print 用于输出想要的内容，类似 echo

awk --profile -v blog="$a" 'BEGIN{print blog}'
# 默认当前路径会生成awkprof.out文件，用来保存格式化后的awk命令

$ tail /etc/services |awk 'BEGIN{print "Service\t\tPort\t\t\tDescription\n==="}{print $0}'
Service         Port                    Description
aigairserver    21221/tcp               # Services for Air Server
ka-kdp          31016/udp               # Kollective Agent Kollective Delivery
ka-sddp         31016/tcp               # Kollective Agent Secure Distributed Delivery
edi_service     34567/udp               # dhanalakshmi.org EDI Service
axio-disc       35100/tcp               # Axiomatic discovery protocol
axio-disc       35100/udp               # Axiomatic discovery protocol
pmwebapi        44323/tcp               # Performance Co-Pilot client HTTP API
cloudcheck-ping 45514/udp               # ASSIA CloudCheck WiFi Management keepalive
cloudcheck      45514/tcp               # ASSIA CloudCheck WiFi Management System
spremotetablet  46998/tcp               # Capture handwritten signatures

END模式是在awk处理文件之后执行的操作

$ tail /etc/services |awk --profile 'BEGIN{print "Service\t\tPort\t\t\tDescription\n==="}{print $0}END{print "===\nEND......"}'
Service         Port                    Description
aigairserver    21221/tcp               # Services for Air Server
ka-kdp          31016/udp               # Kollective Agent Kollective Delivery
ka-sddp         31016/tcp               # Kollective Agent Secure Distributed Delivery
edi_service     34567/udp               # dhanalakshmi.org EDI Service
axio-disc       35100/tcp               # Axiomatic discovery protocol
axio-disc       35100/udp               # Axiomatic discovery protocol
pmwebapi        44323/tcp               # Performance Co-Pilot client HTTP API
cloudcheck-ping 45514/udp               # ASSIA CloudCheck WiFi Management keepalive
cloudcheck      45514/tcp               # ASSIA CloudCheck WiFi Management System
spremotetablet  46998/tcp               # Capture handwritten signatures
END......

# 输出匹配包含tcp的行
tail /etc/services | awk '/tcp/{print $0}'
# 匹配以a开头的行
tail /etc/services | awk '/^a/{print $0}'
# 匹配开头是6个字符的行
tail /etc/services | awk '/^[a-z0-9-]{6} /{print $0}'
# 匹配以Kollective Delivery结尾的行
tail /etc/services | awk '/(Kollective Delivery)$/{print $0}'

$ cat awk.txt 
Remember Kruschev:  he tried to do too many things too fast, and he was
removed in disgrace.  If Gorbachev tries to destroy the system or make too
many fundamental changes to it, I believe the system will get rid of him.
I am not a political scientist, but I understand the system very well.
I believe he will have a "heart attack" or retire or be removed.  He is
up against a brick wall.  If you think they will change everything and
become a free, open society, forget it!
                -- Victor Belenko, MiG-25 fighter pilot who defected in 1976
                        "Defense Electronics", Vol 20, No. 6, pg. 110

匹配过程与sed类似，以行为单位进行匹配

# 匹配文本中有Rem的行作为起始行，匹配文本中有I的行作为结束行
$ awk '/Rem/,/I/' awk.txt
Remember Kruschev:  he tried to do too many things too fast, and he was
removed in disgrace.  If Gorbachev tries to destroy the system or make too

FS: FS 与 -F 选项功能一样，在awk处理文本之前设置 FS 变量

$ tail -n3 /etc/passwd | awk 'BEGIN{FS=":"}{print $1,$2}'
ssjinyao x
redis x
lei x
$ tail -n3 /etc/passwd | awk -F':' '{print $1,$2}'
ssjinyao x
redis x
lei x
$ tail -n3 /etc/passwd | awk -v FS=':' '{print $1,$2}'
ssjinyao x
redis x
lei x

OFS: 上面的例子中默认输出的是 lei x 中间以空格分隔，这个空格就是OFS的值

$ tail -n3 /etc/passwd | awk 'BEGIN{FS=":";OFS="-"}{print $1,$2}'
ssjinyao-x
redis-x
lei-x
$ tail -n3 /etc/passwd | awk -F':' '{print $1"-"$2}'
ssjinyao-x
redis-x
lei-x

RS: RS默认以 \n 分隔每段值

$ echo "www.feiyiblog.com/user/test.html" | awk 'BEGIN{FS="/"}{print $0}'
www.feiyiblog.com/user/test.html
$ echo "www.feiyiblog.com/user/test.html" | awk 'BEGIN{RS="/"}{print $0}'
www.feiyiblog.com
test.html
# 使用正则
$ seq -f "str%02g" 10 | sed 'n;n;a\-----'
str01
str02
str03
-----
str04
str05
str06
-----
str07
str08
str09
-----
str10
$ seq -f "str%02g" 10 | sed 'n;n;a\-----' | awk 'BEGIN{RS="-+"}{print $1}'
str01
str04
str07
str10

ORS: 默认以 \n 作为分隔输出每行内容

$ seq 10 |awk 'BEGIN{ORS="-"}{print $0}'
1-2-3-4-5-6-7-8-9-10-
$ tail -n3 /etc/passwd | awk 'BEGIN{FS=":";OFS="-";ORS=":"}{print $1,$2}'
ssjinyao-x:redis-x:lei-x:
# 默认ORS的原本输出内容为
ssjinyao-x
redis-x
lei-x

$ echo "abcdef" | awk 'BEGIN{FS=""}{print NF}'
# 示例中NF的值为6，也可以直接引用该值来输出最后一个字段 {print $NF}={print $6}
$ echo "abcdef" | awk 'BEGIN{FS=""}{print $NF}'
# 也可以使用NF进行运算来输出指定字段
$ echo "abcdef" | awk 'BEGIN{FS=""}{print $(NF-1)}'
# 也可以直接给字段赋值，不输出最后2个字段
$ echo "abcdef" | awk 'BEGIN{FS=""}{$NF="";$(NF-1)="";print $0}'
a b c d 

$ tail -n4 /etc/services | awk '{print NR,$0}'
1 pmwebapi        44323/tcp               # Performance Co-Pilot client HTTP API
2 cloudcheck-ping 45514/udp               # ASSIA CloudCheck WiFi Management keepalive
3 cloudcheck      45514/tcp               # ASSIA CloudCheck WiFi Management System
4 spremotetablet  46998/tcp               # Capture handwritten signatures
# 输出第3行，第1个字段
$ tail -n4 /etc/services | awk 'NR==3{print $1}'
cloudcheck
# 输出总行数
$ tail -n4 /etc/services | awk 'END{print NR}'
# 输出前2行
$  tail -n4 /etc/services | awk 'NR<=2{print $0}'
pmwebapi        44323/tcp               # Performance Co-Pilot client HTTP API
cloudcheck-ping 45514/udp               # ASSIA CloudCheck WiFi Management keepalive

FNR: awk处理2个文件时，NR会将多个文件的行进行统一编号，而FNR在处理第二个文件时，编号重新计数。

场景：当 FNR==NR 时，说明在处理第一个文件内容，不等于时说明在处理第二个文件内容。

$ cat a 
$ cat b 
$ awk '{print NR,FNR,$0}' a b 
1 1 a 
2 2 b 
3 3 c 
4 1 c 
5 2 d 
6 3 e

处理2个文件时，进行判断