html interface to all docs?
Some time ago on debian-user I and some others were discussing a html front
end for all documents on a system. That is, a system where all documents
could be accessed via a WWW browser. Manual pages and info files would
be turned into html, plain text showed as is, and PostScript and other
difficult formats would be viewed using external viewers (as usual for
WWW). The inital front end page would contain a "menu" of the various
I don't think anything happened. The following is just going out on
comp.os.linux.announce. Perhaps someone could use it as a starting point?
Or at least package it for Debian as well (which shouldn't be too much of
------- Forwarded Message
From: Michael Hamilton <email@example.com>
Subject: VH-Man2html 1.3 - man/mandoc source to HTML
* View man pages using your web browser and httpd.
* HTML GENERATED DIRECTLY FROM NROFF AND TBL SOURCE without the need for
tbl/groff/nroff (sorry eqn isn't supported).
* Follow links to other man pages and C include files.
* Browse HTML-ised whatis man-page subject indexes.
* Browse HTML-ised name-only man-page indexes.
* Optionally add glimpse (a text indexing package) to do full test searches.
* Faster than gunzip|tbl|nroff|less (if your browser is already running).
* Translates both man and mandoc (BSD) man pages.
* True HTML tables generated from tbl.
* Serve man pages to other UNIX hosts.
* Works with netscape and with lynx (except for tables).
* Packaged for Redhat/Caldera linux, but could be tailored for any UNIX
that stores man pages in source form.
* Translator written in C (fast), scripts written in awk.
* RPM available at ftp.redhat.com.
* BSD-style unrestricted use copyright (except for full text searches
because of the dependency on glimpse - suggestions for alternatives to
* See sample output and more info at
* You need to be running an httpd daemon - although I guess you could
use it to pre-translate the pages and store them as files.
* Man pages in source form - with a whatis file - in a normal looking
UNIX man hierarchy.
* Knowledge of Makefile editing if you don't have Redhat/Caldera.
VH-Man2html was was written by Richard Verhoeven (NL:5482ZX35) at
the Eindhoven University of Technology (Email: firstname.lastname@example.org). The
original source is available from his web page at:
There are other man2html programs around - Richard's page has links to
several of them. I've called my version of Richard's program
VH-Man2html to avoid any confusion both with Richards original and
with these other utilities. VH = Richard Verhoeven's Man2html as modified
and packaged up by Michael Hamilton (email@example.com).
Description: CGI-Scripts for turning a web-browser+httpd into a man pager.
Requires man source pages. Designed for Caldera/Redhat
distributions using man-1.4d/g.
Requires web browser with CGI <ISINDEX> capability (and
HTML tables for man pages that have tbl tables).
- C CGI program for man/BSD-mandoc to HTML conversion.
- Works from the unformatted nroff/troff source.
- Does tbl tables (but not eqn equations).
- Generates links to foobar(1), abc@host, and xyzzy.h files
- CGI script for whatis-based alpha-indexes by section.
- CGI script for name-only alpha-indexes by section.
- CGI script for full text search (requires glimpse -
- Front-end script to talk to a pre-launched netscape.
Only tested on Caldera 1.0.
Sample pages (with non-functional links) can be viewed at
Keywords: man mandoc man2html HTML manual pager browser
Author: Man2html.c - Richard Verhoeven (firstname.lastname@example.org).
Mandoc enhancements, index scripts, packaging etc -
Michael Hamilton (email@example.com).
Maintained-by: firstname.lastname@example.org (Michael Hamilton)
Primary-site: sunsite.unc.edu /pub/Linux/system/Manual-pagers
Secondary-site: ftp.redhat.com /pub/contrib
Copying-policy: BSD-style use-for-any-purpose.
Michael Hamilton (http://www.actrix.gen.nz/users/michael)
Happily running Linux since January 1992.
------- End of Forwarded Message