Regular Expression to collect data from html page and place in an array

Hi
My regex is not the best and I am looking to collect some data from this page to add to a database: http://www.iso.org/iso/support/currency_codes_list-1.htm

I am basically hoping to get hold of the entity - currency - alphabetic code - numeric code and place them in an array

Looking at the source code the code for the table begins around line 200 and is formatted like so:
                  <tr >       
                  <td valign="top">
                        AFGHANISTAN
                  </td>
                  <td valign="top">
                        Afghani
                  </td>
                  <td valign="top">
                        AFN
                  </td>
                  <td valign="top">
                        971
                  </td>
                  </tr>
                  <tr class="zebra">                         
                  <td valign="top">
                        ÅLAND ISLANDS
                  </td>
                  <td valign="top">
                        Euro
                  </td>
                  <td valign="top">
                        EUR
                  </td>             
                  <td valign="top">
                        978
                  </td>             
                  </tr>             
                  <tr >                         
                  <td valign="top">
                        ALBANIA
                  </td>             
                  <td valign="top">
                        Lek
                  </td>             
                  <td valign="top">
                        ALL
                  </td>             
                  <td valign="top">
                        008
                  </td>             
                  </tr>             
                  <tr class="zebra">
                  <td valign="top">
                        ALGERIA
                  </td>
                  <td valign="top">
                        Algerian Dinar
                  </td>             
                  <td valign="top">
                        DZD
                  </td>             
                  <td valign="top">
                        012
                  </td>             
                  </tr>  

I have the following code wrote by kambiz for another problem, which works how I would like:

// fetch data  
$exchange_data = file_get_contents('http://www.bloomberg.com/javascripts/currdata.js');
// apply regular expression
preg_match_all("/\['(\w{3}):CUR'\]\s*=\s*(\d+(.\d+)?);/", $exchange_data, $matches);
// create array of CURRENCY => RATE
$exchange_rates = array_combine($matches[1], $matches[2]);

I am hoping someone could help with some regex which helps me collect the data I want from http://www.iso.org/iso/support/currency_codes_list-1.htm

Thanks
dchidAsked:
Who is Participating?
 
Ray PaseurConnect With a Mentor Commented:
Writing a REGEX for this may be more trouble than it is worth.  The data is not well-formatted.  Here is a script that will make sense of the data.  You can install it and run it to see the output which is an array of associative arrays.
<?php // RAY_iso_currency_codes.php
error_reporting(E_ALL);

// READ THE HTML AND REMOVE THE PARTS WE DO NOT WANT
$url = 'http://www.iso.org/iso/support/currency_codes_list-1.htm';
$htm = file_get_contents($url);
$arr = explode('<strong>Numeric code</strong>', $htm);
$poz = strpos($arr[1], '<tr >');
$htm = substr($arr[1], $poz);
$arr = explode('</tbody>', $htm);
$htm = $arr[0];

$htm = str_replace('class="zebra"', '', $htm);
$arr = explode('<tr >', $htm);
unset($arr[0]);

$out = array();
foreach ($arr as $tr)
{
    $a = array();
    $b = array();
    $c = array();

    // PROCESS BY ROWS
    $tr  = str_replace('<br /><br />', '||', $tr);
    $tra = explode('<td valign="top">', $tr);
    $a['e'] = trim(strip_tags($tra[1]));
    $a['c'] = trim(strip_tags($tra[2]));
    $a['a'] = trim(strip_tags($tra[3]));
    $a['n'] = trim(strip_tags($tra[4]));

    // SOME ROWS HAVE TWO OR THREE ELEMENTS
    if (strpos($tr, '||'))
    {
        $cc = explode('||', $a['c']);
        $aa = explode('||', $a['a']);
        $nn = explode('||', $a['n']);
        $a['c'] = $cc[0];
        $a['a'] = $aa[0];
        $a['n'] = $nn[0];

        $b['e'] = $a['e'];
        $b['c'] = $cc[1];
        $b['a'] = $aa[1];
        $b['n'] = $nn[1];

        if (isset($cc[2]))
        {
            $c['e'] = $a['e'];
            $c['c'] = $cc[2];
            $c['a'] = $aa[2];
            $c['n'] = $nn[2];
        }
    }

    // ADD TO THE ARRAY OF ARRAYS
    $out[]
    = array
    ( 'Entity'     => $a['e']
    , 'Currency'   => $a['c']
    , 'AlphaCode'  => $a['a']
    , 'NumberCode' => $a['n']
    )
    ;

    // IF THERE IS A SECOND ELEMENT
    if (!empty($b))
    {
        $out[]
        = array
        ( 'Entity'     => $b['e']
        , 'Currency'   => $b['c']
        , 'AlphaCode'  => $b['a']
        , 'NumberCode' => $b['n']
        )
        ;
    }

    // IF THERE IS A THIRD ELEMENT
    if (!empty($c))
    {
        $out[]
        = array
        ( 'Entity'     => $c['e']
        , 'Currency'   => $c['c']
        , 'AlphaCode'  => $c['a']
        , 'NumberCode' => $c['n']
        )
        ;
    }
}

// SHOW THE WORK PRODUCT
echo "<pre>";
print_r($out);

Open in new window

0
 
käµfm³d 👽Commented:
How about the following? The values will be stored in capture group 1 (Entity), 2 (Currency), 3 (Alphabetic code), and 4 (Numeric code).
<td valign="top">\s*([\S ]*)\s*</td>\s*<td valign="top">\s*([\S ]*)(?:<a[^>]*>[^<]*</a>)?\s*</td>\s*<td valign="top">\s*([A-Z]{3})\s*</td>\s*<td valign="top">\s*(\d{3}|Nil)\s*

Open in new window

0
 
dchidAuthor Commented:
Thank you both for the replies.
Kaufmed, what you supplied seems to pull what I need from the html but I am having trouble looping through it to add into database fields (entity - currency - alphabetic code - numeric code) and have not yet managed to get it to work as required.

