Shell Programming and Scripting

I am trying to figure out a way in nawk to 1) get a count of the number of times a value appears in field 1 and 2) count each time the same value appears in field 2 for each value of field 1. So for example, if I have a text file with the following:

grapes, purple
apples, green
squash, yellow
grapes, green
apples, red
squash, green,
grapes, red
apples, yellow
squash, green


I want to figure out how many times each color occurs for each type of fruit. There may be new fruit thrown in the file from time to time so i don't have a hard-and-fast list of the fruit and would need the script to be able to adjust for new fruit.

I have tried making changes to the following code, but I keep getting syntax errors:

Code:

nawk -F, '{a[$1]++}END{for(i in a){b[$2]++}}END{print i"\t"a[i]"\t"b[i]}' filename.txt

You can use two or more things as array indexes, separated by a comma. It will be treated as one string with SUBSEP between the items. So you can have A["grapes","purple"]=2 for example.

Code:

awk -F"," '{ A[$1]++; B[$1,$2]++ } END { for(X in A) { print X, A[X]; }; for (X in B) { split (B, Z, SUBSEP); print Z[1], Z[2], B[X] }'

Code:

nawk ' { a[$0] ++ } END { for (i in a) print i" = "a[i] } ' FS="\n" filename.txt