Re: Parsing HREFs?

On Mon, 20 Feb 2006, Frank Ellermann wrote:

>> In Jukka's example, the equals sign cannot occur in the
>> reference, so REFC will be implied before it.
>
> Fascinating.  So my "historically interesting browser" is
> actually trying to do the right thing in this case, and it
> was unfortunately a "near miss".

Not really, if you are referring to interpreting &sect in
&section=... as an entity reference for the section sign.
By SGML rules, a conforming browser recognizes &section as an entity 
reference, then checks for any definition for such an entity.

> As often with SGML I'm _very_ happy about my decision to
> start with XHTML 1.0 instead of HTML 4.  Thanks for info.

In this particular detail, using XHTML has actually a benefit: a validator 
will detect any attempt to use an entity reference without a trailing 
semicolon, and this helps to avoid some problems with current browsers.

-- 
Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/

Received on Monday, 20 February 2006 06:34:50 UTC