LinuxQuestions.org - Regex: How do I put this on one line? (Time-date parsing)

- Programming (https://www.linuxquestions.org/questions/programming-9/)

- - Regex: How do I put this on one line? (Time-date parsing) (https://www.linuxquestions.org/questions/programming-9/regex-how-do-i-put-this-on-one-line-time-date-parsing-699142/)

Regex: How do I put this on one line? (Time-date parsing)

I'm not great at regex so I'll try to explain what I have.

Code:



$line = "Log Created: [12/Dec/2008:06:35:15 -0800] by MyScript on 'ubuntu 8.04 LTS' - - Details: 1/2/3";



$stime = $& if $line =~ /\[.+\]/;

$stime =~ s/^\[//;

$stime =~ s/\]$//;

print "$stime\n";

Would like to put this into $stime in one line instead of 3 so that $stime would have "12/Dec/2008:06:35:15 -0800".

Someone know how to do this?

If I understand you, and the lines look like your sample does, it's not too bad. Essentially you want to capture everything in between [ and ] and leave the time stuff in the order it's found. This works using Perl:

Code:

#!/usr/bin/perl

use strict;

use warnings;



my $line = "Log Created: [12/Dec/2008:06:35:15 -0800] by MyScript on 'ubuntu 8.04

LTS' - - Details: 1/2/3";



$line =~ m/\[(.+)\]/;

my $stime = $1;



print "$stime\n";

If you have multiple instances of [stuff] on any given line or stretching across lines, things get uglier. The parentheses () put the captured items into numbered scalars ($1, $2, etc.) for use later. Depending on how reliable your output is, it probably pays to check for the existence of $1 before you try to use it.

Code:

my $stime = $1 if $1;

Thanks! This works. :)