Shell正则表达式与文本处理三剑客(grep、sed、awk)
一、正则表达式
Shell正则表达式分为两种:
基础正则表达式:BRE(basic regular express)
扩展正则表达式:ERE(extend regular express),扩展的表达式有+、?、|和()
1.1 基本正则表达式
1.2 扩展正则表达式
1.3 支持正则表达式的工具
grep:默认不支持扩展表达式,加-E选项开启ERE。如果不加-E使用花括号要加转义符\{\}
egrep:支持基础和扩展表达式
awk:支持egrep所有的正则表达式
sed:默认不支持扩展表达式,加-r选项开启ERE。如果不加-r使用花括号要加转义符\{\}
二、grep
过滤来自一个文件或标准输入匹配模式内容。
除此之外还有:
grep -f:从文件每一行获取匹配模式
grep -m:输出匹配的结果num数
grep -H:打印每个匹配的文件名
grep -h:不输出文件名
grep -q:不输出正常信息
grep -s:不输出错误信息
grep -r:递归目录
grep -B:打印匹配的前几行
grep -A:打印匹配的后几行
grep -C:打印匹配的前后几行
grep --color:匹配的字体颜色
(1)输出b文件在a文件相同的行
> grep -f a b
(2)输出b文件在a文件不同的行
> grep -v -f a b
(3)匹配多个模式
> echo "a bc de" |xargs -n1 |grep -e 'a' -e 'bc'
a
bc
注:xargs为多行输出,-n1表示一行输出1个参数
(4)去除文件内的空行和#开头的行(一般用于查看配置文件)
> grep -E -v "^$|.*#" /etc/httpd/conf/httpd.conf
(5)匹配开头不分大小写的单词
> echo "A a b c" |xargs -n1 |grep -i a
A
a
(6)只显示匹配的字符串
> echo "this is a test" |grep -o 'is'
is
is
(7)输出匹配的前五个结果
> seq 1 20 |grep -m 5 -E '[0-9]{2}'
10
11
12
13
14
(8)统计匹配多少行
> seq 1 20 |grep -c -E '[0-9]{2}'
11
(9)匹配b字符开头的行
> echo "a bc de" |xargs -n1 |grep '^b'
bc
(10)匹配de字符结尾的行并输出匹配的行
> echo "a ab abc abcd abcde" |xargs -n1 |grep -n 'de$'
5:abcde
(11)递归搜索/etc目录下包含ip的conf后缀文件
grep --include:只检索匹配的文件
> grep -r '192.167.1.1' /etc --include *.conf
(12)排除搜索bak后缀的文件
grep --exclude:跳过匹配的文件
> grep -r '192.167.1.1' /opt --exclude *.bak
(13)匹配所有IP
> ifconfig |grep -E -o "[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}"
(14)打印匹配结果及后3行
> seq 1 10 |grep 5 -A 3
5
6
7
8
(15)打印匹配结果及前3行
> seq 1 10 |grep 5 -B 3
2
3
4
(16)打印匹配结果及前后3行
> seq 1 10 |grep 5 -C 3
2
3
4
5
6
7
8
三、sed
流编辑器,过滤和替换文本。
工作原理:sed命令将当前处理的行读入模式空间进行处理,处理完把结果输出,并清空模式空间。然后再将下一行读入模式空间进行处理输出,以此类推,直到最后一行。还有一个空间叫保持空间,又称暂存空间,可以暂时存放一些处理的数据,但不能直接输出,只能放到模式空间输出。
这两个空间其实就是在内存中初始化的一个内存区域,存放正在处理的数据和临时存放的数据。
语法格式:sed [选项] '地址 命令' file
借助一些文本内容作为示例:
[root@openEuler-1 script]# cat sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
3.1 匹配打印(p)
(1)打印匹配blp5开头的行
[root@openEuler-1 script]# sed -n '/^blp5/p' sed.txt
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
(2)打印第一行
[root@openEuler-1 script]# sed -n '1p' sed.txt
nimgtw 48003/udp # Nimbus Gateway
(3)打印第一行至第三行
[root@openEuler-1 script]# sed -n '1,3p' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
(4)打印奇数行
[root@openEuler-1 script]# seq 10 | sed -n '1~2p'
1
3
5
7
9
(5)打印匹配行及后一行
[root@openEuler-1 script]# sed -n '/blp5/,+1p' sed.txt
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
(6)打印最后一行
[root@openEuler-1 script]# sed -n '$p' sed.txt
iqobject 48619/udp # iqobject
(7)不打印最后一行
感叹号也就是对后面的命令取反。
[root@openEuler-1 script]# sed -n '$!p' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
