Write a bash script to calculate the frequency of each word in a text file words.txt.

For simplicity sake, you may assume:

• words.txt contains only lowercase characters and space ' ' characters.
• Each word must consist of lowercase characters only.
• Words are separated by one or more whitespace characters.

For example, assume that words.txt has the following content:

the day is sunny the the

the sunny is is

Your script should output the following, sorted by descending frequency:

the 4

is 3

sunny 2

day 1

# Read from the file words.txt and output the word frequency list to stdout.

cat words.txt | tr -s ' ' '\n' | sort | uniq -c | sort -rn | awk '{print $2" "$1}'

tr -s: 使用指定字符串替换出现一次或者连续出现的目标字符串（把一个或多个连续空格用换行符代替）

sort: 将单词从小到大排序

uniq -c: uniq用来对连续出现的行去重，-c参数为计数

sort -rn: -r 倒序排列， -n 按照数值大小排序（感谢网友 长弓1990 指正）

awk '{ print $2, $1 }': 格式化输出，将每一行的内容用空格分隔成若*分，$i为第i个部分。

本文链接：http://bookshadow.com/weblog/2015/03/24/leetcode-word-frequency/