awk處理統計值的問題
來源：互聯網
上載者：User
file1
0011,aaa,901
0013,bbb,912
0017,ccc,909
0010,ddd,910
0044,eee,911
file2
0011,.3700,20080321
0013,17.3300,20080321
0011,.8800,20080621
0011,27.2200,20080921
0013,17.8800,20080621
0013,.3300,20080921
0010,46.8700,20080321
0010,87.2300,20080621
要得到的結果應該是如下(即將file2中的相同$1的$2值累加,然後和file1中的匹配列印):
0011,aaa,901,28.47
0013,bbb,912,35.54
0010,ddd,910,134.1
最初的做法是:
cat file2 | awk 'BEGIN{FS=",";OFS=","}{array[$1]=array[$1]+$2;}END{for(i in array){while(getline < "file1")if($1==i) print "=:"i,array[i],$2,$3;close("file2");}}'
現在更簡潔的做法:
awk 'BEGIN{FS=OFS=","} NR==FNR{array[$1]+=$2;} NR!=FNR&&($1 in array){print $0,array[$1]}' file2 file1
可以看一下NR和FNR的區別這樣就很好理解上面代碼了(awk帶兩個檔案的用法兩檔案都會讀入)
NR 已經讀出的記錄數
FNR 目前記錄數
awk '{print NR, FNR , $0}' file2 file1
1 1 0011,.3700,20080321
2 2 0013,17.3300,20080321
3 3 0011,.8800,20080621
4 4 0011,27.2200,20080921
5 5 0013,17.8800,20080621
6 6 0013,.3300,20080921
7 7 0010,46.8700,20080321
8 8 0010,87.2300,20080621
9 1 0011,aaa,901
10 2 0013,bbb,912
11 3 0017,ccc,909
12 4 0010,ddd,910
13 5 0044,eee,911