Ray I will look at what you have supplied there and see if I can get that to work and post back with how it goes.

Thanks again
0
Never miss a deadline with monday.com

The revolutionary project management tool is here!   Plan visually with a single glance and make sure your projects get done.

 
kambizConnect With a Mentor Commented:
The currency details on that table are not uniform. In some rows, there are more than one currency, alpha code, and numeric code for one entity (for example, UNITED STATES). Also, there are some rows with only entity field (for example, ANTARCTICA). You may want to process the result associative array to normalize its entries.

// fetch data
$data = file_get_contents('http://www.iso.org/iso/support/currency_codes_list-1.htm');
// apply regex
preg_match_all('@<tr[^>]*>\s*<td[^>]*>\s*([\S ]+)\s*</td>\s*<td[^>]*>\s*([\S ]+)\s*</td>\s*<td[^>]*>\s*([\S ]+)\s*</td>\s*<td[^>]*>\s*([\S ]+)\s*</td>\s*</tr>@s', $data, $matches);
// merge parts into a single associative array
$currencies = array_map('merge_currency_datails', $matches[1], $matches[2], $matches[3], $matches[4]);

// to create an associative array using array_map
function merge_currency_datails($entity, $currency, $alpha_code, $numeric_code) {
  return array(
    'entity'    => htmlspecialchars_decode($entity),
    'currency'  => htmlspecialchars_decode($currency),
    'alpha'     => htmlspecialchars_decode($alpha_code),
    'num'       => htmlspecialchars_decode($numeric_code),
  );
}

Open in new window

0
 
käµfm³d 👽Connect With a Mentor Commented:
Here is code I was playing with to view the extracted data (Note: I had to copy the file locally because I couldn't download it directly from the URL for some reason):
$exchange_data = file_get_contents('test.html');
preg_match_all('#<td valign="top">\s*([\S ]*)\s*</td>\s*<td valign="top">\s*([\S ]*?)\s*(?:<a[^>]*>[^<]*</a>)?\s*</td>\s*<td valign="top">\s*([A-Z]{3})\s*</td>\s*<td valign="top">\s*(\d{3}|Nil)#', $exchange_data, $matches, PREG_SET_ORDER);

print count($matches) . "<br /><br />";

for ($i = 0; $i < count($matches); $i++)
{
	print "<b>Entity:</b> " . $matches[$i][1] . " | <b>Currency:</b> " . $matches[$i][2] . " | <b>Alpha Code:</b> " . $matches[$i][3] . " | <b>Num. Code:</b> " . $matches[$i][4] . "<br />\n";
}

Open in new window

0
 
käµfm³d 👽Commented:
Here is a screenshot of what the results of the above look like.
untitled.PNG
0
 
käµfm³d 👽Commented:
It seems I missed the detail about the data that kambiz mentions when I originally looked through it. The following accounts for this, but the lines that have multiple data would have to be dealt with as a special case--the data is captured in the submatch, but you would have to separate the submatch into separate lines. You may be better off using an approach like Ray_Paseur's.
$exchange_data = file_get_contents('test.html');
preg_match_all('#<td valign="top">\s*([\S ]*)\s*</td>\s*<td valign="top">\s*([\S ]*?(?:<br /><br />[\S ]*?)*)\s*(?:<a[^>]*>[^<]*</a>)?\s*</td>\s*<td valign="top">\s*([A-Z]{3}(?:<br /><br />[A-Z]{3})*)\s*</td>\s*<td valign="top">\s*((?:\d{3}|Nil)(?:<br /><br />(?:\d{3}|Nil))*)#', $exchange_data, $matches, PREG_SET_ORDER);

print count($matches) . "<br /><br />";

for ($i = 0; $i < count($matches); $i++)
{
	print "<b>Entity:</b> " . $matches[$i][1] . " | <b>Currency:</b> " . $matches[$i][2] . " | <b>Alpha Code:</b> " . $matches[$i][3] . " | <b>Num. Code:</b> " . $matches[$i][4] . "<br />\n";
}

Open in new window

0
 
dchidAuthor Commented:
Thank you, you have all been most helpful, I have got all your suggestions partially working how I want, I am going to look into inserting them into the database now and I will post back how this goes.

Kaufmed the for loop you gave has helped alot and with minor alterations I have it working with the other code above.
0
 
dchidAuthor Commented:
Thanks to you all, really appreciate all your help.
0
All Courses

From novice to tech pro — start learning today.