Re: Proposal to remove catman(8)

[Robert Elz <kre%munnari.OZ.AU@localhost>]

    Date:        Tue, 10 Nov 2020 19:28:41 +0100
    From:        Kamil Rytarowski <kamil%netbsd.org@localhost>
    Message-ID:  <c4db1f9d-85ee-f90a-2ea0-c1a6448b39ca%netbsd.org@localhost>

  | I hope this is a typo, and not the indication that you forgot how to use
  | the cat-pages at all and miss a computer to cross-check how these files
  | are named.

As in my reply to Mouse, I just didn't bother to check, I did what I
did, and tested it, and it worked.   But:

  | cat-pages always finish with .0

that works even better.   Even more readable (I guess there's a difference
in the method used to display the file).

  | Personally, I miss ditroff, as I have got some documentation in this
  | format that is not formatted promptly with other tools I checked.

Huh?   There's very little that ditroff (which is just a troff implementation
with a more general set of output drivers than the original troff had, that's
the "di" - device independant) can format that groff does differently (groff
has many extensions, but if they're not being used, that's harmless).

You do need the appropriate macros (whatever the source assumes) of course.
And you need the appropriate pre-processors (if any are used).

  | I didn't differentiate MKCATPAGES=yes from catpages support.

Exactly.   That's what various people have been telling you.

  | I also wrote explicitly to use dynamic manual pages afterward.

Yes, and that's fine too, I guess there are some uses (but beware of
reducing the width, many pages won't work properly at all if the
line width is too small - they'll format, but be extremely ugly).

  | This was already phrased in the original MKCATPAGES mail and repeated;
  | the wanted feature is to have MANWIDTH and it needs reformatting the
  | pages on the fly.

That's fine.

  | This conflicts with the pregenerated pages concept

Yes, so you either just ignore the cat pages, or use them only if
they were generated at the desired width.   It doesn't mean that
support for preformatted pages needs to be deleted.

  | which I find no longer relevant any more for the original reasons (such
  | as performance on multiuser vax780 from 1977?

79 I think.   [Irrelevant note: 20 users on a 780 was nothing, ours
typically had about 60 connected, with many others leaving processes
running in background after logging out - the load average was sometimes
over 100.   But not all of them would have been reading man pages.]

  | FreeBSD implements it by disabling cat-pages behind the scenes. I find
  | this approach waste of time in 2020. It could be maybe tentative in 2011
  | when implemented in FreeBSD.

But I thought you wanted to be compatible with other systems, yet you're
apparently striving to be different to FreeBSD ???

  | I keep having some spurious fallout in builds and the re-addition (seems
  | to be after sqlite3 changes) and I keep having constant build failures.
  | I will add it.

Thanks.   And also for testing before committing - that's always the right way.

kre