Re: [PATCH] HTML output fixes

[Eli Zaretskii]

> From: Ville =?ISO-8859-1?Q?Skytt=E4?= <address@hidden>
> Date: 31 Mar 2002 16:05:55 +0300
> 
> @@ -351,7 +351,7 @@
>              already_outputting_pending_notes++;
>              execute_string ("%s", footnote->note);
>              already_outputting_pending_notes--;
> -            add_word ("</p>\n");
> +            add_word ("</p>\n</li>\n");
>            }

Please be very careful about this change.  The markup generated for
footnotes strikes a fine balance between the need to produce something
that looks like a footnore (a notion that HTML lacks) and the
capabilities of popular browsers.  IIRC, it took a few iterations to
arrive at the current version of the code, and it's quite possible
that it doesn't produce a </li> on purpose.  Do we _really_ need to
add the </li> (yes, I know about Lint-like utilities that could
complain; but do we actually _need_ to comply)?