Shulou(Shulou.com)06/02 Report--

This article analyzes "how to use awk" with you. The content is detailed and easy to understand. Friends who are interested in "how to use awk" can follow the editor's idea to read it slowly and deeply. I hope it will be helpful to you after reading. Let's learn more about "how to use awk" with the editor.

Awk print text content

Awk: Aho, Kernighan and Weinberger

The report generator looks up the text content under specific conditions and then displays it in a specific format

Format of the awk command:

# awk [option] 'script' file1 file2...

# awk [option] 'PATTERN {action}' file1 file2...

PATTERN:

Conditions described by text characters and regular expression metacharacters can be omitted

Action:

Print

Printf specifies the format of the output item; the format must be written

Option options:

-F specifies the text separator

Awk text processing mechanism:

Awk takes out the PATTERN-compliant text line by line and follows the specified delimiter (the default is blank

The-F option allows you to specify a separator) to split, and then output each segment according to a specific format

Output of awk:

1. Print

The format of print:

Print item1,item2,....

Note:

1. Items are separated by commas, while output is separated by white space characters

2. The output item can be a string, a numeric value, a field of the current record ($1), a variable, or an expression of awk. The numeric value is converted to a string and then output.

3. The item after the print command can be omitted, which is equivalent to print $0 ($0 represents the entire line of undivided text), so if you want to output a blank line, you need to use print "".

Segment with white space to display the first and second paragraphs of the text

# awk'{print $1 and 2} 'test.txt

This is

[root@shell ~] # df-hT | sed '1d' | awk' {print "Disk name:", $1, "Mount Point:", $7, "Total Size:", $3, "Free size", $5}'

Example:

Output 3 lines

# awk 'BEGIN {print "line one\ nline two\ nline three"}'

Line one

Line two

Line three

The user name and its uid in the output / etc/passwd

# awk-F:'{print $1 recorder 3}'/ etc/passwd

Awk variable

1. Record variables of awk built-in variables

FS: specifies the line delimiter to be used when reading text, which defaults to a blank character; equivalent to the-F option of awk

OFS: specifies the delimiter of the output, which defaults to a blank character

[root@localhost ~] # head-n 1 / etc/passwd | awk-F:'{print $1 recording 7}'

Root / bin/bash

[root@localhost ~] #

[root@localhost ~] # head-n 1 / etc/passwd | awk 'BEGIN {FS= ":"} {print $1 recording 7}'

Root / bin/bash

[root@localhost ~] # head-n 1 / etc/passwd | awk-F:'{print $1 recording 7}'

Root / bin/bash

[root@localhost ~] # head-n 1 / etc/passwd | awk-F: 'BEGIN {OFS= "- -"} {print $1 "7}'

Root---/bin/bash

[root@localhost ~] #

RS: specifies the newline character to be used when reading text (specify what character to use as the newline character); the default is newline character

ORS: specifies the line delimiter to use when outputting

2. Data variables of awk built-in variables

NR: record the number of lines of text processed by awk. If there are multiple files, all files will be counted together.

[root@localhost ~] # awk'{print ", NR," Line content: ", $0}'/ etc/hosts / etc/issue

Line 1 content: 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4

Line 2: 1 localhost localhost.localdomain localhost6 localhost6.localdomain6

Line 3: CentOS release 6.6 (Final)

Line 4: Kernel\ r on an\ m

Line 5:

Note:

Print does not need to use $when displaying variable values

FNR: record the number of lines of files being processed by awk. If there are multiple files, each file is counted separately.

[root@localhost ~] # awk'{print ", FNR," Line content: ", $0}'/ etc/hosts / etc/issue

Line 1 content: 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4

Line 2: 1 localhost localhost.localdomain localhost6 localhost6.localdomain6

Line 1 content: CentOS release 6.6 (Final)

Line 2: Kernel\ r on an\ m

Line 3:

NF: record the current line being processed by awk is separated into several fields

# cat test.txt

This is a test.

# awk'{print NF} 'test.txt

four

# awk'{print $NF} 'test.txt

Test.

[root@localhost ~] # awk-F:'{print "Number of line:", NF}'/ etc/passwd

[root@shell] # awk-F. '{print "Number of Line:", NF}' / etc/hosts

[root@shell ~] # awk 'BEGIN {FS= "."} {print "Number of Line:", NF}' / etc/hosts

Number of Line: 6

Number of Line: 3

Number of Line: 6

3. User-defined variables

Awk allows users to customize variables. Variable names cannot start with a number and are case-sensitive.

