AWK, the Three Musketeers of Linux

#AWK命令基础显示

Print the second area of the Install.log file that contains the data field row

awk ‘/data/ {print $2}’ install.log

View the first line of Num10.txt

head -n 1 num10.txt结果：10cat !$  #cat上次打开的文件，cat num10.txt

AWK displays only the first line:

NR number of rows in row

awk reads a file one line at a read;

awk ‘NR==1‘ num10.txt结果：10

awk prints the last line:

awk ‘END{print $0}‘ num10.txt结果： 1 最后一行

Show the last column:

NF Number of fields column

awk -F ‘ ‘ ‘{print $NF}‘ qqq结果：lll222

Do not display the first row

awk ‘NR != 1‘ num10.txt

Do not display the last column

awk -F ‘ ‘ ‘BEGIN{OFS="---"}{print $1,$2,$3}‘ qqq  结果：123---123---c1fsda---ddd3---222

ORS, the default is the line break after each line ends

awk -F " " ‘BEGIN{OFS="";ORS=""}{for(i = 1;i < NF; i++){print $i" "}{printf "\n"}}‘ qqq   

Base operations

Suddenly think of a python solution, to two instances ~

There is now a file QQQ, which reads as follows:

1 2 3) 4 5

2 3 4) 5 6

3 4 5) 6 7

实例1：（求每行的和）with open("qqq") as f:    j = 0    for line in f:  # line 是字符串数据类型。        li = line.strip().split()  # split生成列表数据        sum = 0        for i in li:            sum += int(i)            # print(line,sum)        j += 1        print(j,sum)结果：1 152 203 25实例2：（求所有值的和）with open("qqq") as f:    sum = 0    for line in f:  # line 是字符串数据类型。        li = line.strip().split()  #split生成列表数据        for i in li:            sum += int(i)    print(sum)结果：60

Character matching

There is now a file grade, which reads as follows:

john 10 3  78 94 88andrea 20 90 75 90 86jasper 90 450 90 92 84sun 60 50 80 98 87

Print rows for the first column as Sun

awk  ‘{if($1 == "sun")print $0}‘ grade结果：sun 60 50 80 98 87

Prints rows with the first column longer than 4 and the second column greater than 40.

awk ‘{if (length($1) > 4 && $2 > 40) print $0}‘ gradejasper 90 450 90 92 84awk 不指定-F，就是默认空格或者table来分割。

Regular match: ~

awk  ‘{if($1 ~ "su")print $0}‘ grade结果：sun 60 50 80 98 8711. 匹配noarch字段，如果有，则显示整行$awk ‘$2 ~ /noarch/’ install.log12. 匹配不存在noarch字段的行，如果有，则显示整行$awk ‘$2 !~ /noarch/’ install.log

Process Control Two single quotes can have multiple {},if statements, such as the following

Example 1:/etc/passwd change root to admin, and the other is the list of common user.

awk -F: ‘{if ($1 == "root")    printf "%-15s: %s\n", $1,"Admin"else    print "%-15s: %s\n", $1,"Common User"}‘ /etc/passwd  

Example 2: How many ordinary users, UID > 1000

awk -F: -v sum=0 ‘{    # sum=0  这里不能有空格if ($3 > 1000)  sum++}END{  print sum  }‘  /etc/passwd结果：47

The process is as follows,

If process

If‘{if(条件)｛action｝}‘

For process

for‘{for (条件) {action}}‘

Exercise: Print all rows/values that last column is less than or equal to 6 (or use Python ....) )

with open("qqq") as f:    for line in f:        print(line.strip(),type(line.strip()))  # line in f 中 line 是字符串，strip（）去掉换行符。        li = line.strip().split()        a = int(li[-1])        if a > 6:            pass            # print(line)            # print(a)        else:            pass结果：1 2 3 4 5 <class ‘str‘>2 3 4 5 6 <class ‘str‘>3 4 5 6 7 <class ‘str‘>awk ‘{if ($NF <= 6) print $0}‘ qqq  #AWK一条命令解决

--------------------------------------awk Command format

The awk instruction consists of a combination of patterns, actions, or patterns and actions.

‘pattern {action}‘pattern 即模式，也可以理解为条件，你要找谁，高矮，胖瘦，等，都是条件。action：即动作，可以理解为干啥，找到人之后你要干嘛

awk Execution Process

a) awk屈辱第一行内容b) 判断是否符合模式中得条件，匹配则执行action，不匹配，继续读取下一行c) 继续读取下一行d) 直到读取到最后一行（EOF end of file）

Fields and records

Field fields, regions, fields $1,$2, etc., represent the first field, the second field, and so on.

Record records, which can be understood as a whole line

内置变量：RS == record seperator 即行得分隔符NR == number of record 行号ORS == output record seperator 输出时候得分隔符

Awk has a regular match.

^cool 匹配所有以cool开头得字符串cool$ 匹配所有以cool结尾得字符串a+b 匹配一个或多个a 加上b得行ifconfig eth0 |awk -F "[ :]+" ‘NR==2 {print $3 }‘ifconfig eth0 | awk -F "netmask|broadcast" ‘NR==2{print $3}‘怎么匹配1，不匹配到11：awk --posix -F: ‘$3~/1{2,}/{print $0}‘ test.awk    出现两次得可以出来，写成{1，} 也还是都出来了awk -F: ‘$3~/\<1\>/{print $0}‘ test.awk  \< 或 \b：词首锚定；用于单词模式的左侧；\> 或 \b：词尾锚定；用于单词模式的右侧；\<PATTERN\>：匹配整个单词；netstat -lantp | awk -F "[:/ ]+" ‘{print $5,"\t",$10}‘| sort 查询服务得端口

Begin and end modules detailed

找空行：grep -c ‘^$‘ test.awkawk -v a=0 ‘/^$/{a=a+1}END{print a}‘ test.awkawk -v a=0 -F: ‘$3>=5{a=a+1;(用;号分割)print a（a++）}END{print a}‘ test.awk

AWK, the Three Musketeers of Linux