Regular expression

Grep

It's mainly used for filtering.

Cp / etc/passwd 1.txt

Grep-- color 'root' 1.txt # prints matching lines and displays them in color

Vim .bashrc

# add

Alias grep='grep-color'

Grep-c 'root' 1.txt # filter line number

Grep-n 'root' 1.txt # prints the line that meets the requirements and displays the line number

Grep-v 'root' 1.txt # upside down, printing does not match the line you want to go

Grep-E = egrep # indicates a special symbol such as () +?, do not escape, escape symbol\

-A: followed by a number (with or without spaces). For example,-A2 prints lines that meet the requirements and the following two lines

-B: followed by a number. For example,-B2 means to print the lines that meet the requirements and the above two lines.

-C: followed by a number. For example,-C2 means to print lines that meet the requirements and two lines up and down.

-r: traverses all the files under the directory

Grep 'r.t' 1.txt # r and t must be printed with a character in the middle

Grep 'rystt' 1.txt # means t must print, r can have zero or more

Grep 'r.roomt' 1.txt # r and t must be printed, with any character in the middle

Grep'[0-9] '1.txt # prints all lines containing numbers

Grep'[^ 0-9] '1.txt # prints all non-numeric lines, such as characters, blank lines

Grep'^ [0-9] '1.txt # prints all lines that start with a number

Grep'[0-2] 'txt = grep' [0min1d2] '1.txt

Grep'^ $'1.txt # print blank lines

Grep 'in$' 1.txt # prints lines ending in in

Grep'[a Murz] '1.txt # prints all small letter lines

Grep'[a-zA-Z] '1.txt # prints lines of all uppercase and small letters

Grep 'root\ | mysql' 1.txt # or, but to escape\

^: what does it start with

$: how does it end

*: means * the character preceding it can have zero or more characters.

. *: any character

+: one or more characters preceding +

Zero or one? The preceding character

Characters that need to be escaped:? + () []\

Sed

Grep can be used by sed, but sed has more functions. You can delete d and replace g.

-r means to use regular special symbols without ambiguity.

Sed '10roomp-n 1.txt # p is printed,-n cancels automatic printing, prints lines as required, and repeats printing without-n

P is usually used with-n, that is, printing is usually used with-n.

Sed-n'/ root/p' 1.txt # prints a line containing a character. You can use regular ^, * $, and / / when you want to use the key word.

-e can be used to perform multiple tasks at the same time, sed-eBay / body/p'-n 1.txt, also can be used.

Sed-n'/ root/p;/body/p' 1.txt

Sed'/ root/d' 1.txt # Delete lines

Sed '1d' 1.txt # delete the first line and print the rest

Sed '1Jing 10d' 1.txt # delete the first line to the tenth line and print the rest

Sed'1 sed 2sCharger to 1.txt # s replace, g global replace, otherwise only replace the first line in 1-2 lines, replace ot with to, the rest can be printed as usual, / can be #, @, these only take effect if the replacement takes effect

Sed's / [0-9] / / g '1.txt # delete all numbers, that is, replace them with empty ones

Sed'/ [0-9] / d'1.txt # deletes all numeric lines

Sed's / [^ 0-9] / / g '1.txt # delete all non-numbers, that is, replace the number with an empty one

Head-N2 1.txt | sed's/\ (root\)\ (. *\)\ (bash\) /\ 3\ 2\ 1 swap the positions of two strings, note the escape symbols, and there are three /

Sed's / ^. * $/ & loging/g' 1.txt # add loging at the end, it must be replaced before it can be added

Sed-I's 1.txt to Universe directly modify the contents of the file

Sed exercises:

Copy / etc/passwd to / root/test.txt

Print all lines with sed: sed'/. * / p'- n 1.txt

Print 3 to 10 lines of test.txt: sed '3Jing 10p'-n 1.txt

Print the line containing 'root'' in test.txt: sed / root/p'-n 1.txt

Delete 15 lines of test.txt and all subsequent lines: sed'15 1.txt

Delete the line in test.txt that contains' bash'; sed'/ bash/d' 1.txt

Replace 'root' with' toor' in test.txt

Replace'/ sbin/nologin' to'/ bin/login' in test.txt

Delete all the numbers from lines 5 to 10 in test.txt: sed-e's / [0-9] / / g'- e '5je 10p'-n 1.txt

Delete all special characters (except numbers and uppercase and lowercase letters) from test.txt: sed's / [^ a-zA-Z0-9] / / g '1.txt

Change the position of the first word and the last word in test.txt: sed-r's / (^. *) (:. *:) (. * $) /\ 3\ 2\ 1 1.txt

Replace the first number and the last word in test.txt

Move the first number in test.txt to the end of the line

Add 'aaa:':sed-e's / ^. * $/ & aaa/g'-e' 10p'- n 1.txt to the end of line 20 of test.txt

Awk

You can print by segment

Awk-F':'{print $1} '1.txt # intercepts the first paragraph of the document with: as the delimiter

Awk-F':'{print $1 "#" $2} '1.txt # uses: as the delimiter, intercepts the first and second paragraphs of the document and delimits them with #. Custom delimiters can only be defined in double quotation marks, not single quotation marks.

Awk-F': 'OFS= "@" {print$1,$3}' 1.txt # and the above meaning

Awk'/ oo/' 1.txt # matches a character or string

Awk-F':'$1 ~ / oo/' 1.txt # match for a certain segment

Awk'/ root | mysql/' 1.txt # matches multiple

Awk-F':'/ root/ {print $1 camera 3}; $1 MySQL 1.txt # match multiple times and show each match.

Awk-F':'$1~/r*o/ {print $3} '1.txt # prints matching lines

The conditional operator = =, >, =, which should be in the form 'root@/bin/bash'): awk-F':''OFS= "@" {print $1 minus 7}' 1.txt

Use':'as the delimiter to add the fourth paragraph of the entire document to sum:

Awk-F': 'sum=sum+$4 {print sum}' 1.txt | tail-N1