perl正则表达式模式匹配

作为GMF文件输入：

CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Non-Smartphone Package|3126|GB| | 
CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Smartphone Package|3126|GB| | 
CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Non-Smartphone Package -  Charged|3126|GB|7500000|234446

在perl代码中，我使用下面的命令从行中提取字符串

if($line=~m/^(CUSTEVSUMMROW_GPRS|CUSTEVSUMMROW).*?\s(.*?)\|(\d+)\|.*\|(.*?)$/)
{
    $tag=$1;
    $lineTxt=$2;
    $usage = $3;
    $amt = $4;
}

输出：

tag :: CUSTEVSUMMROW_GPRS  lineTxt :: GPRS - Nova Subscriber Non-Smartphone Package  usage :: 3126  amt ::
tag :: CUSTEVSUMMROW_GPRS  lineTxt :: GPRS - Nova Subscriber Smartphone Package  usage :: 3126  amt ::
tag :: CUSTEVSUMMROW_GPRS  lineTxt :: GPRS - Nova Subscriber Non-Smartphone Package - Charged usage :: 3126 amt :: 234446

---

我如何检索/打印使用的单位是MB或GB。有人能帮我吗。

您不能在

\d+

之后捕获列。为此添加括号

*

是贪婪的，即它尽可能匹配。添加一个

？

以使其更省钱：

您还可以根据需要重写备选方案

(CUSTEVSUMMROW(?:_GPRS)?)

---

在

\d+

之后不捕获列。为此添加括号

*

是贪婪的，即它尽可能匹配。添加一个

？

以使其更省钱：

您还可以根据需要重写备选方案

(CUSTEVSUMMROW(?:_GPRS)?)

---

考虑到你所拥有的：

if($line=~m/^(CUSTEVSUMMROW_GPRS|CUSTEVSUMMROW).*?\s(.*?)\|(\d+)\|(.*?)\|(.*?)$/)
{
    $tag=$1;
    $lineTxt=$2;
    $usage = $3;
    $units = $4;
    $amt = $5;
}

但我认为这不是解决这个问题的最佳方法——我会考虑使用

split

并分别处理第一个字段

可能是这样的：

#!/usr/bin/env perl
use strict;
use warnings;

use Data::Dumper;

my @fields = qw ( tag lineTxt usage units amt );

while (<DATA>) {
    my ( $first_field, @record )  = split '\|';

    #split the first field on _just_ the first space.
    unshift( @record, $first_field =~ m/^(\w+) (.*)$/ );

    #use a hash slice to put that record into a hash of named keys.
    my %data;
    @data{@fields} = @record;
    print Dumper \%data;

    # can of course, make this an array of hashes quite easily. 
}


__DATA__
CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Non-Smartphone Package|3126|GB| | 
CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Smartphone Package|3126|GB| | 
CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Non-Smartphone Package -  Charged|3126|GB|7500000|234446

---

考虑到你所拥有的：

if($line=~m/^(CUSTEVSUMMROW_GPRS|CUSTEVSUMMROW).*?\s(.*?)\|(\d+)\|(.*?)\|(.*?)$/)
{
    $tag=$1;
    $lineTxt=$2;
    $usage = $3;
    $units = $4;
    $amt = $5;
}

但我认为这不是解决这个问题的最佳方法——我会考虑使用

split

并分别处理第一个字段

可能是这样的：

#!/usr/bin/env perl
use strict;
use warnings;

use Data::Dumper;

my @fields = qw ( tag lineTxt usage units amt );

while (<DATA>) {
    my ( $first_field, @record )  = split '\|';

    #split the first field on _just_ the first space.
    unshift( @record, $first_field =~ m/^(\w+) (.*)$/ );

    #use a hash slice to put that record into a hash of named keys.
    my %data;
    @data{@fields} = @record;
    print Dumper \%data;

    # can of course, make this an array of hashes quite easily. 
}


__DATA__
CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Non-Smartphone Package|3126|GB| | 
CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Smartphone Package|3126|GB| | 
CUSTEVSUMMROW_GPRS GPRS - Nova Subscriber Non-Smartphone Package -  Charged|3126|GB|7500000|234446