(8)匹配范围blp5开头的行~com开头的行
以逗号分开两个样式选择某个范围。
[root@openEuler-1 script]# sed -n '/^blp5/,/^com/p' sed.txt
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
3.2 匹配删除(d)
删除与打印使用方法类似,打印是把匹配的打印出来,删除是把匹配的删除,删除只是不用-n选项。
(1)删除包含blp5的行
[root@openEuler-1 script]# sed '/^blp5/d' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(2)删除1~3行
[root@openEuler-1 script]# sed '1,3d' sed.txt
isnetserv 48128/udp # Image Systems Network Services
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(3)去除http.conf文件空行和开头#号的行
[root@openEuler-1 script]# sed '/^$/d;/.*#/d' /etc/httpd/conf/httpd.conf
3.3 替换(s///)
(1)替换blp5字符串为test
[root@openEuler-1 script]# sed 's/blp5/test/' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
test 48129/tcp # Bloomberg locator
test 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(2)替换开头是blp5的字符串并打印
[root@openEuler-1 script]# sed -n 's/blp5/test/p' sed.txt
test 48129/tcp # Bloomberg locator
test 48129/udp # Bloomberg locator
(3)使用&命令引用匹配内容并替换
[root@openEuler-1 script]# sed 's/48049/&.0/' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049.0/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(4)给IP加引号
g:全局替换
[root@openEuler-1 script]# echo "192.168.121.11 172.1.1.2 223.5.5.5" | sed -r 's/[^ ]+/“&”/g'
“192.168.121.11” “172.1.1.2” “223.5.5.5”
(5)对1-5行的blp5进行替换
[root@openEuler-1 script]# sed '1,5s/blp5/test/' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
test 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(6)对匹配行进行替换
[root@openEuler-1 script]# sed '/48129\/tcp/s/blp5/test/' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
test 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(7)二次匹配替换
[root@openEuler-1 script]# sed 's/blp5/test/;s/3g/4g/' sed.txt
nimgtw 48003/udp # Nimbus Gateway
4gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
test 48129/tcp # Bloomberg locator
test 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(8)注释匹配行后的多少行
[root@openEuler-1 script]# seq 10 | sed '/5/,+3s/^/#/'
1
2
3
4
#5
#6
#7
#8
9
10
3.4 多重编辑(-e)
[root@openEuler-1 script]# sed -e '1,2d' -e 's/blp5/test/' sed.txt
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
test 48129/tcp # Bloomberg locator
test 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
# 也可以使用;分隔
[root@openEuler-1 script]# sed '1,2d;s/blp5/test/' sed.txt
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
test 48129/tcp # Bloomberg locator
test 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
3.5 添加新内容(a,i、c)
(1)在blp5上一行添加test
[root@openEuler-1 script]# sed '/blp5/i \test' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
test
blp5 48129/tcp # Bloomberg locator
test
