Bug 167009 - grep(1): GNU grep -q can exit !0 even if strings found
Summary: grep(1): GNU grep -q can exit !0 even if strings found
Status: Closed FIXED
Alias: None
Product: Base System
Classification: Unclassified
Component: gnu (show other bugs)
Version: 8.3-PRERELEASE
Hardware: Any Any
: Normal Affects Only Me
Assignee: Jilles Tjoelker
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-04-16 22:40 UTC by david
Modified: 2021-10-09 01:30 UTC (History)
1 user (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description david 2012-04-16 22:40:04 UTC
	The -q flag for GNU grep is described as:

`-q'
`--quiet'
`--silent'
     Quiet; do not write anything to standard output.  Exit immediately
     with zero status if any match is found, even if an error was
     detected.  Also see the `-s' or `--no-messages' option.

`-s'
`--no-messages'
     Suppress error messages about nonexistent or unreadable files.
     Portability note: unlike GNU `grep', traditional `grep' did not
     conform to POSIX.2, because traditional `grep' lacked a `-q'
     option and its `-s' option behaved like GNU `grep''s `-q' option.
     Shell scripts intended to be portable to traditional `grep' should
     avoid both `-q' and `-s' and should redirect output to `/dev/null'
     instead.

	Despite this, there are cases where the use of "grep" to find a
	string will succeed (with an exit status of 0), while an
	otherwise-identical search using "grep -q" will merely exit with
	a non-zero status code.

	While it's possible that I'm confused, that seems to contradict
	the above info excerpt.  It's also possible that there's
	something peculiar about the FreeBSD environment (from the
	perspective of the GNU folks).

How-To-Repeat: 	I encountered this in looking for certain pathnames in
	executables, thus:

albert(8.3-P)[1] which grep
/usr/bin/grep
albert(8.3-P)[2] grep --version
grep (GNU grep) 2.5.1-FreeBSD

Copyright 1988, 1992-1999, 2000, 2001 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

albert(8.3-P)[3] grep -alw /usr/local /usr/local/bin/bison; echo $?
/usr/local/bin/bison
0
albert(8.3-P)[4] strings -a /usr/local/bin/bison | grep -w /usr/local; echo $?
/usr/local/lib
/usr/local/share/locale
/usr/local/share/bison
/usr/local/bin/gm4
/usr/local/lib
0
albert(8.3-P)[5] strings -a /usr/local/bin/bison | grep -qw /usr/local ; echo $? 
141
albert(8.3-P)[6] 

	Most of the folks to whom I've shown this have agreed that
	0 != 141.  :-}
Comment 1 Jilles Tjoelker freebsd_committer freebsd_triage 2012-04-17 20:08:12 UTC
FreeBSD PR gnu/167009:
> [strings -a /usr/local/bin/bison | grep -qw /usr/local fails]

The peculiar thing is the shell, tcsh.

What happens is that grep -q terminates early, so that strings receives
SIGPIPE when it tries to write further data. Whereas the exit status
of a pipeline in sh is always the exit status of the last element so
that your command has the expected behaviour, this is different in tcsh.
Tcsh looks backwards for a failing command if the last element has exit
status 0, and returns the 141 corresponding to SIGPIPE.

If you do not use -q, grep reads all of its input and strings will not
be hit by SIGPIPE, so the exit status of the pipeline is the exit status
of grep.

Something similar happens if 'set -o pipefail' is in effect in bash or
ksh93.

I can't help but be happy that I'm not using tcsh, because
  tcsh -c 'yes | :'
does return 0 (apparently it doesn't look backwards if the last element
is a builtin) and
  tcsh -c 'yes | if (1) true'
hangs. Also, the double prompt I see after typing
  yes | :
in an interactive tcsh doesn't really inspire confidence.

-- 
Jilles Tjoelker
Comment 2 david 2012-04-17 20:12:15 UTC
Huh.  Thanks for the explanation.  :-}

Peace,
david
-- 
David H. Wolfskill				david@catwhisker.org
Depriving a girl or boy of an opportunity for education is evil.

See http://www.catwhisker.org/~david/publickey.gpg for my public key.
Comment 3 Jilles Tjoelker freebsd_committer freebsd_triage 2012-04-17 23:05:26 UTC
State Changed
From-To: open->closed

This is not a bug but a result of exit status of tcsh pipelines. 


Comment 4 Jilles Tjoelker freebsd_committer freebsd_triage 2012-04-17 23:05:26 UTC
Responsible Changed
From-To: freebsd-bugs->jilles

Take.