Re: makemandb errors from pkgsrc

[Dave Huang <khym%azeotrope.org@localhost>]

On Mar 5, 2012, at 5:38 PM, Joerg Sonnenberger wrote:
> The same argument has been used to complain about compiler warnings.
> Guess what -- often enough they point out real issues. The major reason
> why so many man page have smaller or bigger markup bugs is that it
> always was way too easy to ignore the warnings groff throws. The
> messages are parse errors. They signify that the markup or content is
> broken in a way that the document isn't indexed. As such it is the
> perfect example of what should not be silenced.

Yes, maybe there are real issues with the manpage. And what are the 
consequences? If you ignore a compiler warning, maybe the program will dump 
core or have a security vulnerability. Sounds pretty bad. If you ignore a 
manpage indexing error, maybe apropos will only show you a one-line summary, as 
it did in NetBSD 5.x and earlier. Certainly not ideal, but not too bad... *nix 
has been working like that for decades.

And what's an end user supposed to do with these manpage errors? Compiler 
warnings are seen by the developers writing the code, and when they see the 
warning, they can/should do something to fix it. makemandb's errors are seen by 
end-users of the manpages, most of whom did not write the manpage, and can't do 
anything to fix it. The issue is that makemandb is showing me errors that I 
can't do anything about. Manpage authors should lint their pages, and pkgsrc 
package maintainers should lint their package's pages--but end users shouldn't 
get daily reminders of manpage authors' failure to do so.

-- 
Name: Dave Huang         |  Mammal, mammal / their names are called /
INet: khym%azeotrope.org@localhost |  they raise a paw / the bat, the cat /
FurryMUCK: Dahan         |  dolphin and dog / koala bear and hog -- TMBG
Dahan: Hani G Y+C 36 Y++ L+++ W- C++ T++ A+ E+ S++ V++ F- Q+++ P+ B+ PA+ PL++