Re: Converting preformatted documents to HTML

Eelco van Asperen (
Fri, 7 Aug 92 15:30:22 +0200

> I agree that it would be useful to have a mono-spaced litteral
> text format which would still accept tags other than the closing
> tag. Because tags, especially if
> automatically generated, can easily produce very long lines
> unless they are broken up, it might be useful for
> thiat format to have explicit new lines. (The paragraph
> mark <P> could be used, with hte convention that it
> represents only a new line). Like you, I have come
> across the need for it mainly in converted information.
> It would be very easy to put in the browsers.

ok, I'll look into that; any thoughts on a name for this format ?

> By the way, have you seen the Gnu Info convertion
> which Philippe Defert does, on ?
> He uses a perl script. Would you both like to publish
> your scripts? They would be useful examples for anyone
> writing converters, and people might wantto set up their
> own Gnu Info servers too.

Yes, I've seen this script. My own script is also in perl but
performs a more complete conversion. For example, each Info
'node' has previous/next references and embedded in the text
can be references to other documents. Also, my script is explicitly
designed to perform the conversion for a single topic when the user
wants to read this topic. This means that there's no need to store
the converted document and that the data is always up to date.
The disadvantage of this approach is that the scripts has to been
every time a request is made.

The scripts and the modifications I made to the HTTP daemon are
now available via anonymous ftp from in the
/pub/www directory, files Daemon4.0.patch, info2html and man2html.
Or try

I'm on holiday next week so problems/flames/praise will have to
wait until I'm back.

Eelco van Asperen. | Erasmus University Rotterdam
-----------------------------| Department of Computer Science, room H4-32
internet: | PObox 1738, 3000 DR Rotterdam, The Netherlands