awk之按特征去重并按特征内的特征计数

文本：

hash:0af1403ac9021973529b723cf9479633 date:7.6.0.2001

hash:0af1403ac9021973529b723cf9479633 date:7.6.0.2001

hash:f80cf7f9629c91aaa899e2256a6b4dac date:7.6.0.2001

hash:f80cf7f9629c91aaa899e2256a6b4dac date:7.6.0.2001

hash:f80cf7f9629c91aaa899e2256a6b4dac date:7.6.0.2001

hash:f80cf7f9629c91aaa899e2256a6b4dac date:7.6.0.2001

hash:f80cf7f9629c91aaa899e2256a6b4dac date:7.6.0.2001

hash:f80cf7f9629c91aaa899e2256a6b4dac date:7.6.0.2001

hash:f80cf7f9629c91aaa899e2256a6b4dac date:7.6.0.2001

hash:f80cf7f9629c91aaa899e2256a6b4dac date:7.6.0.2009

根据date统计hash的个数，重复的只算一个，得到结果：

7.6.0.2001 2

7.6.0.2009 1

awk -F"[ :]" '!a[$2,$NF]++{b[$NF]++}END{for(i in b)print i,b[i]}' file

[解析]

把第2个字段和第4字段防入数组a中，去重后执行数组b的计数。

awk -F"[ :]" -vc=1 '{if(x==$NF){if($2!=y)c++}else{if(NR>1){print x,c;c=1}}}{x=$NF;y=$2}' file <(echo)

[解析]

这个也是按照流程来执行比较，计数。