extracting data from complex string

victor78

I am working on a project on sourceforge for a game called Everquest. I have a script where the user pastes a raw log into a text box. What I need to do is extract the times, dates, and name into an array. I want the dates formated in the 00-00-0000 format, and the times converted to unix time stamps. I can most likely figure out this part, what I can't figure out is how to get the right information into an array.

Here is an example of a pasted log. The time and date is in the beginning, while the name is between the level class and race.

[Fri Mar 01 23:53:57 2002] [52 Vicar] Sharahelous (Human) <Shining Path>
[Fri Mar 01 23:53:57 2002] [60 Hierophant] Mangue (Unknown) <Shining Path>
[Fri Mar 01 23:53:57 2002] [60 Crusader] Caan (Human) <Shining Path>
[Fri Mar 01 23:53:57 2002] [53 Pathfinder] Aedwlan (Unknown) <Shining Path>
[Fri Mar 01 23:53:57 2002] [56 Myrmidon] Chandos (Wood Elf) <Shining Path>
[Fri Mar 01 23:53:57 2002] [58 Evoker] Andjor (High Elf) <Shining Path>
[Fri Mar 01 23:53:57 2002] [47 Rogue] Alkarin (Dark Elf) <Shining Path>
[Fri Mar 01 23:53:57 2002] [58 Beguiler] Lamaen (High Elf) <Shining Path>
[Fri Mar 01 23:53:57 2002] [56 Conjurer] Senye (High Elf) <Shining Path>
[Fri Mar 01 23:53:57 2002] [ANONYMOUS] Fasad <Shining Path>
[Fri Mar 01 23:53:57 2002] [60 Crusader] Barraind (Human) <Shining Path>
[Fri Mar 01 23:53:57 2002] [58 Templar] Larorf (Dwarf) <Shining Path>
[Fri Mar 01 23:53:57 2002] [60 Hierophant] Tovok (Wood Elf) <Shining Path>
[Fri Mar 01 23:53:57 2002] [ANONYMOUS] Perkins
[Fri Mar 01 23:53:57 2002] [58 Myrmidon] Madine (Barbarian) <Shining Path>

ednark

two ways to do this, i will break it up into two posts

1) regular expressions

i prefer the preg_ functions

read more about them here:
http://www.php.net/manual/en/function.preg-match.php

if you don't know regular expressions already, they are hard to learn. but you can refer to php.net to help learn.

i will give you code below with a regular expression that pulls out the info you want from your lines... there are other ways to write the expression that will work also.

you will end up with the an array where
0 => part of code which was looked at
1 => month
2 => day of month
3 => time
4 => year

with this you should be able to proceed

<?php

$lines_of_file = array( 
    "[Fri Mar 01 23:53:57 2002] [52 Vicar] Sharahelous (Human) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [60 Hierophant] Mangue (Unknown) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [60 Crusader] Caan (Human) <Shining Path>"
);

$regex = "/^.{4}\s(.{3})\s(.{2})\s(.{8})\s(.{4})/";

print "<pre>";
foreach ( $lines_of_file as $line ) {
    $results = array();
    preg_match( $regex, $line, $results );
    print_r($results);
}
print "</pre>";

?>

ednark

the other method is using the substr function

    foreach ( $lines_of_file as $line ) {
        $month = substr( $line, 5,  3 );
        $day   = substr( $line, 9,  2 );
        $year  = substr( $line, 21, 4 );
        $time  = substr( $line, 12, 8 );
    }

ednark

sorry for the large size of this post

i ran a time test with Pear's Benchmark iteration class...

you can read more abou that here:
pear.php.net

here are the results of several runs:

regex fetching : 0.000209

substr fethcing : 0.000185

regex fetching : 0.000212

substr fethcing : 0.000181

regex fetching : 0.000218

substr fethcing : 0.000183

regex fetching : 0.000215

substr fethcing : 0.000184

here is the code i used

<?php

include "Benchmark/Iterate.php";

$lines_of_file = array( 
    "[Fri Mar 01 23:53:57 2002] [52 Vicar] Sharahelous (Human) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [60 Hierophant] Mangue (Unknown) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [60 Crusader] Caan (Human) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [53 Pathfinder] Aedwlan (Unknown) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [56 Myrmidon] Chandos (Wood Elf) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [58 Evoker] Andjor (High Elf) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [47 Rogue] Alkarin (Dark Elf) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [58 Beguiler] Lamaen (High Elf) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [56 Conjurer] Senye (High Elf) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [ANONYMOUS] Fasad <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [60 Crusader] Barraind (Human) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [58 Templar] Larorf (Dwarf) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [60 Hierophant] Tovok (Wood Elf) <Shining Path>",
    "[Fri Mar 01 23:53:57 2002] [ANONYMOUS] Perkins",
    "[Fri Mar 01 23:53:57 2002] [58 Myrmidon] Madine (Barbarian) <Shining Path>"
);

function regex_test( $lines_of_file ) {
    $regex = "/^.{4}\s(.{3})\s(.{2})\s(.{8})\s(.{4})/";
    foreach ( $lines_of_file as $line ) {
        $results = array();
        preg_match( $regex, $line, $results );
    }
}

function substr_test( $lines_of_file ) {
    foreach ( $lines_of_file as $line ) {
        $month = substr( $line, 5,  3 );
        $day   = substr( $line, 9,  2 );
        $year  = substr( $line, 21, 4 );
        $time  = substr( $line, 12, 8 );
    }
}

$benchmark = new Benchmark_Iterate;

$benchmark->run(10000, 'regex_test', $lines_of_file );
$result = $benchmark->get();

echo "regex fetching : {$result['mean']}<br>";

$benchmark->run(10000, 'substr_test', $lines_of_file );
$result = $benchmark->get();

echo "substr fethcing : {$result['mean']}<br>";

?>

ednark

sigh... sorry or being a post hog... but since i started answering this i better get it right... this should teach me to sleep at night....

you should NOT use the substr method, as you want the name, and it is not in the same place every time.... substr just grabs characters in the same place... use preg_match for finding patterns in changing strings

a new regex to pull out the name would look like this:

$regex = "/^{.{4}\s(.{3})\s(.{2})\s(.{8})\s(.{4})]\s[.+]\s(.+)\s(/";}

and would produce:

0 => all parts which were looked at
1 => month
2 => day of month
3 => time
4 => year
5 => name

this expression looks monsterous, but is just a really long straight forward example of the simplist application of regular expressions.
if you read up a bit you should be able to expand my example to pick out all the information in your string.