php - converting badly formatted txt to csv - php

Question

i have a badly formatted text file which i would like to convert to csv.
Here's an example:
100910 NA/1-2013-99636 VIA DEI PESCATORI 2/A LODI APR 8 2013 4:24PM DANNEGGIATO -10% 200 2700 0 0 NO
148013 NA/1-2014-146194 CAVALLOTTI SNC LODI GEN 3 2014 3:37PM DANNEGGIATO -10% 0 0 2 0 NO
160032 NA/1-2014-158129 PAOLO GORINI SNC LODI MAG 6 2014 11:51AM DANNEGGIATO -10% 2 0 2 0 NO
54900 NA/1-2014-158070 STRADA VECCHIA CREMONESE SNC LODI MAG 6 2014 9:53AM DANNEGGIATO +10% 10 0 10 0 NO
100910 NA/1-2013-99636 VIA DEI PESCATORI 2/A LODI APR 8 2013 4:24PM DANNEGGIATO -10% 200 2700 0 0 NO
147959 NA/1-2014-146140 DOSSENA SNC LODI GEN 3 2014 10:45AM DANNEGGIATO -10% 200 0 200 0 NO
That is roughly in this form :
[number] [id] [awfully formatted street] ['LODI'] [timestamp] [damaged or not] [percentage] [squaremeters] [squaremeters] [squaremeters] [squaremeters] [asbest-crumbled or not]
My problem is how to extract the 3rd part, [awfully formatted street].
Basically it's the string after [id] preceding the string ['LODI'] (but ['LODI'] must be just before [timestamp] )
Should i explode() each line by spaces and then traversing the array backwards, overtake [timestamp], overtake ['LODI'] and joining the values before array[id], i.e array [1]? Or is there a smarter (elegant) way to do this, perhaps with preg_match()?
Thanks for any hint!

Answer 1

<?php
// read file line by line
$line = '148013 NA/1-2014-146194 CAVALLOTTI SNC LODI GEN 3 2014 3:37PM DANNEGGIATO -10% 0 0 2 0 NO';
//start by seperating the string on LODI
$lodi_split = explode('LODI', $line);
// Now split the first occ into an array on space
$bits = explode(' ', $lodi_split[0]);
$address = '';
// start reading occurance from occ 2 to loose the first 2 fields
for ($i=2; $i < count($bits); $i++ ) {
$address .= $bits[$i] . ' ';
}
echo $address . PHP_EOL;
Result is
CAVALLOTTI SNC

Answer 2

This should work to extract the address from a row.
<?php
$row = "100910 NA/1-2013-99636 VIA DEI PESCATORI 2/A LODI APR 8 2013 4:24PM DANNEGGIATO -10% 200 2700 0 0 NO";
$row_array = preg_split('/\s+/', $row);
array_shift($row_array);
array_shift($row_array);
for($i=0; $i<12; $i++){
array_pop($row_array);
}
$address = implode(" ", $row_array);
?>

Answer 3

I think explode won't do here. I propose using regexp. For Instance, If you read your .txt file as one string(where data strings are separated with \n):
$f = fopen($fname="file.txt", "rt");
$str = fread($f, filesize($fname)));
fclose($f);
Then use preg_match_all() like this:
$re = "/^(\\d+)\\s*(.*)(LODI)\\s*(.+(?:AM|PM))\\s*(\\w+)\\s+(-?\\d{1,3}%)\\s+(\\d+)\\s+(\\d+)\\s+(\\d+)\\s+(\\d+)\\s+(\\w+)$/m";
preg_match_all($re, $str, $matches,PREG_SET_ORDER );
echo "<pre>\n";
print_r($matches);
echo "</pre>\n";
The output would look like this:
Array
(
[0] => Array
(
[0] => 100910 NA/1-2013-99636 VIA DEI PESCATORI 2/A LODI APR 8 2013 4:24PM DANNEGGIATO -10% 200 2700 0 0 NO
[1] => 100910
[2] => NA/1-2013-99636 VIA DEI PESCATORI 2/A
[3] => LODI
[4] => APR 8 2013 4:24PM
[5] => DANNEGGIATO
[6] => -10%
[7] => 200
[8] => 2700
[9] => 0
[10] => 0
[11] => NO
)
[1] => Array
(
[0] => 148013 NA/1-2014-146194 CAVALLOTTI SNC LODI GEN 3 2014 3:37PM DANNEGGIATO -10% 0 0 2 0 NO
[1] => 148013
[2] => NA/1-2014-146194 CAVALLOTTI SNC
[3] => LODI
[4] => GEN 3 2014 3:37PM
[5] => DANNEGGIATO
[6] => -10%
[7] => 0
[8] => 0
[9] => 2
[10] => 0
[11] => NO
)
..........// And so on
I used the text that you provided above in this example. So in the output you recieve your data formated as list of arrays. So you can do whatever you want with it. $matches[$i][0] - will store the whole match so just skip it and use $matches[$i][1]....$matches[$i][11] as your data.