请教怎么用perl处理文本取值

xylonxiang

文本数据如下：
2016-02-26 17:29:42 Type:[JMSDestination] Name:[pay_jmsmodule_01!pay_jmsqueue_01] ConsumersHighCount:[1] MessagesCurrentCount:[0] MessagesPendingCount:[0] MessagesHighCount:[10] MessagesReceivedCount:[373184] ConsumersTotalCount:[1]
2016-02-26 17:29:42 Type:[JMSDestination] Name:[pay_jmsmodule_01!Queue-urm-log] ConsumersHighCount:[1] MessagesCurrentCount:[0] MessagesPendingCount:[0] MessagesHighCount:[4] MessagesReceivedCount:[126072] ConsumersTotalCount:[1]
2016-02-26 17:29:42 Type:[JMSDestination] Name:[pay_jmsmodule_01!Queue-pay-log] ConsumersHighCount:[0] MessagesCurrentCount:[0] MessagesPendingCount:[0] MessagesHighCount:[0] MessagesReceivedCount:[0] ConsumersTotalCount:[0]
2016-02-26 17:29:42 Type:[JMSDestination] Name:[pay_jmsmodule_01!pay_deadqueue] ConsumersHighCount:[0] MessagesCurrentCount:[0] MessagesPendingCount:[0] MessagesHighCount:[0] MessagesReceivedCount:[0] ConsumersTotalCount:[0]


我想把文本格式化成下面这样，即取"[]"中的值请问这用perl怎么处理呢？谢谢！
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!pay_jmsqueue_01|1|0|0|10|373184|1

用一个pl脚本来处理也行

yakczh_cu

'1668 59 3. 7 1713 45 2. 7 1766 53 3. 1 1835 69 3. 9'
这种按4个数据一组怎么处理

RE_HASH

$> cat aa|perl -ne '($T)= /^(.+?:\d\d) /; @V = /\[(.+?)\]/g; printf "$T|%s\n", join "|", @V;'
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!pay_jmsqueue_01|1|0|0|10|373184|1
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!Queue-urm-log|1|0|0|4|126072|1
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!Queue-pay-log|0|0|0|0|0|0
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!pay_deadqueue|0|0|0|0|0|0

sunzhiguolu

回复 3# RE_HASH
能想您问下, -n 在这里的作用吗?

   

RE_HASH

$> perl -h

Usage: perl [switches] [--] [programfile] [arguments]
  -0[octal]         specify record separator (\0, if no argument)
  -a                autosplit mode with -n or -p (splits $_ into @F)
  -C[number/list]   enables the listed Unicode features
  -c                check syntax only (runs BEGIN and CHECK blocks)
  -d[:debugger]     run program under debugger
  -D[number/list]   set debugging flags (argument is a bit mask or alphabets)
  -e program        one line of program (several -e's allowed, omit programfile)
  -E program        like -e, but enables all optional features
  -f                don't do $sitelib/sitecustomize.pl at startup
  -F/pattern/       split() pattern for -a switch (//'s are optional)
  -i[extension]     edit <> files in place (makes backup if extension supplied)
  -Idirectory       specify @INC/#include directory (several -I's allowed)
  -l[octal]         enable line ending processing, specifies line terminator
  -[mM][-]module    execute "use/no module..." before executing program
  -n                assume "while (<>) { ... }" loop around program
  -p                assume loop like -n but print line also, like sed
  -s                enable rudimentary parsing for switches after programfile
  -S                look for programfile using PATH environment variable
  -t                enable tainting warnings
  -T                enable tainting checks
  -u                dump core after parsing program
  -U                allow unsafe operations
  -v                print version, patchlevel and license
  -V[:variable]     print configuration summary (or a single Config.pm variable)
  -w                enable many useful warnings
  -W                enable all warnings
  -x[directory]     ignore text before #!perl line (optionally cd to directory)
  -X                disable all warnings

Run 'perldoc perl' for more help with Perl.

sunzhiguolu

回复 5# RE_HASH
谢谢大神帮助, 向您学习!

