- From: Rick Bogart <rick@sun.Stanford.EDU>
- Date: Thu, 20 Sep 2007 10:08:16 -0700
- To: Rui del-Negro <w3validator@dvd-hq.info>
- Cc: www-validator@w3.org
> >I have seen several messages go through on this topic, and it appears > >that the "CHARSET=" tag is not supposed to be case-sensitive. It is > >particularly odd because I can change the case of the other tag values > >and everything is fine: > > > ><META HTTP-EQUIV="CONTENT-TYPE" CONTENT="TEXT/HTML; charset=ISO-8859-1"> > > > >validates with no problem; only the term "charset" is case-sensitive! > >Is this going to be fixed, or do I have to change all my META tags to > >get their pages to validate? Thanks, > > "charset" isn't a tag. The "tag" there is <meta>. "text/html; > charset=iso-8859-1" is all part of the *value* of the attribute "content". > In other words, there is no reason why it should behave the same way or be > subject to the same rules as the tags (or attribute names). > > As a general rule, if something isn't explicitly defined as being > case-insensitive, one should assume it is case-sensitive. > > Having said that, I believe section 3.7 of RFC2616 states that "parameter > attribute names are case-insensitive". And, according to that same section > of that same RFC, that's exactly what "charset" is, in that context: the > name of a parameter attribute. > > http://www.faqs.org/rfcs/rfc2616.html > > Am I missing something? No, you are not missing something. I apologize for using the wrong terminology. But if we are agreed that "charset", as the name of a parameter attribute, should be case-insensitive, are we not then also agreed that the validator should not treat "charset" differently from "CHARSET", as it now does?
Received on Thursday, 20 September 2007 17:08:26 UTC