Re: Some wrong HTML code validates correctly.

Le lun 10/02/2003 à 23:56, Grzegorz Adam Hankiewicz a écrit :
> Hopefully this is the right mailing list, in the feedback page it
> said "If your page doesn't validate...", but the problem is that
> the page should NOT validate.

This is the right mailing list, yes.

> The whole discussion can be read here:
> 
>   http://groups.google.com/groups?hl=es&lr=&ie=UTF-8&threadm=mailman.1044810540.18789.python-list%40python.org&rnum=1&prev=/groups%3Fq%3Dgrzegorz%2Badam%2Bhankiewicz%2Bhtmlparser%2Bbug%26hl%3Des%26lr%3D%26ie%3DUTF-8%26sa%3DN%26tab%3Dwg
> 
> Basically, the following is an incorrect validation:
> 
>   http://validator.w3.org/check?uri=http://www.terra.es/personal7/gradha/test.html

(note that this page is not available anymore apparently)
 
> According to http://www.w3.org/TR/1998/REC-html40-19980424, 3.2.2,
> attributes have to be separated by spaces. They aren't, yet the
> page is considered valid. Is this a bug in the validator?

Yes, it is. The validator relies on OpenSP to parse XML, and as stated
in the results of validation, OpenSP supports has some limitations:
http://openjade.sourceforge.net/doc/xml.htm
"OpenSP does not enforce the following XML constraints: 
[...]
# XML does not allow a parameter separator that is adjacent to a
delimiter to be omitted."

Which is the case in your example.

Dom
-- 
Dominique Hazaël-Massieux - http://www.w3.org/People/Dom/
W3C/ERCIM
mailto:dom@w3.org