Trying to be more rigorous with attempts to demonstrate the question in simple code examples results in me being unable to duplicate the problem . . .

Now I am left to ponder if the problem I thought I fixed with the gsub commands was even a problem and I somehow accidently fixed somthing else (entropy notwithstanding) . . .

Code:

 
$ echo "abc,d e f,xyz" | awk -F, 'NR==1 { a[1]=$1; a[2]=$2; a[3]=$3; c[a[1]a[2]a[3]]="value"; if ( a[1]a[2]a[3] in c) print "index found" }'
index found
 
$ echo "abc,d e f,xyz" | awk -F, 'NR==1 { a[1]=$1; a[2]=$2; a[3]=$3; c[a[1]a[2]a[3]]="value"; if ( "abcd e fxyz" in c) print "index found" }'
index found
 
$ echo "abc,d e f,xyz" | awk -F, 'NR==1 { a[1]=$1; a[2]=$2; a[3]=$3; c[a[1]a[2]a[3]]="value"; if ( "something else" in c ) print "index found" }'
 
$ echo "abc,d e f,xyz" | awk -F, 'NR==1 { a[1]=$1; a[2]=$2; a[3]=$3; c[a[1]a[2]a[3]]="value"; if ( $1$2$3 in c ) print "index found" }'
index found
 
$ echo "abc,d e f,xyz" | awk -F, 'NR==1 { a[1]=$1; a[2]=$2; a[3]=$3; c[$1$2$3]="value"; if ( a[1]a[2]a[3] in c ) print "index found" }'
index found

Mike

---------- Post updated at 12:31 PM ---------- Previous update was at 12:20 PM ----------

Thanks for the suggestion but I think the error is creaping in through another source since my first awk invovation uses only $0 which appears to never get parsed and my second omitted awk statement uses the alternative delimiter.

Code:

$ echo " this is a line with beginning and end spaces plus several internal spaces ( ) " | awk 'NR==1 {print "x"$0"x"}'
x this is a line with beginning and end spaces plus several internal spaces ( ) x

Actually AWK doesn't but UNIX.com does so you'll have to take my work for it

However, you may be right about the error creaping in from something else in my input file. I will look in that direction since I have exhausted other avenues.

BTW: I have already kicked myself several times for not starting with tab seperated values from the very beginning of the project Tabs never exist in my fields.

Mike