Hello Bob.

27 Sep 03 21:23, you wrote to Vince Coen:

 BJ> Ok.  I don't think we've had this platform tested before.  I suspect
 BJ> that you are running newer tools that Bo or I am in some areas.
 BJ> [We're both running Debian setups, probably with older GCC compilers,
 BJ> and similar age for other apps.]  So, there might be something in the
 BJ> code that's biting us with newer tools.....  Hmmmm......

What I noticed in his posted log was that it was failing in the exact same
location where it had failed when I first tried compiling it.  Bo and I
discovered that it was due to a call in the squish makefile to
"nawk" rather then "awk".  I editted it and it then
compiled successfully.  Since then Bo had updated the cvs and I tried it
from scatch, downloading, configuring and compiling without edits and it
again compiled successfully.  We both decided that using "awk" as
a call in the makefile was a good idea seeing that no matter what
"awk" was, or at least should be, a symbolic link on any Linux
dist. I have seen.  On Slackware "awk" is a symbolic link to
"gawk".  It looks to me that Vince might be using a previous
download from the cvs prior to Bo uploading the new and improved makefile. 
Just a hunch but I believe a good one.  If Vince could check
maximus/squish/Makefile there should be a line that reads:

ftscprod.c:
        awk -F, -f mkprod.awk  ftscprod.c

If it says "nawk -F, ..." then he should edit it to say
"awk" and try recompiling.

Maurice

--- GoldED+/LNX 1.1.4.7