Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT sources DOT redhat DOT com Delivered-To: mailing list cygwin AT sources DOT redhat DOT com Message-Id: <5.1.0.14.2.20010703163921.02795758@pop3.cris.com> X-Sender: rrschulz AT pop3 DOT cris DOT com X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Tue, 03 Jul 2001 17:32:53 -0700 To: jonadab AT bright DOT net, cygwin AT cygwin DOT com From: Randall R Schulz Subject: RE: documentation for windows In-Reply-To: <3B420E44.27706.3FBCCF@localhost> References: <004c01c10326$10eb2110$051865c6 AT merdelyxp> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Jonadab, Well, there's more to the manual than section 1, but that would be accommodated with this variation: grep /usr/man/man*/* Then there's the fact that you'd get all the lines from all the man page files, when what you want is to know the names of the commands on which to run "man." So: grep -l /usr/man/man*/* But the real problem is that you'll probably get too many false hits this way, since any old mention of the search pattern, regardless of whether the man page author considered it a "key word" in the synoptic description of the command. You should be aware, too, that these files are man page _sources_ written in nroff for the man macros (plus, possibly, tbl, eqn or others). This could cause further false hits. In some cases, it might even cause a miss. The pre-formatted pages for things that have already been "manned" are in /usr/man/cat*, but the formatting, including lots of embedded grunk (like backspace-encoded overtyping to get bold) and hyphenation, make grepping problematic. If you have no luck with apropos / man -k, then the grep technique might work. Once up a time on a non-Unix system (MPW), I created an "apropos" command that had options to control how much of a help file entry (the MPW equivalent to man pages) would be searched. The default was still just the synopsis. By the way, I don't think speed would be a problem with your suggestion for me, but I have two fast CPUs and pretty much the fastest disks and I/O adaptor available today. From this standpoint, I'd suggest egrep, which I use for all my "grepping" needs. Randall Schulz At 15:26 2001-07-03, Jonadab the Unsightly One wrote: ># It may also be worth noting, since you are recommending this for ># newbies, that you have to type "/usr/sbin/makewhatis" to create the ># whatis database for apropos to be useful. ;-) > >Wouldn't it be just as easy to grep /usr/man/man1/* >for your key word? Or would that take longer? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/