Re: Parsing HTML using SGMLS and HTML.DTD

R J Partington ([email protected])
Tue, 15 Aug 1995 00:05:24 +0100 (BST)


Joe English wrote...
> R J Partington <[email protected]> wrote:
>
> > I've got a problem trying to parse HTML using the `html.dtd'
> > (comes with GF, written by Daniel W. Connolly :) using sgmls.
>
> [ Aside: Dan Connolly didn't write GF; Gary Houston did. ]

Oops. The 'written by DWC' bit was meant to be referring to the
HTML.DTD. Guess I wasn't clear enough. :|

> You need to supply the SGML declaration on the command line:
>
> sgmls html.decl yourfile.html

Guess who figured this out about 5 minutes after mailing the message?
:)

> Don't do that :-)
>
> It will choke most Web browsers, and it really isn't necessary.
> You can keep (a single copy of) the SGML declaration in a
> separate file.

Thanks.

As a seperate thread, what do people think about Netscapism's?
I normally use lynx as my WWW browser (I use Linux, and it's not
usually worth my while to go into X with 8MB of memory), and some
``Netscape enhanced'' pages look _absolutely_ terrible.

The best one I've seen (for terrible-ness) was a page where
the first letter of each word was enlarged using <font size+1>
(or whatever the correct one is).

For a commercial web-site, ie one that provides services etc,
is it a good idea to check the web-pages usings SGMLS and HTML.DTD
to make sure they conform (just as a courtesy to the users)?

Also, what's the state of the HTML 3 DTD? Is there one I can
download from somewhere (w3.org?) or is it still in a state of
flux? (I have tables on my pages and I'd like to check them
using a DTD+SGMLS).

rjp