Example:

Method 1: use the-v option

[root@localhost ~] # head-n 3 / etc/passwd | awk-v test= "hello"-F:'{print test, $1}'

Hello root

Hello bin

Hello daemon

Method 2: define variables in BEGIN {} mode

[root@localhost ~] # head-n 3 / etc/passwd | awk-F: 'BEGIN {test= "hello"} {print test, $1}'

Hello root

Hello bin

Hello daemon

[root@localhost ~] #

Second, printf, format the output content

Format:

Printf format,item1,item2,...

Main points:

1. Specify the format format when printf is exported

2. Format is used to specify the format of each subsequent item output

3. The printf statement does not automatically print newline characters\ n

Format:

% c: display a single character

% dline% I: decimal integer

% eForce% E: display value by scientific counting method

% f: show floating point number

% gpeng% G: displays values in scientific counting format or floating point format

% s: display string

% u: unsigned integer

%%: show% itself

Modifier:

N: display width, N is a number

-: left alignment, default is right alignment

+: display numeric symbol

Example:

[root@localhost ~] # head-n 3 / etc/passwd | awk-F:'{printf "%-10s%-8s%-20s\ n", $1

Root / root / bin/bash

Bin/ bin/ sbin/nologin

Daemon / sbin/ sbin/nologin

[root@shell ~] # df-hT | sed '1d' | awk' {printf "%-15s%-10s\ n", $1m 7}'

/ dev/vda2 /

Tmpfs / dev/shm

/ dev/vda1 / boot

[root@shell ~] #

[root@shell ~] # awk-F:'{printf "%-15s%-10d%-25s%-15s\ n", $1 for 3 15s%-10d%-25s%-15s, 6 for 7}'/ etc/passwd

Root 0 / root / bin/bash

Bin 1 / bin/ sbin/nologin

Daemon 2 / sbin/ sbin/nologin

Adm 3 / var/adm / sbin/nologin

Lp 4 / var/spool/lpd / sbin/nologin

[root@node2 ~] # df-hT | sed '1d' | awk' {printf "%-10s%-10s%-12s%-10s%-15s%-6s%-8s%-4s\ n", "DiskName:", $1, "MountPoint:", $7, "TotalSize:", $3, "Usage:", $6}'

DiskName: / dev/sda3 MountPoint: / TotalSize: 77g Usage: 6

DiskName: tmpfs MountPoint: / dev/shm TotalSize: 491m Usage: 0

DiskName: / dev/sda1 MountPoint: / boot TotalSize: 190m Usage: 15

# awk [option] 'PATTERN {action}' file1 file2...

Action:

Print

Printf

Options:

-F

-v

PATTERN representation:

1. Regular expression with the format of / regex/

Display the first paragraph of the line starting with r in / etc/passwd with a colon as the delimiter

# awk-F:'/ ^ r / {print $1}'/ etc/passwd

Root

Rpc

Rtkit

Rpcuser

[root@localhost ~] # awk'/ ^ (r | s) / {print $0}'/ etc/passwd

[root@localhost ~] # awk'/ ^ [rs] / {print $0}'/ etc/passwd

[root@localhost ~] # ls-l / tmp/ | awk'/ ^ d / {print $9}'

[root@localhost ~] # netstat-antp | awk'/ ^ tcp/ {print $0}'

[root@localhost ~] # netstat-antp | sed '1mai 2d'

2. Expression, an expression consisting of the following operators

Operators for awk

1. Arithmetic operator

-x negative value

+ x is converted to numeric value, positive value

X ^ y, x ^ 3 to the power of x ^ y

Xfanty

X/y

Xfanty

XMury

X% y

2. String operator

+: implement string concatenation "ab" + "cd" abcd "ab" + "12" ab12

3. Assignment operator

=

+ = axiomb a=a+b a+=2a=a+2

-=

* =

/ =

% =

^ = x ^ = y x ^ y

* =

+ + x++x=x+1

--

4. Comparison operator

X

< y x y x >

= y

X = y

X! = y

X ~ y x is a string, y is a pattern, true if x can be matched by a pattern, otherwise false "abc" ~ ^ a

X! ~ y

5. Logical relation character

& &

| | |

Displays user names and uid with uid greater than or equal to 500

# awk-F:'$3 > = 500 {print $1

Nfsnobody 65534

Wjc 500

Displays the user name and shell name with the default shell of / bin/bash

# awk-F:'$7 million = "/ bin/bash" {print $1 recorder 7}'/ etc/passwd

Or

# awk-F:'$7 ~ "bash$" {print $1 recording 7}'/ etc/passwd

Root / bin/bash

Mysql / bin/bash

Wjc / bin/bash

[root@node2 ~] # netstat-antp | awk'$6pm = "ESTABLISHED" {print $0}'

