This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
Reporterd by Ernest Unrau http://lists.w3.org/Archives/Public/www-validator/2007Aug/0053.html OK: <META HTTP-EQUIV="content-type" CONTENT="text/html; charset=ISO-8859-1"> NOT Recognised: <META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=ISO-8859-1"> TODO: * check HTTP specification for case-sensitivity (I believe there is nothing mentioned precisely, but HTTP is case-sensitive) * make the validator's detection less strict * throw a warning if loose detection caught improper construct
*** Bug 4953 has been marked as a duplicate of this bug. ***
This is true not only for charsets specified via the <meta/> element but as well for those specified within HTTP headers. (Content-Type: text/html; charset=utf-8 is fine, while Content-Type: text/html; Charset=utf-8 is not.) Hope this isnt too obvious to be pointed out here.
I reported a bug to LWP, arguing that this module, responsible for the parsing of headers, should make sure it normalizes the case: http://rt.cpan.org/Public/Bug/Display.html?id=28713 The author of HTML::Encoding, which is used by the validator, actually thinks this is not LWP's responsibility and ack'd the bug report: http://lists.w3.org/Archives/Public/www-validator/2007Aug/0068.html http://search.cpan.org/dist/HTML-Encoding/
a new version of the HTML::Encoding library fixes this issue: http://lists.w3.org/Archives/Public/www-validator/2007Dec/0006.html tested on the dev server, I am updating the production service with the new library and closing this bug.