PATCH: tcsetpgrp test switch

PATCH: tcsetpgrp test switch

[Clint Adams]

This provides a configure switch to prevent the testing of tcsetpgrp.

--- configure.in.bak	Tue Nov  2 14:43:47 1999
+++ configure.in	Tue Nov  2 15:05:47 1999
@@ -250,6 +250,13 @@
   AC_DEFINE(MAILDIR_SUPPORT)
 fi])
 
+dnl Do you want to verify that tcsetpgrp works?
+undefine([tcsetpgrp-check])dnl
+AC_ARG_ENABLE(tcsetpgrp-check,
+[  --disable-tcsetpgrp-check     Disable test of tcsetpgrp],
+[tcsetpgrp-check="$enableval"], [tcsetpgrp-check=yes])
+
+
 dnl ------------------
 dnl CHECK THE COMPILER
 dnl ------------------
@@ -1240,8 +1247,12 @@
 dnl -----------
 dnl if found tcsetpgrp, test to see if it actually works
 dnl for instance, BeOS R4.51 does not support it yet
+dnl When building where stdin is not a terminal, this
+dnl check will fail, resulting in no job control support
+dnl being available.  Because this is undesirable, allow
+dnl it to be skipped.
 dnl -----------
-if test $ac_cv_func_tcsetpgrp=yes; then
+if test $ac_cv_func_tcsetpgrp=yes -a $tcsetpgrp-check=yes; then
     AC_CACHE_CHECK(if tcsetpgrp() actually works,
     zsh_cv_sys_tcsetpgrp,
     [AC_TRY_RUN([

Re: PATCH: tcsetpgrp test switch

[Bart Schaefer]

On Nov 2,  3:10pm, Clint Adams wrote:
} Subject: PATCH: tcsetpgrp test switch
}
} This provides a configure switch to prevent the testing of tcsetpgrp.

There doesn't seem to be much point in performing this test at all when
it can't possibly succeed.  Rather than a command-line switch, why not:

Index: configure.in
===================================================================
@@ -1233,7 +1233,7 @@
 dnl if found tcsetpgrp, test to see if it actually works
 dnl for instance, BeOS R4.51 does not support it yet
 dnl -----------
-if test $ac_cv_func_tcsetpgrp=yes; then
+if test -t 0 -a $ac_cv_func_tcsetpgrp=yes; then
     AC_CACHE_CHECK(if tcsetpgrp() actually works,
     zsh_cv_sys_tcsetpgrp,
     [AC_TRY_RUN([


One reason "why not" may be that it's better to err on the side of no job
control rather than broken job control; but perhaps in that case we should
try something more clever -- such as, build in the job control functions
but default the state of the option to NO_MONITOR.

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Re: PATCH: tcsetpgrp test switch

[Clint Adams]

> One reason "why not" may be that it's better to err on the side of no job
> control rather than broken job control; but perhaps in that case we should
> try something more clever -- such as, build in the job control functions
> but default the state of the option to NO_MONITOR.

If someone with BeOS can confirm that test -t 0 works then I
don't see why not.

Re: PATCH: tcsetpgrp test switch

[Will Day]

[-- Attachment #1: Type: text/plain, Size: 1228 bytes --]

A short time ago, at a computer terminal far, far away, Clint Adams wrote:
>> One reason "why not" may be that it's better to err on the side of no job
>> control rather than broken job control; but perhaps in that case we should
>> try something more clever -- such as, build in the job control functions
>> but default the state of the option to NO_MONITOR.
>
>If someone with BeOS can confirm that test -t 0 works then I
>don't see why not.

Yes, "test -t" does appear to work under BeOS R4.5:

   $ /bin/sh
   sh-2.02# test -t 0; echo $?
   0
   sh-2.02# test -t 1; echo $?
   0
   sh-2.02# test -t 2; echo $?
   0
   sh-2.02# test -t 3; echo $?
   1
   sh-2.02# test -t 0 < /dev/null; echo $?
   1
   sh-2.02# test -t 1 > /dev/null; echo $?
   1
   sh-2.02# test -t 2 2> /dev/null; echo $?
   1

-- 
Will Day     <PGP mail preferred>     OIT / O&E / Technical Support
willday@rom.oit.gatech.edu            Georgia Tech, Atlanta 30332-0715
  -> Opinions expressed are mine alone and do not reflect OIT policy <-
Those who would give up essential Liberty, to purchase a little temporary
Safety, deserve neither Liberty nor Safety.
    Benjamin Franklin, Pennsylvania Assembly, Nov. 11, 1755

[-- Attachment #2: Type: application/pgp-signature, Size: 344 bytes --]