Re: false --help vs. false --version

[Jim Meyering]

address@hidden (Paul Jarc) wrote:
> false --help exits 0.  false --version exits 1.  I think false --help
> should be changed, but certainly one of them should be.

Thank you for pointing that out!
I've fixed it for the upcoming coreutils-5.0.

        * src/Makefile.am (false.c): Change all occurrences of
        `(EXIT_SUCCESS)' to `(EXIT_FAILURE)' so that false exits
        unsuccessfully also with --help.  Reported by Paul Jarc,
        * tests/misc/false: New test for the above.

Index: src/Makefile.am
===================================================================
RCS file: /fetish/cu/src/Makefile.am,v
retrieving revision 1.18
retrieving revision 1.19
diff -u -p -u -r1.18 -r1.19
--- src/Makefile.am     19 Mar 2003 23:17:12 -0000      1.18
+++ src/Makefile.am     2 Apr 2003 07:21:40 -0000       1.19
@@ -178,7 +178,7 @@ false.c: true.c
        sed \
          -e s/true/false/g \
          -e s/success/failure/g \
-         -e 's/^  exit (EXIT_SUCCESS/  exit (EXIT_FAILURE/g' \
+         -e 's/(EXIT_SUCCESS)/(EXIT_FAILURE)/g' \
          $(srcdir)/true.c > address@hidden
        chmod a-w address@hidden
        mv address@hidden $@


Here's what motivated me to change `false --version' to exit nonzero.

  From: Felix Lee <address@hidden>
  Subject: sh-utils 1.16: "false" isn't always false
  To: address@hidden
  Date: Wed, 03 Mar 1999 15:16:11 -0800

  The command
      false --version

  exits with zero status.  I think 'false' should always exit
  with non-zero status.

  Why this matters:

    tries="awk gawk nawk mawk"
    for try in false $tries; do
      if ($try --version) >/dev/null 2>&1; then
        echo "my favorite awk is $try"
      fi
    done

  (note, the bash built-in 'false' doesn't do anything with
  '--version' and always exits non-zero.)

-----------------
I see (in CVS log) that I made this change,

  date: 2002/07/20 15:18:20;  author: meyering;  state: Exp;  lines: +1 -1
  (false.c): Convert only the final EXIT_SUCCESS
  into EXIT_FAILURE.  Otherwise, false --help and false --version
  would fail.

but, surprisingly, there's no corresponding ChangeLog entry.
I've just added it.

-----------------
In any case, I've also added a test case, so this should
not happen again.