Shell basics 67: execution sequence (| and &, {} and () and regular table expressions

From: http://bbs.chinaunix.net/viewthread.php? Tid = 445229

Quote: $ unset
$ [-N "$ A"] & ["$ A"-lt 100] | echo 'too big! '
Too big!

Why can we get the above results? Mark

Question 10
10) & | what is the difference?
Http://bbs.chinaunix.net/forum/viewtopic.php? T = 218853 & show_type = new & postdays = 0 & postorder = ASC & START = 60

Question 6:
6) What is the difference between exec and source?
Http://www.chinaunix.net/forum/viewtopic.php? T = 194191

Summary:
Fork: how to change the environment in the sub-itinerary does not affect the environment of the parent itinerary.

Normally, when we execute a shell script, we first generate a sub-Shell Sub-itinerary, and then the sub-shell will generate the sub-itinerary of the command line.
That is, when we run a script normally:

[Copy to clipboard] [-]

Code:

./My. Script

Source: the so-called source is to allow the script to be executed in the current shell, rather than generating a sub-shell for execution.
Because all execution results are completed in the current shell, if the script environment changes, of course, it will change the current environment.

[Copy to clipboard] [-]

Code:

Source./My. Script
Or:
../My. Script

() And {}

Quote: If you want to combine several commands for execution, s h e l provides two methods. You can execute a group of commands either in the current s h e l or in the child s h e l.
1 ,():
To execute a group of commands in the current s h e l, you can use the command separator to separate each command and enclose all the commands in parentheses.
It generally takes the following form:

[Copy to clipboard] [-]

Code:

(Command 1; command 2 ;...)

2 ,{}:
Use {} to replace (), then the corresponding command will be executed as a whole in the sub-s h e l, only when the output of all commands in {} is redirected as a whole, the commands are placed in the sub-s h e l for execution, otherwise, it is executed in the current s h e l. It generally takes the following form:

[Copy to clipboard] [-]

Code:

{Command 1; command 2 ;...}

This section is incorrect. The following is now available:

[Copy to clipboard] [-]

Code:

(Cmd1; cmd2;...; cmdn) # execute a group of commands in a sub-shell
{Cmd1; cmd2;...; cmdn} # execute a group of commands in the Current Shell
This is a basic concept

[Jason @ firewall Jason] $ A = 1; echo $ A; {A = 2 ;}; echo $
1
2
[Jason @ firewall Jason] $ A = 1; echo $ A; (a = 2;); echo $
1
1

{A = 2;} changed the current shell variable value.
(A = 2;) The current shell variable value is not changed

You can view the original link
Http://bbs.chinaunix.net/forum/viewtopic.php? Show_type = & P = 3386007 #3386007
Thank you for pointing out.

Example 1: In the above example:

[Copy to clipboard] [-]

Code:

$ Comet month_end.txt | exit

If the script fails to be executed, I want to send an email to myself first and then exit. You can use the following method to implement this:

[Copy to clipboard] [-]

Code:

$ Comet month_end | (echo "Comet did no work" | mail Sam; Exit)

In the above example, because only command delimiters are used and they are not combined, s h e l will directly execute the last command (e x I t ).

Example 2: The original example is as follows:

[Copy to clipboard] [-]

Code:

$ Sort quarter_end.txt> quarter. Sorted & LP quarter. Sorted

Use the command combination method. If the s o r t command is successfully executed, first copy the output file to a log area and then print it.

[Copy to clipboard] [-]

Code:

$ Sort quarter_end.txt> quarter. Sorted & (CP quarter. Sorted/logs/quarter. Sorted; LP quarter. Sorted)

Shell 13th question 7:
7) What is the difference between () and?
Http://bbs.chinaunix.net/forum/viewtopic.php? T = 218853 & show_type = new & postdays = 0 & postorder = ASC & START = 45

At the beginning of the RegEx below, there are not many posts in this part, and there are also a lot of posts in the Forum.

In the future, we will post the regular expression rules in the Forum:
Regular expression syntax
Http://bbs.chinaunix.net/forum/viewtopic.php? T = 149723

Post an old article about regular expressions, which is really helpful for beginners.
Http://www.chinaunix.net/forum/viewtopic.php? T = 63273

[Post] examples of regular expressions that are frequently used
Http://www.chinaunix.net/forum/viewtopic.php? T = 190716

My common Regular Expressions
Http://bbs.chinaunix.net/forum/viewtopic.php? T = 190088

Regular Expression expression (For details, refer to the experts)
Http://bbs.chinaunix.net/forum/viewtopic.php? T = 247322

Regular Expression fog, regular expression syntax comparison among various tools
Http://bbs.chinaunix.net/forum/viewtopic.php? T = 425827

[Original] On the greedy nature of Regular Expressions
Http://www.chinaunix.net/forum/viewtopic.php? T = 15825

[Copy to clipboard] [-]

Code:

"Match the beginning and end of a row.
"Match the dataset.
"Only matches letters and numbers.
"Matches string sets within a certain range.

When extracting or filtering text from a file or command output, you can use a regular expression (r e), which is a set of special or not special string modes.

