groff
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Groff] The permuted index from Hell


From: Clarke Echols
Subject: Re: [Groff] The permuted index from Hell
Date: Wed, 14 Jun 2006 19:33:56 -0600
User-agent: Thunderbird 1.5.0.4 (Windows/20060516)

When I first had responsibility for HP's manpages, I was completely
appalled at the uselessness of the permuted index that was taken
from the NAME line.  For example, on cp(1), the line was:

   cp, mv, ln - copy, move, or link files

I solved the problem by embedding comments in each file such as
.\" toc cp(1)<tab>copy file, files, or directory subtrees
.\" toc mv(1)<tab>move or rename file or files or directory
.\" index rename file<tab>mv(1)
.\" index create zero-length file<tab>touch(1)
etc.

as appropriate for each file.  I used grep to pull the comments from
the files, ran them through sed, sort, etc. and produced a masterful
and highly useful index with lots of goodies.

What started that was my early experiences with unix where I went
to the permuted index and looked for "rename a file".  I was not
plesed when the only thing I could find is the rename(2) system call.

It's called USABILITY.  AT&T's original approach may have been
"neat" but it wasn't valuable as a useful tool...

Clarke


Larry Kollar wrote:
So I got this bright idea to build a permuted index for the MIBs we support in our products at work. It might be helpful for customers who are trying to figure out which MIB object would have the information they're looking for.

So I grabbed the textutils package and built it, checking over the documentation several times, installing an "eign" file in the right place, and adding the #define to the ptx source code so it will *find* the ignore list. Then once I fed some sample data in, I figured out which options I needed (settling on "ptx -frRO -F '...'"), then wrote an awk script to pull what I wanted out of the MIB files.

After a bit of debugging, and writing a Q&D ".xx" macro to format the index, I was ready to see what I had. What I had was 253 pages of 8-point text, formatted properly.

Back to the drawing board, I guess.

--
Larry Kollar     k  o  l  l  a  r  @  a  l  l  t  e  l  .  n  e  t
Unix Text Processing: "UTP Revival"
http://unixtext.org/




_______________________________________________
Groff mailing list
address@hidden
http://lists.gnu.org/mailman/listinfo/groff





reply via email to

[Prev in Thread] Current Thread [Next in Thread]