CODEWITHSUNDEEP
regexperl
doubando
doubando

Reputation: 77

Perl Regex to Grep a word not starting with special character _

i want a perl regex to grep a word from the following output:

Process Completed;Result= Volume in drive D has no label.

 Volume Serial Number is 328A-C899

 Directory of D:\Program

07/14/2013  12:09 PM    <DIR>          .
07/14/2013  12:09 PM    <DIR>          ..
06/16/2013  01:07 PM    <DIR>          IPS
07/14/2013  12:10 PM    <DIR>          IPS1
07/14/2013  12:12 PM    <DIR>          IPS2
07/14/2013  12:16 PM    <DIR>          IPS3
07/14/2013  01:50 PM    <DIR>          IPS4
07/14/2013  12:17 PM    <DIR>          IPS5
07/14/2013  12:17 PM    <DIR>          IPS6
07/14/2013  12:18 PM    <DIR>          IPS7
07/14/2013  12:18 PM    <DIR>          IPS8
06/16/2013  01:10 PM    <DIR>          IPSCommon
07/08/2013  12:32 PM    <DIR>          _IPS10
07/08/2013  12:32 PM    <DIR>          _IPS11
07/08/2013  12:32 PM    <DIR>          _IPS12
07/08/2013  12:32 PM    <DIR>          _IPS13
07/08/2013  12:32 PM    <DIR>          _IPS14
07/08/2013  12:57 PM    <DIR>          _IPS15
07/08/2013  12:32 PM    <DIR>          _IPS16
07/08/2013  03:38 PM    <DIR>          _IPS17
07/08/2013  12:32 PM    <DIR>          _IPS18
07/08/2013  12:32 PM    <DIR>          _IPS9
               0 File(s)              0 bytes
              22 Dir(s)  770,968,162,304 bytes free

I have used the following regex IPS\d+\d*$ but this will grep words that starts with _ also

How can I specify ! not _ ?

Upvotes: 2

Views: 742

Answers (5)

TLP
TLP

Reputation: 67910

You are trying to parse the output of the Windows dir command, which I assume is an attempt from you to list files in a directory. You should know that this is not a good way to accomplish that, and I will show you some alternatives. The answers to the regex are already given, so I won't bother with that.

Using a glob, < ... >, which is pretty much an emulation of how a shell expands wildcards.

my @ips = grep /^IPS\d+/,      # only IPS with number
          grep -d,             # only directories
          <D:/Program/IPS*>;   # list IPS file in the target dir

Using opendir. 

opendir my $dh, "D:/Program" or die $!;
my @ips = grep /^IPS\d+/, readdir($dh);   
closedir $dh;

Using File::Find. Do note that this option is recursive (will also list files in subdirectories):

use File::Find;    # core module in Perl 5
my @ips;
find(sub { push @ips, $File::Find::name if /^IPS\d+/ }, "D:/Program");

Each method has its own advantages. In your case, the most similar method to use would be the top one, the one using a glob.

Upvotes: 4

Gansi
Gansi

Reputation: 11

need to capture only Start with IP.

Please use the below regex.

(^IPS\d+)$

Upvotes: 0

dirtydexter
dirtydexter

Reputation: 1073

add this before your regex it will look for word boundary and still not include in result before IPS

(?<=\b)

so your final regex looks like

(?<=\b)IPS\d+\d*$

Upvotes: 0

Barmar
Barmar

Reputation: 781868

Use the \b regexp operator to match a word boundary:

\bIPS\d+$

Upvotes: 2

Zaid
Zaid

Reputation: 37146

You can use a negated character class:

[^_]IPS\d+$

Note that the second \d in your original regex is redundant because \d+ will be greedy.

Upvotes: 1

Related Questions