W3C home > Mailing lists > Public > www-validator@w3.org > April 2014

Re: [VE] [html5] spurious "must have child option" when all options in optgroups

From: Philip Taylor <P.Taylor@Rhul.Ac.Uk>
Date: Wed, 16 Apr 2014 15:12:22 +0100
Message-ID: <534E8FC6.5030303@Rhul.Ac.Uk>
To: David Dorward <david@dorward.me.uk>
CC: Ken Geis <kgeis@berkeley.edu>, www-validator@w3.org

David Dorward wrote:

 > I just created and validated this document (where all options are
 > children of opt groups) and the validator passed it.
 > Can you provide a reproducible test case to go with your bug report?
 > <!DOCTYPE html>
 > <title>test</title>
 > <select size=1>
 > <optgroup label=jkldf>
 >     <option>hello
 > </optgroup>
 > </select>

But your example, David, unless you deliberately over-ride the 
Validator's DOCTYPE sniffing, will check the document against
an editor's draft of the provisional HTML 5 specification;
Mr Geis, on the other hand, specifically cited the HTML 4.01
specification, so a valid comparison would require an HTML 4.01
compliant example with appropriate DOCTYPE.

Philip Taylor
On 15 Apr 2014, at 23:16, Ken Geis wrote:

>> I got the following error when validating my document. The HTML
>> 4.01 spec says that the select element must contain an option, but
>> it does not say that it must be a child of the select element. The
>> validator throws the message when all options are inside optgroups,
>> and I think that should not be an error.
Received on Wednesday, 16 April 2014 14:13:03 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 14:18:10 UTC