Bug 686 – [html4/xhtml1] head/@profile holds a white-space separated list of URIs

NOTE: The current preferred location for bug reports is the GitHub issue tracker.

Bug 686 - [html4/xhtml1] head/@profile holds a white-space separated list of URIs


Summary:	[html4/xhtml1] head/@profile holds a white-space separated list of URIs

Status:	RESOLVED FIXED

Product:	Validator.nu
Classification:	Unclassified
Component:	XHTML 1.0 schema
Version:	HEAD
Hardware:	All All

Importance:	P2 normal
Assigned To:	Michael[tm] Smith

URL:

Depends on:
Blocks:
	Show dependency tree / graph

Reported:	2009-11-27 16:59 CET by Julian Reschke
Modified:	2009-12-01 16:14 CET (History)
CC List:	1 user (show)

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Julian Reschke 2009-11-27 16:59:57 CET

head/@profile takes a list of URIs, not a single URI.

Example for incorrect error message:

"Error: Bad value http://www.w3.org/2006/03/hcard http://dublincore.org/documents/2008/08/04/dc-html/ for attribute profile on element head: WHITESPACE in PATH.

From line 4, column 17; to line 4, column 116

lang="en"><head profile="http://www.w3.org/2006/03/hcard http://dublincore.org/documents/2008/08/04/dc-html/">↩

Syntax of IRI reference:
    Any URL. For example: /hello, #canvas, or http://example.org/. Spaces should be escaped as %20. "

Comment 1 Michael[tm] Smith 2009-12-01 02:19:13 CET

OK, I see that the HTML4 DTD has this:

http://www.w3.org/TR/html4/struct/global.html#edef-HEAD

<!ATTLIST HEAD
  %i18n;                               -- lang, dir --
  profile     %URI;          #IMPLIED  -- named dictionary of meta info --
  >

and this:

  Attribute definitions
  profile = uri

... so I can see why somebody might assume that "%URI" and "uri" meant to be a single URI.

But that's followed by spec prose that says, "This attribute specifies the location of one or more meta data profiles, separated by white space. For future extensions, user agents should consider the value to be a list even though this specification only considers the first URI to be significant...".

So, I guess that wins.

It would have helped a lot of they had instead used "url-list" or something.

I'll take a look at changing this now.

Comment 2 Michael[tm] Smith 2009-12-01 02:37:14 CET

validator r335
http://whattf.browse.cvsdude.com/validator?view=rev&revision=335

Change is live on http://qa-dev.w3.org:8888/ so you can test there.

Comment 3 Julian Reschke 2009-12-01 09:35:21 CET

Note that Manu's "profile" spec contains an erratum for HTML 4.01. I think the HTML WG should just consider to approve that formally, so we avoid this kind of confusion in the future.

Comment 4 Julian Reschke 2009-12-01 16:14:28 CET

(In reply to comment #2)
> validator r335
> http://whattf.browse.cvsdude.com/validator?view=rev&revision=335
> 
> Change is live on http://qa-dev.w3.org:8888/ so you can test there.

Looks good to me, see, for instance http://qa-dev.w3.org:8888/?doc=http%3A%2F%2Fgreenbytes.de%2Ftech%2Fwebdav%2Frfc2291.html&parser=html5