Tcp 0 52 192.168.87.102:22 192.168.87.1:49541 ESTABLISHED 1940/sshd

Tcp 0 0 192.168.87.102:22 192.168.87.1:49650 ESTABLISHED 2135/sshd

[root@node2 ~] #

[root@localhost ~] # df-hT | awk'+ $6 > 10 {print $1pm NF}'

/ dev/sda1 / boot

/ dev/sr0 / mnt

3. Specify a range in the format of pattern1,pattern2

Use colons as delimiters to display the user names, uid and shell of all lines in the middle between uid=0 and the last field ending with nologin

# awk-F:'$3pm / 7 ~ "nologin$" {print $1 / etc/passwd

Root 0 / bin/bash

Bin 1 / sbin/nologin

Sample formatted output

# awk-F:'$3 printf 7 ~ "nologin$" {printf "%-10s%-10s%-20s\ n" / etc/passwd

Root 0 / bin/bash

Bin 1 / sbin/nologin

4. BEGIN/END, special mode

BEGIN indicates that awk performs an operation before processing

END indicates that awk performs an operation before the end of the last line

Print the header using BEGIN

# awk-F: 'BEGIN {printf "%-10s%-10s%-20s\ n", "Username", "Uid", "Shell"} $3 million 0 7 ~ "nologin$" {printf "%-10s%-10s%-10s\ n", $1 penny 3 printf 7} / etc/passwd

Username Uid Shell

Root 0 / bin/bash

Bin 1 / sbin/nologin

Use END to print the footer

# awk-F: 'BEGIN {printf "%-10s%-10s%-20s\ n", "Username", "Uid", "Shell"} $3 million 0END OF File... 7 ~ "nologin$" {printf "%-10s%-10s%-10s\ n", $1 END {print "END OF File..."}' / etc/passwd

Username Uid Shell

Root 0 / bin/bash

Bin 1 / sbin/nologin

END OF File...

# awk [option] 'PATTERN {action}' file1 file2...

Logic control statement

1 、 if...else

Format:

If (conditional) {statement; statement} else {statement 1; statement 2}

If statement has only one statement, {} may not write

[root@localhost ~] # awk-F:'{if ($3 million / 0) {print $1, "is administrator."}'/ etc/passwd

Use the colon as the separator to determine the first field. If it is root, the user name Admin is displayed, otherwise the user name Common User is displayed.

# awk-F:'{if ($1percent = "root") print $1, "Admin"; else print $1, "Common User"}'/ etc/passwd

Root Admin

Bin Common User

Daemon Common User

Adm Common User

Lp Common User

Sync Common User

Shutdown Common User

Formatted output

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

Root: Admin

Bin: Common User

Daemon: Common User

Adm: Common User

Lp: Common User

Sync: Common User

Shutdown: Common User

Halt: Common User

Statistics on the number of users of the system

[root@localhost] # awk-F:'{if ($3 > 0 & & $350) sum++} END {print "The number of user:", sum}'/ etc/passwd

The number of user: 16

Remove those with disk utilization greater than 10%

# df-h | sed '1d' | awk' {if (+ $5 > 10) {print $0}}'

/ dev/sda1 194M 26M 158m 15% / boot

/ dev/sr0 2.9G 2.9G 0100% / mnt

[root@localhost ~] # awk-F:'{if ($3 million) {print $1} else {print $7}}'/ etc/passwd

[root@localhost ~] # awk-F:'{if ($3 million / 0) {count++} else {ionization +}} END {print "Administrative users:", count, "number of ordinary users:", I}'/ etc/passwd

[root@localhost ~] # awk-F:'/ bash$/ | | / nologin$/ {if ($7clients = "/ bin/bash") {print +} else {print "bash users:", I, "number of nologin users:", j}'/ etc/passwd

Number of bash users: 38 nologin users: 28

2 、 while

Format:

While (condition) {statement 1; statement 2.}

Use colons as delimiters to determine that if the length of each field in each line is greater than 4, it is displayed.

# awk-F:'{iposit while (iTun4) {print $I}; iTunes +}}'/ etc/passwd

Count the words of length 5 in test.txt files

[root@localhost] # awk'{iposit 1; while (i5) {print $I}; iTunes +}} 'test.txt

3. For traversal array

Format:

For (variable definition; condition under which the loop terminates; statement that changes the condition of the loop) {statement; statement;.}

For (iTunes 1)

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.