Basic metacharacters and their meanings

Quote: The original post is published by "net player:
ABC: three character characters of ABC, but they are independent from each other rather than a set. (three Char. set can be written in a simple form)
(ABC): a set of three character characters ABC. (The unit can be regarded as a char. Set)
A | B: a single character, or A or B.
(ABC | XYZ): represents one of the two Char. Sets, ABC or XYZ. (Limit 2)
[ABC]: A single character, which can be A, B, or C. (same as the [ABC] principle of wildcard)
[^ ABC]: A single character, Not A, B, or C. (with wildcard [! ABC] Same principle)
.: Represents any single character. (Which of the following is wildcard? Same principle)

[Copy to clipboard] [-]

Code:

^ Match only the beginning of a row
$ Only matches the end of a row
* Only one single character followed by *, matching 0 or more single characters
[] Only matches the characters in. It can be a single character or a character sequence. Available-
Indicates the range of character sequences in []. For example, [1-5] is used to replace [1 2 3 4 5].
/Is only used to block the special meaning of a metacharacter. Because sometimes some metacharacters in s h e l contain
Special meaning. /Can make it meaningless
. Only match any single character
P a t e r n/{n/} is only used to match the occurrence times of the preceding p a t e r n. N is the number of times
P a t e r n/{n,/} m only means the same as above, but the minimum number of times is n
P a t e r n/{n, m/} only indicates the same as above, but the occurrence times of p a t e r n are between N and M.
The special meanings are described in detail.

1. Use a period to match a single character

Example 1: Beng. N: starts with B e g and contains any character in the middle.
Example 2:... x C...: a total of 10 characters. The first four are followed by XC.
Example 3: List directories or files with write permissions for all users:

[Copy to clipboard] [-]

Code:

Ls-L | grep... x. x.. x

2. Match the string or Character Sequence with ^ at the beginning of the line
^ Only matching characters or words at the beginning of a line are allowed.

For example, use the LS-l command and match the directory.

[Copy to clipboard] [-]

Code:

$ LS-L | grep ^ d

3. match strings or characters at the end of a row with $
It can be said that $ is the opposite of ^. It matches strings or characters at the end of a row, and $ is placed after matching words.

Example 1: List all rows ending with the word common in the file httpd1.conf.

[Copy to clipboard] [-]

Code:

$ Grep common $ httpd1.conf
Or
$ Cat httpd1.conf | grep common $

Example 2: match all empty rows: ^ $

Example 3: Only one character row is returned: ^. $

4. Use/to block the meaning of a special character

The following characters are generally considered special characters:

[Copy to clipboard] [-]

Code:

$. '"* [] ^ | ()/+?

Quote: for example:
/.
The character after the backslash is no longer a special character, but a common character, that is, a period.

Quote: If you want to match the rows that contain ^, put the backslash before it to block its special meaning:
/^

Quote: in a regular expression, match all objects ending with *. p a s:
/*/. P A S
To block the specific meaning of the character.

5. the number of times that the/{/} match results appear
Use * to match all matching results at any time, but if you only need to specify the number of times, use /{/},

Quote: This mode has three forms:
Pattern/{n/} match mode appears n times.
Pattern/{n,/} match mode appears at least N times.
Pattern/{n, m} match mode appears between N to m times, n, m is 0-2 5 any integer in 5.

Example 1: match letter A appears twice and end with B:

[Copy to clipboard] [-]

Code:

A/{2/} B

The matching value is a B.

Example 2: Match A at least 4 times:

[Copy to clipboard] [-]

Code:

A/{4,/} B

The result is a B or a A a B, but not a B.

Example 3: If the range of occurrences is given, for example, a appears twice to 4 times:

[Copy to clipboard] [-]

Code:

A/{2, 4/} B

The result is a B, A a B, A a B, rather than a B or a A a B.

Example 4: extract code from the following list:

Quote: 1234xc9088
4523xx9001
0011xa9912
9931xc3445

The format is as follows: the first 4 characters are numbers, followed by X, and the last 4 characters are numbers. The operation is as follows:
[0-9]/{4/} X [0-9]/{4 /}

Quote: the meanings are as follows:
1) The matching number appears four times.
2) followed by code X.
3) the number appears four times.

The result is as follows:

Quote: 1234xc9088-No match
4523xx9001-match
0011xa9912-No match
9931xc3445-No match

Examples of Regular Expressions

[Copy to clipboard] [-]

Code:

^ First line
$ End of line
^ [T h e] For rows starting with t h e
[S] I g n a [L] for matching words s I g n a l, s I g N A L
[Ss] igna [ll]/. Same as above, but add a sentence
[M a y m a Y] a row containing uppercase or lowercase letters of m a y
^ U s e r $ rows that only contain U S E R
[Tty] $ pairs of rows ending with T Y
/. For rows with periods
^ D. x directory that has executable permissions for users, user groups, and other user group members
^ [^ L] list of directories for excluding associated Directories
[. * 0] add any character before or after 0
[0 0 0 *] 0 0 0 or more
[II] uppercase or lowercase I
[I] [n] uppercase or lowercase I or N
[^ $] Empty rows
[^. * $] Match any string in the row
^ ...... $ A row containing 6 Characters
[A-Za-Z] For any single character
[A-Z] [A-Z] * at least one lowercase letter
[^ 0-9/$] identifies a non-digit or dollar
[^ 0-0 A-z A-Z] for non-numbers or letters
[1 2 3] a number between 1 and 3
[D] e v I c e for words d e v I c e or d e v I c e
The first two letters of d e. c e are d e, followed by two arbitrary characters, and finally c e.
/^ The Q pair starts with ^ Q.
^. $ Rows with only one character
^/. [0-9] [0-9] indicates the row starting with a period and two numbers.
'"D e v I c e"' indicates the word d e v I c e.
D e [V v] I c e/. For words d e v I c e or d e v I c e
[0-9]/{2/}-[0-9]/{2/}-[0-9]/{4/} pair Date Format D-m -Y
[0-9]/{3 /}/. [0-9]/{3 /}/. [0-9]/{3 /}/. [0-9]/{3/} on the I p address format NNN. nnn. nnn. nnn
[^. * $] Match any row