Jeremy Kister wrote:
looks like we need to test for setenv vs putenv at configure time.
No need to test. Putenv is the preferred option, according to:
http://www.gnu.org/software/automake/manual/autoconf/Function-Portability.html
We're already using putenv several other places in the code, and just
this one instance of setenv. I've changed it to putenv in CVS. There
are a couple of other issues I want to resolve before the next release.
why does the environment variable "SHELL" have to be set for in this case?
No idea, so I don't want to change it.
!DSPAM:473eafe132001131818550!