   

stanley_tam

来一发不用正则表达式的

1. #!perl
   
2. use strict;
   
3. use warnings;
   
4. use autodie;
   
5. 
   
6. # input file name
   
7. my $file = 'a.txt';
   
8. 
   
9. open my $fh, '<', $file;
   
10. while ( defined (my $line  = readline $fh) ) {
    
11.     # body...
    
12.     chomp $line;
    
13.     my $replacement        = q{};
    
14.     my $date               = substr $line, 0, 19, $replacement;
    
15.    
    
16.     my $is_in_bracket      = 0;
    
17.     my $str_within_bracket = q{};
    
18. 
    
19.     my $line_needed        = $date;
    
20. 
    
21.     for my $char ( unpack("U*", $line) ){
    
22.         $char = pack('U', $char);
    
23. 
    
24.         if ( $char eq '[' ) {
    
25.             $is_in_bracket       = 1;
    
26.             $str_within_bracket  = q{};
    
27.         }
    
28.         elsif ( $char eq ']' ){
    
29.             $is_in_bracket       = 0;
    
30.             $line_needed        .= "|$str_within_bracket";
    
31.         }
    
32.         elsif ( $is_in_bracket ) {
    
33.             $str_within_bracket .= $char;
    
34.         }
    
35.         else {
    
36.             # do nothing
    
37.         }
    
38.     }
    
39. 
    
40.     print "$line_needed$/";
    
41. }
    
42. close $fh;
    
43. 
    
44. 
    
45. __END__

复制代码

sunzhiguolu

我也来个 map 操作符的:

1. #!/usr/bin/perl
   
2. use strict;
   
3. use warnings;
   
4. 
   
5. print map {join ('|', join (' ', (split)[0,1]), m/\[([^\]]*)\]/g), "\n"} <DATA>;
   
6. 
   
7. __DATA__
   
8. 2016-02-26 17:29:42 Type:[JMSDestination] Name:[pay_jmsmodule_01!pay_jmsqueue_01] ConsumersHighCount:[1] MessagesCurrentCount:[0] MessagesPendingCount:[0] MessagesHighCount:[10] MessagesReceivedCount:[373184] ConsumersTotalCount:[1]
   
9. 2016-02-26 17:29:42 Type:[JMSDestination] Name:[pay_jmsmodule_01!Queue-urm-log] ConsumersHighCount:[1] MessagesCurrentCount:[0] MessagesPendingCount:[0] MessagesHighCount:[4] MessagesReceivedCount:[126072] ConsumersTotalCount:[1]
   
10. 2016-02-26 17:29:42 Type:[JMSDestination] Name:[pay_jmsmodule_01!Queue-pay-log] ConsumersHighCount:[0] MessagesCurrentCount:[0] MessagesPendingCount:[0] MessagesHighCount:[0] MessagesReceivedCount:[0] ConsumersTotalCount:[0]
    
11. 2016-02-26 17:29:42 Type:[JMSDestination] Name:[pay_jmsmodule_01!pay_deadqueue] ConsumersHighCount:[0] MessagesCurrentCount:[0] MessagesPendingCount:[0] MessagesHighCount:[0] MessagesReceivedCount:[0] ConsumersTotalCount:[0]

复制代码

rubyish

回复 3# RE_HASH

good, very good
3 Q ~
   

RE_HASH

$> cat aa|perl -npe ' s/( Type:\[)|(\].+?\[)|(\])/{($3)? "":"|"}/ge'
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!pay_jmsqueue_01|1|0|0|10|373184|1
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!Queue-urm-log|1|0|0|4|126072|1
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!Queue-pay-log|0|0|0|0|0|0
2016-02-26 17:29:42|JMSDestination|pay_jmsmodule_01!pay_deadqueue|0|0|0|0|0|0