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(2)在blp5下一行添加test
[root@openEuler-1 script]# sed '/blp5/a \test' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
blp5 48129/tcp # Bloomberg locator
test
blp5 48129/udp # Bloomberg locator
test
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
(3)将blp5替换新行
[root@openEuler-1 script]# sed '/blp5/c \test' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
test
test
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
3.6 读取文件并追加到匹配行后(r)
[root@openEuler-1 script]# cat a.txt
123
456
[root@openEuler-1 script]# sed '/blp5/r a.txt' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
blp5 48129/tcp # Bloomberg locator
123
456
blp5 48129/udp # Bloomberg locator
123
456
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
3.7 将匹配行写入文件(w)
[root@openEuler-1 script]# sed '/blp5/w b.txt' sed.txt
nimgtw 48003/udp # Nimbus Gateway
3gpp-cbsp 48049/tcp # 3GPP Cell Broadcast Service Protocol
isnetserv 48128/tcp # Image Systems Network Services
isnetserv 48128/udp # Image Systems Network Services
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
com-bardac-dw 48556/tcp # com-bardac-dw
com-bardac-dw 48556/udp # com-bardac-dw
iqobject 48619/tcp # iqobject
iqobject 48619/udp # iqobject
[root@openEuler-1 script]# cat b.txt
blp5 48129/tcp # Bloomberg locator
blp5 48129/udp # Bloomberg locator
3.8 读取下一行(n和N)
n 读取下一行到模式空间。
N 追加下一行内容到模式空间,并以换行符\n分隔。
(1)打印匹配的下一行
[root@openEuler-1 script]# seq 5 |sed -n '/3/{n;p}'
4
(2)打印偶数
[root@openEuler-1 script]# seq 10 | sed -n 'n;p'
2
4
6
8
10
# sed先读取第一行1,执行n命令,获取下一行2,此时模式空间是2,
# 执行p命令,打印模式空间。 现在模式空间是2,sed再读取3,
# 执行n命令,获取下一行4,此时模式空间为4,执行p命令,以此类推.
(3)打印奇数
[root@openEuler-1 script]# seq 10 |sed 'n;d'
1
3
5
7
9
# sed先读取第一行1,此时模式空间是1,并打印模式空间1,
# 执行n命令,获取下一行2,执行d命令,删除模式空间的2,sed再读取3,
# 此时模式空间是3,并打印模式空间,再执行n命令,获取下一行4,
# 执行d命令,删除模式空间的3,以此类推.
3.9 打印和删除模式空间的第一行(P和D)
P 打印模式空间的第一行。
D 删除模式空间的第一行。
(1)打印奇数
[root@openEuler-1 script]# seq 6 | sed -n 'N;P'
1
3
5
(2)保留最后一行
[root@openEuler-1 script]# seq 6 | sed 'N;D'
6
# 读取第一行1,执行N命令读取下一行并追加到模式空间,
# 此时模式空间是1\n2,执行D命令删除模式空间第一行1,剩余2.
# 读取第二行,执行N命令,此时模式空间是3\n4,
# 执行D命令删除模式空间第一行3,剩余4,以此类推.
3.10 保持空间操作(h与H、g与G和x)
h 复制模式空间内容到保持空间(覆盖)。
H 复制模式空间内容追加到保持空间。
g 复制保持空间内容到模式空间(覆盖)。
G 复制保持空间内容追加到模式空间。
x 模式空间与保持空间内容互换。
(1)将匹配的内容覆盖到另一个匹配
[root@openEuler-1 script]# seq 6 | sed -e '/3/{h;d}' -e '/5/g'
1
2
4
3
6
# h命令把匹配的3复制到保持空间,d命令删除模式空间的3.
# 后面命令再对模式空间匹配5,并用g命令把保持空间3覆盖模式空间5.
(2)将匹配的内容放到最后
[root@openEuler-1 script]# seq 6 | sed -e '/3/{h;d}' -e '$G'
1
2
4
5
6
3
# 这里的$代表最后一行,G表示将保持空间的内容追加到当前行的后面
(3)交换模式空间和保持空间
[root@openEuler-1 script]# seq 6 | sed -e '/3/{h;d}' -e '/5/x' -e '$G'
