Linux Shell编程实战---统计特定文件中单词的词频
方法1：使用sedShell>cat a1.txt
123a123,555
456.333
566。555!88,thisis a good boy.
Shell>cat a1.txt|sed 's/[[:space:]|[:punct:]]/\n/g'|sed '/^$/d'|sort|uniq -c|sort -n-k1 -r
2 555
1 this
1 is
1 good
1 boy
1 a123
1 a
1 88
1 566
1 456
1 333
1 123
Shell>
sed 's/[[:space:]|[:punct:]]/\n/g'
[]表示正则表达式集合，[:space:]代表空格。[:punct:]代表标点符号。
[[:space:]|[:punct:]]代表匹配空格或者标点
s/[[:space:]|[:punct:]]/\n/g代表把空格或标点替换成\n换行符
sed '/^$/d' 删除掉空行
方法2：使用awk
#!/bin/bash
filename=$1
cat$filename|awk '{
#getline var;
split($0,a,/[[:space:]|[:punct:]]/);
for(i in a) {
word=a;
b++;
}
}
END{
printf("%-14s%s\n","Word","Count");
for(i in b) {
printf("%-14s%d\n",i,b)|"sort-r -n -k2";
}
}
'
运行结果
# cat a1.txt
123a123,555
456.333
566。555!88,thisis a good boy.
# ./word_freq.sh a1.txt
Word Count
555 2
this 1
is 1
good 1
boy 1
a123 1
a 1
88 1
566 1
456 1
333 1
123 1
1
#
方法3：使用tr
# cat a1.txt
123a123,555
456.333
566i555!88,this is a good boy.
# cat a1.txt |tr '[:space:]|[:punct:]' '\n'|tr -s '\n'|sort|uniq -c|sort -n-k1 -r
2 555
1 this
1 is
1 good
1 boy
1 a123
1 a
1 88
1 566i
1 456
1 333
1 123
#
页:
[1]