bug-autoconf
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: _AS_ECHO_N_PREPARE - bug correction in ksh88 on AIX 6.1


From: Ralf Wildenhues
Subject: Re: _AS_ECHO_N_PREPARE - bug correction in ksh88 on AIX 6.1
Date: Tue, 30 Jun 2009 23:06:07 +0200
User-agent: Mutt/1.5.20 (2009-06-15)

Hi Eric,

* Eric Blake wrote on Tue, Jun 30, 2009 at 01:56:15PM CEST:
> According to Ralf Wildenhues on 6/30/2009 12:07 AM:
> > 
> > Thus I suggest this patch.  OK?
> 
> Yes, once these nits are fixed:

Thanks.  Pushed after fixing all of them the way you suggested, except
as noted below, and adding Jan to THANKS.  I did ensure that the new
test passes on AIX 5.3 only with the rest of the patch.

> > +entirely drop the output of the next echo in a command substitution.
> 
> Would it be worth an @example showing this weird behavior, or do you think
> the paragraph was sufficient?  I could go either way.

I think it is sufficient.

Cheers,
Ralf

    Avoid AIX 6.1 ksh88 ECHO_C command substitution bug.
    
    * lib/m4sugar/m4sh.m4 (_AS_ECHO_N_PREPARE): Ensure more than
    one character is output with `\c'; reset echo output state
    if buggy ksh was detected, and set ECHO_T instead of ECHO_C.
    * doc/autoconf.texi (Limitations of Builtins): Document it.
    * tests/m4sh.at (ECHO_C): New test.
    * THANKS: Update.

diff --git a/doc/autoconf.texi b/doc/autoconf.texi
index 455d1c4..5908bbb 100644
--- a/doc/autoconf.texi
+++ b/doc/autoconf.texi
@@ -15733,7 +15733,10 @@ Limitations of Builtins
 Solaris outputs 2, but Bash and Zsh (in @command{sh} emulation mode) output 1.
 The problem is truly @command{echo}: all the shells
 understand @samp{'\n'} as the string composed of a backslash and an
address@hidden
address@hidden  Within a command substitution, @samp{echo 'string\c'} will
+mess up the internal state of ksh88 on AIX 6.1 so that it will print
+the first character @samp{s} only, followed by a newline, and then
+entirely drop the output of the next echo in a command substitution.
 
 Because of these problems, do not pass a string containing arbitrary
 characters to @command{echo}.  For example, @samp{echo "$foo"} is safe
diff --git a/lib/m4sugar/m4sh.m4 b/lib/m4sugar/m4sh.m4
index c7b0bff..b2cabd7 100644
--- a/lib/m4sugar/m4sh.m4
+++ b/lib/m4sugar/m4sh.m4
@@ -780,14 +780,19 @@ m4_defun_init([_AS_ECHO_LOG],
 # display the checking message.  In addition, caching something used once
 # has little interest.
 # Idea borrowed from dist 3.0.  Use `*c*,', not `*c,' because if `\c'
-# failed there is also a newline to match.
+# failed there is also a newline to match.  Use `xy' because `\c' echoed
+# in a command substitution prints only the first character of the output
+# with ksh version M-11/16/88f on AIX 6.1; it needs to be reset by another
+# backquoted echo.
 m4_defun([_AS_ECHO_N_PREPARE],
 [ECHO_C= ECHO_N= ECHO_T=
-case `echo -n x` in @%:@((((
+case `echo -n x` in @%:@(((((
 -n*)
-  case `echo 'x\c'` in
+  case `echo 'xy\c'` in
   *c*) ECHO_T='        ';;     # ECHO_T is single tab character.
-  *)   ECHO_C='\c';;
+  xy)  ECHO_C='\c';;
+  *)   echo `echo ksh88 bug on AIX 6.1` > /dev/null
+       ECHO_T='        ';;
   esac;;
 *)
   ECHO_N='-n';;
diff --git a/tests/m4sh.at b/tests/m4sh.at
index 065a9dd..f936a7c 100644
--- a/tests/m4sh.at
+++ b/tests/m4sh.at
@@ -1412,3 +1412,23 @@ AT_CHECK_M4SH
 AT_CHECK([./script])
 
 AT_CLEANUP
+
+
+## -------- ##
+## ECHO_C.  ##
+## -------- ##
+
+AT_SETUP([ECHO_C])
+
+AT_DATA_M4SH([script.as], [[dnl
+AS_INIT
+_AS_PREPARE
+foo=`echo foobar`
+echo "$foo"
+]])
+
+AT_CHECK_M4SH
+AT_CHECK([./script], [], [foobar
+])
+
+AT_CLEANUP




reply via email to

[Prev in Thread] Current Thread [Next in Thread]