Regular expressions in the shell

Regular expressions in the shell

Vbird said that learning Linux, mastering the shell and the regular is equivalent to open the two-channel, after the ability of the growth will be rapid.

The Shell's basic learning has been summed up in a blog post: http://www.cnblogs.com/jyzhao/p/4485553.html

This article summarizes the regular expressions in the shell and the commonly used character processing commands, which lays the foundation ^_^ for opening the two veins of the governor.

1. An example of a basic regular expression

2. Character Intercept command

   • Cut command
   • awk command
   • SED command

3. Character processing commands

   • Sort command
   • WC command

1. Examples of basic regular expressions

*,.,\,^word,word$,[list],[^list],[n1-n2],\{n\},\{n,m\}

Example 1: Easy to determine date format

For example: 2015-05-11, simply judge the format of the numbers.

^[0-9]\{4\}-[0-9]\{2\}-[0-9]\{2\}$

Example 2: Easy to determine IP address format

For example: 192.168.1.100, just simple to judge the format of numbers.

[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}

Example 3: Find a line that starts with a letter in a 1.txt text file and does not start with a letter.

Lines that start with a letter:

grep --color=auto "^[a-z,A-Z]" 1.txt

Lines that do not start with a letter:

grep --color=auto "^[^a-z,A-Z]" 1.txt

2. Character intercept command cut command

Purpose: Specific interception of certain columns for the text of the delimiter specification is simple and easy to use.

Example 4: Cut intercepts the user name and corresponding uid,gid, filtering out rows containing '/sbin '.

grep -v "/sbin" /etc/passwd | cut -f 1,3,4 -d ":"

Examples of interception results:

root:0:0oracle:500:500grid:501:500

awk command

Purpose: For the truncation of irregular text, the feature is powerful, syntax is more complex than the cut command.

awk ‘条件类型1{动作1}条件类型2{动作2} ...‘ filename

1. awk built-in variables:

   NF 每一行（$0）拥有的字段总数NR 目前awk所处理的是“第几行”数据FS 目前的分隔字符，默认是空格键

2. The logical operator of awk:

   >, <, >=, <=, ==, !=

Example 5: Querying the PID of the Pmon process.

ps -ef | grep pmon | grep -v grep | awk ‘{print $2}‘

Example 6: Print out the uid<3 users and their UID in/etc/passwd.

cat /etc/passwd | awk ‘BEGIN{FS=":"} $3 < 3 {print $1 "\t" $3}‘

NOTE: Notice that begin is used here, without the first line of the begin display will be incorrect.

Example 7: Capturing the disk space utilization of the system root directory

df -h | awk ‘{print $5}‘|cut -f 1 -d "%"

SED command

SED can replace, delete, add, and select specific rows of data.

-n 只有经过sed处理的才显示（默认显示全部）-e 直接在命令行模式上进行sed的动作编辑-f 将sed的动作写入一个文件，然后-f filename执行filename中的sed命令-r 支持扩展型正则表达式语法（默认是基础正则表达式语法）-i 直接修改读取的文件内容，而不是默认的屏幕输出结果

Example 8: Display line numbers and print out/etc/hosts, deleting the first two lines of a file.

nl /etc/hosts | sed ‘1,2d‘

Example 9: Replace all jy-db in the/etc/hosts file with alfred-db display

sed -e ‘s/JY-DB/Alfred-DB/g‘ /etc/hosts

Note: It is also possible not to write -e parameters here, if two or more SED commands require each of the previous write -e parameters.

Example: Modify the/etc/hosts file directly, insert a new line under line 2nd of the file "192.168.1.100 jy-db"

sed -i ‘2a 192.168.1.100 JY-DB‘ /etc/hosts

3. Character Processing Command Sort command

Purpose: The results are sorted and displayed.

du -sk * | sort -rn
sort -t ":" -k "3,3" /etc/passwd
sort -n -t ":" -k "3,3" /etc/passwd

WC command

Purpose: Count the number of rows, words, and characters of the result.

Statistics Linewc -l

Example: Count the number of user processes in Oracle

ps -ef | grep LOCAL=NO | grep -v grep | wc -l

Statistical wordswc -w

Example: Count the number of words in a/etc/issue file

wc -w /etc/issue

Statistical characterswc -m

Example: Count the number of characters in a/etc/issue file

wc -m /etc/issue

Regular expressions in the shell