man2html.1   [plain text]

'\" t
.\" Man page for man2html
.\" aeb, 980101
.TH man2html 1 "1 January 1998"
.LO 1
man2html \- format a manual page in html
man2html [options] [file]
.B man2html
converts a manual page as found in
.I file
(or stdin, in case no file argument, or the argument "-", is given)
from man-style nroff into html, and prints the result on stdout.
It does support tbl but does not know about eqn.
The exit status is 0. If something goes wrong,
an error page is printed on stdout.

This can be used as a stand-alone utility, but is mainly intended
as an auxiliary, to enable users to browse their man pages using
a html browser like
.BR lynx (1),
.BR xmosaic (1)
.BR netscape (1).
./" (See
./" .BR man (1)
./" for info on how to browse man pages via
./" .BR man2html .
./" Usually it would suffice to put "MANHTMLPAGER=/usr/bin/lynx"
./" in the environment.)

The main part of
.B man2html
is the troff-to-html engine written by Richard Verhoeven (
It adds hyperlinks for the following constructs:
l l.
foo(3x)	"http://localhost/cgi-bin/man/man2html?3x+foo"
method://string	"method://string"	""	""
name@host	"mailto:name@host"
<string.h>	"file:/usr/include/string.h"
(The first of these can be tuned by options - see below.)
No lookup is done - the links generated need not exist.
Also an index with internal hyperlinks to the various sections
is generated, so that it is easier to find one's way
in large man pages like
.BR bash (1).

When reading from stdin, it is not always clear how to
do .so expansion. The \-D option allows a script to define
the working directory.
.B \-\^D pathname
Strip the last two parts from the pathname, and do a
\fIchdir\fP(\fIdir\fP) before starting the conversion.
The \-E option allows the easy generation of error messages
from a cgi script.
.B \-\^E string
Output an error page containing the given error message.
The general form of a hyperlink generated for a man page reference is
with a default as shown above. The parts of this hyperlink
are set using the various options.
.B \-\^h
Set method:cgipath to http://localhost. This is the default.
.BI \-\^H " host[.domain][:port]"
Set method:cgipath to
.RI http:// host.domain:port .
.B \-\^l
Set method:cgipath to
.RI lynxcgi: /home/httpd .
.BI \-\^L " dir"
Set method:cgipath to
.RI lynxcgi: dir .
.BI \-\^M " man2htmlpath"
Set the man2htmlpath to use. The default is
.IR /cgi-bin/man/man2html .
.B \-\^p
Set separator to '/'.
.B \-\^q
Set separator to '?'. This is the default.
.B \-\^r
Use relative html paths, instead of cgi-bin paths.
On a machine without running
.BR httpd ,
one can use
.B lynx
to browse the man pages, using the lynxcgi method.
When some http daemon is running, lynx, or any other browser,
can be used to browse the man pages, using the http method.
The option \-l (for `lynxcgi') selects the former behaviour.
With it, the default cgipath is \fI/home/httpd\fP.

In general, a cgi script can be called by
and the environment variables PATH_INFO and QUERY_STRING
will be set to <more_path> and <query>, respectively.
Since lynxcgi does not handle the PATH_INFO part, we generate
hyperlinks with `?' as a separator by default.
The option \-p (for `path') selects '/' as a separator, while
the option \-q (for `query') selects '?' as a separator.

The option \-H \fIhost\fP will specify the host to use
(instead of \fIlocalhost\fP).  A cgi script could use
man2html -H $SERVER_NAME
if the variable SERVER_NAME is set.  This would allow your machine
to act as a server and export man pages.

There are many heuristics.  The output will not always be perfect.
The lynxcgi method will not work if lynx was compiled without
selecting support for it.  There may be problems with security.

.BR lynx (1),
.BR man (1)