httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Pieter Nagel <>
Subject Re: bogus html? </p> without <p>
Date Mon, 07 Jul 1997 19:10:54 GMT
On Wed, 2 Jul 1997, Alexei Kosut wrote:

> You get the same effect by changing </p> to <p>.  It's a bogus tag in my
> opinion.  I'm surprised that it's only my files that do it, I swear I saw
> it elsewhere. 

As far as I know, an HTML parser is *supposed* to be able to infer
misssing tags if their ommision is unambiguous - this is why pages
with the <HTML> or <HEAD> or <BODY> tags ommited still work.

I think this type of behaviour is specced at the SGML level and is
therefore not readily apparent from reading the HTML spec alone.

Precisely *what* "unambiguous" means, I don't know. I suppose if you
read the SGML DTD grammar for HTML, it'll boil down to infering the
missing <P> tag at the end of the previous paragraph, </H1>, </H2>
etc. - there are boundaries a <P>...</P> pair can't cross.

     /_)              /| /
    /   i e t e r    / |/ a g e l

    PGP: finger 

View raw message