Unix_ Perl's single-line command stunt

Transferred from: superdebug.blog.chinaunix.net

Replace

Replace Foo in all C programs with bar and old files back to. bak

Perl-p-i.bak-e ' s/\bfoo\b/bar/g ' *.c

Very powerful features, especially refactoring in large programs. I remember it was only used in UltraEdit. If you don't want to back up, just write perl-p-i-e or simpler perl-pie, well, pie is a good word.

Add a value to each file that appears in the

Perl-i.bak-pe ' s/(\d+)/1 + $1/ge ' file1 file2 ....

Replace the line break \ r \ n with \ n

Perl-pie ' s/\r\n/\n/g ' file

With the Dos2unix command.

Replace the line break \ n \ r \ n

Perl-pie ' s/\n/\r\n/g ' file

With the Unix2dos command.

Take out part of a file

Field 0-4 and field 6 are displayed, the delimiter for the field is a space

Perl-lane ' print ' @f[0..4] $F [6] "' File

Very good and powerful, with awk ' print $ $, $ $, $ $4, $ $7 '. Parameter name Lane is also very well remembered.

If the field delimiter is not a space but a colon, use the

Perl-f:-lane ' print "@f[0..4]\n" '/etc/passwd

Show the section between Start and end

Perl-ne ' Print if/^start$/. /^end$/' File

I'm afraid this operation can only be done by sed ...

Instead, the part between start and end is not displayed

Perl-ne ' Print unless/^start$/. /^end$/' File

Display the beginning 50 lines:

Perl-pe ' exit if $. > ' File

Same command Head-n 50

Do not display the beginning 10 lines:

Perl-ne ' Print unless 1.. ' File

Display 15 rows to 17 rows:

Perl-ne ' Print if 15.. "File

Each row takes the first 80 characters:

Perl-lne ' Print substr ($_, 0, +) = "" ' File

The first 10 characters are discarded per line:

Perl-lne ' Print substr ($_, ten) = "" ' File

Search

To find the comment string:

Perl-ne ' Print if/comment/' Duptext

This is the normal grep command.

To find rows that do not contain comment strings:

Perl-ne ' Print unless/comment/' Duptext

The reverse of grep, i.e. grep-v.

To find a line that contains comment or Apple:

Perl-ne ' Print if/comment/| | /apple/' Duptext

The same function will be used to egrep, the syntax is more complex, I will not ...

Calculation

Calculate the sum of field 4 and the penultimate field:

Perl-lane ' Print $F [4] + $F [-2] '

If you use awk, you have to write awk ' {i=nf-1;print $5+ $i} '

Sort and invert

The files are sorted by row:

Perl-e ' Print sort <> ' file

Equivalent to a simple sort command.

The file is sorted by paragraph:

Perl-00-e ' Print sort <> ' file

Multiple files are sorted by file contents and returned to the merged file:

Perl-0777-e ' Print sort <> ' file1 file2

File is reversed by line:

Perl-e ' Print reverse <> ' file1

Do you have the appropriate command? Yes...... But quite biased, TAC (cat's reversal)

Numerical calculation

10-in-turn 16-system:

Perl-ne ' printf "%x\n", $_ '

10-in-Turn 8-in: Perl-ne ' printf '%o\n ', $_ '

16-in-turn 10-system:

Perl-ne ' Print hex ($_). " \ n "'

8-in-turn 10-system:

Perl-ne ' Print Oct ($_). " \ n "'

Simple calculator.

