Holger Weiss typed (on Sat, Jun 16, 2007 at 10:34:16PM +0200):
| * Jean-Pierre Radley <[EMAIL PROTECTED]> [2007-06-16 15:33]:
| > Derek Martin typed (on Sat, Jun 16, 2007 at 01:03:47PM -0400):
| > | It does not apply to other system commands, because:
| > |
| > | - The standards assume that "standard utilities" have a non-zero
| > | exit status when an error condition has occured
| >
| > Depends on the command. My man page for grep|egrep|fgrep says:
| >
| > Exit values
| >
| > Exit status is 0 if any matches are found,
| > 1 if no matches are found, and 2 for syntax
| > errors or inaccessible files.
|
| This is a nice counter-example to the behaviour of Solaris' Vi, as it
| shows an extremly useful definition of exit values (used in countless
| scripts world-wide) as opposed to the completely useless values Solaris'
| Vi returns (which forces the parent-process to ignore them entirely).
|
| > Not finding a certain string in some files means that it is *false* that
| > this string resides in these files. I wouldn't characterize that as any
| > kind of *error*.
|
| Me not either, and therefore I wouldn't even consider the behaviour of
| Solaris' Vi POSIX compliant, since the standard says that Vi shall exit
| non-zero only if an *error* occurs. However, no matter whether or not
| such stupid behaviour is formally covered by some standard, the question
| is whether Mutt should explicitely support this behaviour or whether it
| should be left to the user. Personally, I'd opt for the latter.
Current POSIX seems to say that, but go back a few years, and the POSIX
standard said that a search error while in vi adds to the exit status
number; many people still use a vi binary that conforms to that older
standard.
--
JP
==> http://www.frappr.com/cusm <==