1
2
4
3
6
5
# 在模式空间匹配5并将保持空间的3与5交换,5就变成了3.
# 最后把保持空间的5追加到模式空间.
(4)倒序输出
[root@openEuler-1 script]# seq 5 |sed '1!G;h;$!d'
5
4
3
2
1
# 1!G 第一行不执行把保持空间内容追加到模式空间,因为现在保持空间还没有数据
# h 将模式空间放到保持空间暂存
# $!d 最后一行不执行删除模式空间的内容。
(5)每行后面添加新空行
[root@openEuler-1 script]# seq 5 |sed G
1
2
3
4
5
3.11 忽略大小写匹配(I)
[root@openEuler-1 script]# echo -e "a\nA\nb\nc" |sed 's/a/1/Ig'
1
1
b
c
3.12 获取总行数
[root@openEuler-1 script]# seq 10 |sed -n '$='
10
四、awk
awk 是一个处理文本的编程语言工具,能用简短的程序处理标准输入或文件、数据排序、计算以及 生成报表等。
基本的命令语法:awk option 'pattern {action}' file
其中pattern表示AWK在数据中查找的内容,而action是在找到匹配内容时所执行的一系列命令。花括号用于根据特定的模式对一系列指令进行分组。
awk 处理的工作方式与数据库类似,支持对记录和字段处理,这也是grep和sed不能实现的。在awk中,缺省的情况下将文本文件中的一行视为一个记录,逐行放到内存中处理,而将一行中的某一部分作为记录中的一个字段。用1,2,3...数字的方式顺序的表示行(记录)中的不同字段。用$后跟数字,引用对应的字段,以逗号分隔,0表示整个行。
4.1 选项、模式
(1)从文件读取awk程序处理文件
[root@openEuler-1 script]# cat test.awk
{print $2}
[root@openEuler-1 script]# tail -n3 /etc/services | awk -f test.awk
45514/udp
45514/tcp
46998/tcp
(2)指定分隔符,打印指定字段(默认以空格分割)
[root@openEuler-1 script]# tail -n3 /etc/passwd | awk -F':' '{print $1}'
apache
mysql
nginx
(3)指定多个分隔符作为同一个分隔符处理
[root@openEuler-1 script]# tail -n3 /etc/services
cloudcheck-ping 45514/udp # ASSIA CloudCheck WiFi Management keepalive
cloudcheck 45514/tcp # ASSIA CloudCheck WiFi Management System
spremotetablet 46998/tcp # Capture handwritten signatures
[root@openEuler-1 script]# tail -n3 /etc/services | awk -F'[/#]' '{print $1}'
cloudcheck-ping 45514
cloudcheck 45514
spremotetablet 46998
[root@openEuler-1 script]# tail -n3 /etc/services | awk -F'[/#]' '{print $2}'
udp
tcp
tcp
[root@openEuler-1 script]# tail -n3 /etc/services | awk -F'[/#]' '{print $3}'
ASSIA CloudCheck WiFi Management keepalive
ASSIA CloudCheck WiFi Management System
Capture handwritten signatures
(4)变量赋值
[root@openEuler-1 script]# awk -v a=123 'BEGIN{print a}'
123
(5)输出awk全局变量到文件
[root@openEuler-1 script]# seq 5 | awk --dump-variables '{print $0}'
1
2
3
4
5
[root@openEuler-1 script]# cat awkvars.out
ARGC: 1
ARGIND: 0
ARGV: array, 1 elements
BINMODE: 0
CONVFMT: "%.6g"
ENVIRON: array, 25 elements
ERRNO: ""
FIELDWIDTHS: ""
FILENAME: "-"
FNR: 5
FPAT: "[^[:space:]]+"
FS: " "
FUNCTAB: array, 41 elements
IGNORECASE: 0
LINT: 0
NF: 1
NR: 5
OFMT: "%.6g"
OFS: " "
ORS: "\n"
PREC: 53
PROCINFO: array, 21 elements
RLENGTH: 0
ROUNDMODE: "N"
RS: "\n"
RSTART: 0
RT: "\n"
SUBSEP: "\034"
SYMTAB: array, 28 elements
TEXTDOMAIN: "messages"
(6)BEGIN和END
BEGIN模式是在处理文件之前执行该操作,常用于修改内置变量、变量赋值和打印输出的页眉或标 题。END模式是在程序处理完才会执行。
示例:
[root@openEuler-1 script]# cat frequent_ip.sh
#!/bin/bash
#########################
#File name:frequent_ip.sh
#Email:obboda@163.com
#Created time:2025-01-14 11:46:11
#Description:访问nginx日志,得到访问ip最多的前10个
#########################
log_path="/var/log/nginx/access.log"
# 拿取按照访问次数排序过的ip
sort_data=`awk '{print $1}' /var/log/nginx/access.log | sort | uniq -c`
# 按照sort_data的数据进行优化处理,打印输出
echo "$sort_data" | awk '
BEGIN{printf "%-10s %-10s %-10s \n","访问排名","访问IP","访问次数"}
{printf "%-10s %-20s %-10s \n",NR,$2,$1}
END{print "end............"}
'
[root@openEuler-1 script]# bash frequent_ip.sh
访问排名 访问IP 访问次数
1 192.168.121.1 8
2 192.168.121.12 3
3 192.168.121.13 2
4 192.168.121.131 1
5 192.168.121.51 4
end............