Perl-ne ' Print eval ($_). " \ n "'

Other

To start interactive Perl:

Perl-de 1

To view the contents of the Include path:

Perl-le ' Print for @INC '

Note

Perl command-line parameters related to one-liner:

-0< Digital > (in 8 notation) specifies the record delimiter ($/variable), the default is newline-00 paragraph mode, that is, the continuous change behavior delimiter-0777 disables the delimiter, the entire file as a record-a automatic separation mode, separated by a space $_ and saved to @f. Equivalent to @f = Split '. The delimiter can use the-f parameter to specify-F to specify the delimiter for-a, and you can use the regular expression-E to execute the specified script. -i< extension > Replace the file in place and back up the old file with the specified extension. Do not back up without specifying a name extension. -L automatically chomp the input content, automatically add line wrap-n auto-loop to the output, equivalent to while (<>) {script;}-P auto loop + Auto output, equivalent to while (<>) {script; print;}

1--> Show history Command usage frequency

[Email protected]:~$ History | Perl-f "\| | <\ (|;| \ ' |\\$\ ('-alne ' foreach (@F) {print $ if/\b (?! DO) [a-z]+] \b/i} ' | Sort | uniq-c | Sort-nr
169 ls
98 VIM
Python
Man

2--> adds a value to each file that appears in the

Perl-i.bak-pe ' s/(\d+)/1 + $1/ge ' file1 file2 ....

[Email protected]:~$ cat Ptt1.txt
204.108.13.15 ABC [] serverpath=/home/html/pics 62ms
214.92.113.13 xxx [code=5] serverpath=/home/html/pages 32ms

[Email protected]:~$ perl-i.bak-pe ' s/(\d+)/1 + $1/ge ' Ptt1.txt

Match the characters after Servrerpath and print them out

[Email protected]:~$ perl-ne ' print ' $1\n ' if/serverpath= (\s+)/g ' ptt1.txt
/home/html/pics
/home/html/pages

3--> View the contents of the containing path:

Perl-le ' Print for @INC '

4--> Remove part of a file

Field 0-4 and field 6 are displayed, the delimiter for the field is a space

perl-lane ' print ' @f[0..4] $F [6] "' File

Very good and powerful, with awk ' print $ $, $ $, $ $4, $ $7′. Parameter name Lane is also very well remembered.

If the field delimiter is not a space but a colon, use the

perl-f:-lane ' print "@f[0..4]\n" '/etc/passwd

Show the section between Start and end

perl-ne ' Print if/^start$/./^end$/' File

I'm afraid this operation can only be done by sed ...

Instead, the part between start and end is not displayed

perl-ne ' Print unless/^start$/./^end$/' File

Display the beginning 50 lines:

Perl-pe ' exit if $. > ' File

Same command Head-n 50

Do not display the beginning 10 lines:

perl-ne ' Print unless 1 ' file

Display 15 rows to 17 rows:

perl-ne ' Print if: ' File

Each row takes the first 80 characters:

perl-lne ' Print substr ($_, 0, +) = "" ' File

The first 10 characters are discarded per line:

perl-lne ' Print substr ($_, ten) = "" ' File

5--> UrlEncode and UrlDecode are often required to prepare keyword tests

Here are 2 common single-line Perl scripts (regular Expressions): input as a log or keyword list

UrlEncode: Yes \ n not transcoding

Perl-p-E ' s/([^\w\-\.\@])/$1 eq "\ n"? "\ n": sprintf ("%%%2.2x", Ord ($))/eg ' Keywords.list

UrlDecode:

Perl-p-E ' s/% (..) /pack ("C", Hex ($))/eg ' Query.log

Note

Perl command-line arguments related to one-liner

-0< Digital >
(in 8 notation) specifies the record delimiter ($/variable), the default is line wrapping
-00
Paragraph mode, which is a continuous-change-behavior delimiter
-0777
Disable delimiters, which will be the entire file as a record
-A
Auto-delimited mode, separating $_ with spaces and saving to @f. Equivalent to @f = Split ". The delimiter can be specified with the-f parameter
-F
Specify a delimiter for-a, you can use regular expressions
-E
Executes the specified script.
-i< extension >
Replace the file in place and back up the old file with the specified extension. Do not back up without specifying a name extension.
-L
Automatically chomp the input content and automatically add line breaks to the output
-N
Auto Loop, equivalent to while (<>) {script;}
-P
Auto loop + Auto output, equivalent to while (<>) {script; print;}

Unix_ Perl's single-line command stunt