svn commit: r186737 - head/sbin/geom/class/virstor

[Hiroki Sato]

Christian Brueffer <chris at unixpages.org> wrote
  in <20090104194844.GB1257 at haakonia.hitnet.RWTH-Aachen.DE>:

ch> On Mon, Jan 05, 2009 at 03:22:11AM +0900, Hiroki Sato wrote:
ch> >  I mean I am wondering if rewriting "1" with ">0" is reasonable or
ch> >  not.  "1>0" is always true, but "1" is not equal to ">0".
ch> >
ch> >  Some other manual pages have the description "1 on error.".  If we
ch> >  have a consensus on that this rewriting is reasonable, we should
ch> >  also rewrite them in consistency.
ch> >
ch>
ch> Interesting question, I have no strong opinion for either of the
ch> alternatives.  I agree that we should standardize on one though.

 IMO, the following guidelines are reasonable:

 1. If the command conforms a standard such as POSIX, the description
    in the manual page should match the standard.  For example,
    ls(1) returns >0 on error, but what(1) returns 1 on error.

 2. Otherwise, EXIT STATUS section should match the reality as
    possible.  This means if a command returns 0 on success, 1 on
    error, and the other values are not returned, the description in
    the manual page should be "it exits 0 on success, and 1 if an
    error occurs".

--
| Hiroki SATO
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/freebsd-doc/attachments/20090105/71acab03/attachment.sig>