Ticket #636 (closed enhancement: fixed)
Process to get most recent man pages to wiki
| Reported by: | dclark | Owned by: | dclark |
|---|---|---|---|
| Priority: | minor | Milestone: | Bcfg2 1.1.0 Release |
| Component: | bcfg2-doc | Version: | |
| Keywords: | Cc: |
Description
There is some stuff that is only doc'ed in bcfg2 man pages (eg filemonitor=pseudo); it would be useful to have that stuff be in the wiki so it's searchable along with the rest of the doc. I'm guessing something like a cron jobs that runs man2html and then wraps that in trac's "this is html" syntax. For now this is also just a reminder for me to do this manually...
Attachments
Change History
comment:2 Changed 14 years ago by dclark
See ManualPages? for current state. Also above should actually read:
man2html -M /projects/bcfg2/wiki -p (man file) > (html file)
Also you'll need to wrap the stuff you cut between the body tags in:
{{{
#!html
(your html here)
}}}
comment:3 Changed 14 years ago by dclark
- Owner changed from desai to dclark
- Status changed from new to accepted
comment:4 Changed 14 years ago by dclark
Option 1: push from source to trac
So each man page should have embedded in it the URL of the web version.
As part of the new release version process, a script is run that looks through the man pages for these strings, and does the trac updates - script will prompt for trac username and password.
Option 2: trac pulls from source
An alternative would be to write a man page parser for trac, and then use a include macro or something like that. Man pages are written in the troff or groff language.
http://trac.edgewall.org/wiki/ProcessorBazaar - http://trac.lighttpd.net/xcache/wiki/ShellExample - looks really useful
Man page format converters
man2html a Perl program that reads formatted nroff text from standard input (stdin) and writes a HTML document to standard output (stdout). A CGI program is provided to allow hypertext navigation of a system's manpages via a web browser.
- http://sourceforge.net/projects/polyglotman/ - last update Sat Jul 26 2003
Parse formatted man pages and man page source from most flavors of UNIX. Convert to HTML, ASCII, TkMan?, DocBook?, and other formats.
comment:5 Changed 13 years ago by http://fab.id.fedoraproject.org/
Perhaps the 'man page renderer plugin' [1] for Trac can help with this issue.
The command line I'm using to do this manually is:
And then cut-and-paste everything between but not including <BODY> and </BODY>.