CSV join some rows that have the same id - ruby

Question

I have a CSV file like this
1,A,abc
2,A,def
1,B,smthing
1,A,ghk
5,C,smthing
Now I want to join all the rows that have the same value at row 2. In this case is row with the second element is A. The return file should be
1,A,abcdef,ghk
3,B,smthing
5,C,smthing
I'm trying with awk and I can get the second and the third fields but not whole file like this
awk -F, '{a[$2]=a[$2]?a[$2]$3:$3;}END{for (i in a)print i","a[i];}' old_file.csv > new_file.csv
Update
I solved my problem with 2 command. First create a new_file.csv (command above)
Second command will join old_file with new_file
awk -F, 'NR == FNR {a[$1] = $2;} NR != FNR && a[$2] {print $1","$2","a[$2];}' new_file.csv old_file.csv > last_file.csv
The last_file.csv looks like this
1,A,abcdefghk
2,A,abcdefghk
1,B,smthing
1,A,abcdefghk
5,C,smthing
So, how should I make a better command from those 2 commands?
Thank you!

Answer 1

One awk is enough:
awk 'NR==FNR{a[$2]=a[$2]==""?$3:a[$2] $3;next}{$3=a[$2]}1' FS=, OFS=, file file
1,A,abcdefghk
2,A,abcdefghk
1,B,smthing
1,A,abcdefghk
5,C,smthing
Explanation
NR==FNR{a[$2]=a[$2]==""?$3:a[$2] $3;next} merge records to array a (key is column 2)
$3=a[$2] read the input file again, change column 3 with new value.
Add the command to remove the duplicate records (column 2), keep the first one.
awk 'NR==FNR{a[$2]=a[$2]==""?$3:a[$2] $3;next}!b[$2]++{$3=a[$2];print}' FS=, OFS=, file file
1,A,abcdefghk
1,B,smthing
5,C,smthing