4.2 内置变量
(1)FS和OFS
在程序开始前重新赋值FS变量,改变默认分隔符为冒号,与-F一样。
[root@openEuler-1 script]# head -n5 /etc/passwd | awk 'BEGIN{FS=":"}{print $1,$2}'
root x
bin x
daemon x
adm x
lp x
# 也可以使用-v来重新赋值这个变量,并指定输出分隔符
[root@openEuler-1 script]# head -n5 /etc/passwd | awk -vFS=':' '{print $1,$2}'
root x
bin x
daemon x
adm x
lp x
# 也可以指定输出分隔符
[root@openEuler-1 script]# head -n5 /etc/passwd | awk 'BEGIN{FS=":";OFS="#"}{print $1,$2}'
root#x
bin#x
daemon#x
adm#x
lp#x
(2)RS和ORS
# 指定以某个字符作为分隔符来处理记录:
[root@openEuler-1 script]# echo "www.baidu.com/user/test.html" |awk 'BEGIN{RS="/"}{print $0}'
www.baidu.com
user
test.html
# 将输出的换行符替换为+号:
[root@openEuler-1 script]# seq 10 | awk 'BEGIN{ORS="+"}{print $0}'
1+2+3+4+5+6+7+8+9+10+[root@openEuler-1 script]#
(3)NF
# NF是字段个数
[root@openEuler-1 script]# echo "a b c d e f" | awk '{print NF}'
6
# 打印最后一个字段
[root@openEuler-1 script]# echo "a b c d e f" | awk '{print $NF}'
f
(4)NR和FNR
NR统计记录编号,每处理一行记录,编号就会+1,FNR不同的是在统计第二个文件时会重新计数。
[root@openEuler-1 script]# tail -n5 /etc/services | awk '{print NR,$0}'
1 axio-disc 35100/udp # Axiomatic discovery protocol
2 pmwebapi 44323/tcp # Performance Co-Pilot client HTTP API
3 cloudcheck-ping 45514/udp # ASSIA CloudCheck WiFi Management keepalive
4 cloudcheck 45514/tcp # ASSIA CloudCheck WiFi Management System
5 spremotetablet 46998/tcp # Capture handwritten signatures
NR和FNR的区别:
[root@openEuler-1 script]# awk '{print NR,FNR,$0}' a b
1 1 a
2 2 b
3 3 c
4 1 d
5 2 e
6 3 f
(5)ARGC和ARGV
ARGC是命令行参数数量
ARGV是将命令行参数存到数组,元素由ARGC指定,数组下标从0开始
[root@openEuler-1 script]# awk 'BEGIN{print ARGC}' 1 2 3
4
[root@openEuler-1 script]# awk 'BEGIN{print ARGV[0],ARGV[1]}' 1 2 3
awk 1
(6)ARGIND
ARGIND是当前正在处理的文件索引值,第一个文件是1,第二个文件是2,以此类推,从而可以通过这种方式判断正在处理哪个文件。
[root@openEuler-1 script]# awk '{print ARGIND,$0}' a b
1 a
1 b
1 c
2 d
2 e
2 f
(7)ENVIRON
ENVIRON调用系统变量。
[root@openEuler-1 script]# awk 'BEGIN{print ENVIRON["HOME"]}'
/root
(8)FILENAME
FILENAME是当前处理文件的文件名。
[root@openEuler-1 script]# awk '{print FILENAME,$0}' a b
a a
a b
a c
b d
b e
b f
(9)IGNORECASE
IGNORECASE=1表示忽略大小写
[root@openEuler-1 script]# echo "A a b c" |xargs -n1 |awk 'BEGIN{IGNORECASE=1}/a/'
A
a
4.3 操作符
