Re: W3C html validation issue

Handling is incorrect for all browsers :-) if a browser reads a htmlfile (how comes he knows is of no importance, DTD, guess whatever) it ALLWAYS knows about &amp; being entity representaion for character '&' if in any HTML/XML/SGML file a browser reads the character '&' an entity begins.... therefore reading the line "<a href='foo.php?bar=1&amp;myvar=2'>" the browser gets the url "foo.php?bar=1&myvar=2" so when clicking on the url you should read "foo.php?bar=1&myvar=2" and not ""foo.php?bar=1&amp;myvar=2" so do you write to all browser vendors that missunderstood HTML? regards marcus At 00:50 23.07.2002, Walter A. Boring IV wrote:

Howdy, This may be a tad off topic, but it is related to how php deals with building the $_GET superglobal w/ query string vars. According to the W3C HTML validator, it is illegal to build a query string for a url such as foo.php?bar=1&myvar=2

             ^^^^^^^^

They say you HAVE to use foo.php?bar=1&amp;myvar=2

             ^^^^^^^^^^^^

in php's case it builds $_GET["bar"] = 1; $_GET["amp;myvar"] = 2; I don't know of ANYONE on the planet that builds query string vars in this way. Unless you do, your html pages will not validate through the W3C validator. The official W3c spec explanation is at http://www.htmlhelp.com/tools/validator/problems.html#amp I think changing this could potentially break many sites. Does anyone know what other web languages do with &amp;myvar=2 ? I think the W3C validator is broken/wrong in this respect, and wanted to know what other folks think about it. Walt -- PHP Development Mailing List <http://www.php.net/> To unsubscribe, visit: http://www.php.net/unsub.php