2001 01:19:06 UTC