(1) awk -F, '{a[$2]++}END{for (i in a) print i" "a[i]}'   文件名.txt

(2) cat 文件名.txt |cut -d, -f2 |sort |uniq -c |sort -n




https://blog.csdn.net/qq_37085158/article/details/127246941
https://blog.csdn.net/cy413026/article/details/91490714
https://blog.csdn.net/qq_37788558/article/details/81780748
https://blog.csdn.net/laobai1015/article/details/91455406