sure.  you're fix is simplier.

but my final code is simplier.  ☺ and it doesn't have the prefix code, which 
seems
sure to be a bug.

i think it's important to remove the prefix code (i didn't actually remove it.
i if(0)'d it.)  that was key to &pid being interpreted as &pi.  and, 
unfortunately,
to prevent side-effects from removing the prefix code, ti thought it was 
important
to make sure that any unexpected character in an entity reference terminated 
the name.  i'm guessing that was the original reason for the prefix code.

- erik

p.s. your png bug is interesting.  it doesn't happen on p9p. (the image formats
are handled differently.)  it doesn't happen with png -v.  somehow there is an
endian problem RGB → BGR.

p.p.s. when the w3c "strongly suggests" that usually means that all the major 
browsers
do things in defiance of the suggestion. ;-)

On Sat Jul  8 13:13:02 CDT 2006, [EMAIL PROTECTED] wrote:

> ok, this was my time to be wrong, I had to remove /386/lib/libhtml.a
> and then rebuild it in order to get it working.
> 
> I still prefer my way :)
> 
> http://www.w3.org/TR/html401/charset.html#entities :
> "
> Note.
> In SGML, it is possible to eliminate the final ";" after a character 
> reference in some cases (e.g., at a line break or immediately
> before a tag). In other circumstances it may not be eliminated (e.g., in the 
> middle of a word). We strongly suggest
> using the ";" in all cases to avoid problems with user agents that require 
> this character to be present. "
> 
> Federico G. Benavento

Reply via email to