[pmwiki-users] url parms with ampersands and w3c validation

[Chris Cox]

Patrick R. Michaud wrote:
> On Wed, Oct 19, 2005 at 01:28:09AM -0500, Chris Cox wrote:
>>According to w3c, you can't have raw apersands in URLs.
> 
> One cannot have raw ampersands in an *HTML document*, because the 
> ampersand is used to denote character entities.  Thus within an 
> HTML document, one has to use "&" instead of "&", even in 
> things like src="..." and href="...".  HTML authors often forget
> about this when creating HTML.
> 
> It's perfectly valid to have raw ampersands in urls; in fact, the
> CGI specification explicitly says to use "&" to separate parameters.
> 
>>So I figured something like
>>
>>[[whatever?blah=1&foo=2]]
>>
>>would work.  But the & becomes a regular ampersand inside
>>of the link in html and fails validation.
> 
> In markup, this should be
> 
>     [[whatever?blah=1&foo=2]]
> 
> i.e., the actual url as it would appear in the address bar of a browser.  
> PmWiki already takes care of converting ampersands to "&" as 
> needed in the HTML output.
> 

I noticed that PmCalendar pages won't pass the w3c validator
though ... and it seems to indicate that it's a problem with
ampersands in the urls.

Curious.