注意:在awk中,有3种情况表达式为假:数字是0,空字符串和未定义的值。且数值运算,未定义变量初始值为0。字符运算,未定义变量初始值为空。
(1)截取整数
[root@openEuler-1 script]# echo "123abc abc123 123abc123" |xargs -n1 | awk '{print +$0}'
123
0
123
[root@openEuler-1 script]# echo "123abc abc123 123abc123" |xargs -n1 | awk '{print -$0}'
-123
0
-123
(2)打印奇数偶数行
# 打印奇数行
[root@openEuler-1 script]# seq 6 | awk 'i=!i'
1
3
5
# 打印偶数行
[root@openEuler-1 script]# seq 6 | awk '!(i=!i)'
2
4
6
(3)管道符的使用
[root@openEuler-1 script]# seq 5 | shuf | awk '{print $0|"sort"}'
1
2
3
4
5
# 其中shuf会将原来的顺序打乱
(4)三目运算符
[root@openEuler-1 script]# awk 'BEGIN{print 1==1?"yes":"no"}'
yes
4.4 流程控制
(1)if语句
格式:if (condition) statement [ else statement ]
# 单分支
[root@openEuler-1 script]# seq 5 | awk '{if($0==3)print $0}'
3
# 双分支
[root@openEuler-1 script]# seq 5 |awk '{if($0==3)print $0;else print "no"}'
no
no
3
no
no
(2)while语句
格式:while (condition) statement
[root@openEuler-1 script]# echo "1 2 3 4 5" | awk '{i=1;while(i<=NF){print $i;i++}}'
1
2
3
4
5
(3)for语句C语言风格
格式:for (expr1; expr2; expr3) statement
[root@openEuler-1 script]# cat file
1 2 3
4 5 6
7 8 9
[root@openEuler-1 script]# awk '{for(i=1;i<=NF;i++)print $i}' file
1
2
3
4
5
6
7
8
9
(4)for语句遍历数组
格式:for (var in array) statement
[root@openEuler-1 script]# seq -f "str%.g" 5 |awk '{a[NR]=$0}END{for(v in a)print v,a[v]}'
1 str1
2 str2
3 str3
4 str4
5 str5
(5)break和continue语句
break跳过所有循环,continue跳过当前循环。
[root@openEuler-1 script]# awk 'BEGIN{for(i=1;i<=5;i++){if(i==3){break};print i}}'
1
2
[root@openEuler-1 script]# awk 'BEGIN{for(i=1;i<=5;i++){if(i==3){continue};print i}}'
1
2
4
5
(6)exit语句
格式:exit [ expression ]
exit退出程序,与shell的exit一样。[ expr ]是0-255之间的数字。
[root@openEuler-1 script]# seq 5 |awk '{if($0~/3/)exit (123)}'
[root@openEuler-1 script]# echo $?
123
4.5 内置函数
[root@openEuler-1 script]# cat c
123abc
abc123
123abc123
[root@openEuler-1 script]# awk '{print int($0)}' c
123
0
123
4.6 I/O语句
# 获取匹配行的下一行
[root@openEuler-1 script]# seq 5 |awk '/3/{getline;print}'
4
4.7 printf语句
格式化输出,默认打印字符串不换行。
格式:printf [format] arguments
# 左对齐宽度10
[root@openEuler-1 script]# awk 'BEGIN{printf "%-10s %-10s %-10s\n","ID","Name","Passwd"}'
ID Name Passwd