Parse HTML between td-tags?

Guldstrand

Can someone please, PLEASE explain to me how i can extract/parse all the text between the td-tags?
Ex. Blyfri 95, 12,33 kr, 12,18 kr, - 10 öre, 30 dec

<table class="table-standard" style="width: 90.05%; height: 209px;">
    <thead>
        <tr>
            <th>Drivmedel</th>
            <th>Bemannad</th>
            <th>&nbsp;Obemannad</th>
            <th>
            <p>Senaste</p>
            <p><span style="background-color: #ededed;">&nbsp;</span>&auml;ndring</p>
            </th>
            <th>&nbsp;G&auml;ller fr.o.m.</th>
        </tr>
    </thead>
    <tbody>
        <tr>
            <td>&nbsp;Blyfri 95</td>
            <td>&nbsp;12,33 kr</td>
            <td>&nbsp;12,18 kr</td>
            <td><span>&nbsp;-&nbsp;10 &ouml;re</span></td>
            <td>&nbsp;&nbsp;30 dec</td>
        </tr>
        <tr>
            <td>&nbsp;Blyfri 98</td>
            <td>&nbsp;12,83 kr</td>
            <td>&nbsp;12,68 kr</td>
            <td>&nbsp;-&nbsp;10 &ouml;re</td>
            <td>&nbsp;&nbsp;30 dec</td>
        </tr>
        <tr>
            <td>&nbsp;Etanol E85</td>
            <td>&nbsp; 9,43 kr</td>
            <td>&nbsp; 9,28 kr</td>
            <td>&nbsp;+&nbsp;10 &ouml;re</td>
            <td>&nbsp;&nbsp;22&nbsp;dec</td>
        </tr>
        <tr>
            <td>&nbsp;Diesel</td>
            <td>&nbsp;12,62 kr</td>
            <td>&nbsp;12,47 kr</td>
            <td>&nbsp;-&nbsp;10 &ouml;re&nbsp;</td>
            <td>&nbsp;&nbsp;30&nbsp;dec</td>
        </tr>
    </tbody>
</table>

Thanks in advance...

dalecosp

Yes. You'll probably need to use DOMDocument. You will first need to isolate the code. If you want every TD in the file, something as simple as:

$data = file_get_contents($file_name);

should work. Otherwise, you will need to do some magic with strtok() or explode() based on some delimiter you can expect to find within the document.

After that:

$doc = new DOMDocument();
$doc -> loadHTML($data);

$cells = $dom->getElementsByTagName("td");

foreach ($cells as $table_cell) {
   echo $table_cell -> nodeValue.PHP_EOL;
}

HTH,

Guldstrand

That didn't work.
I got:

Fatal error: Call to a member function getElementsByTagName() on a non-object

NogDog

I believe $dom-> should have been $doc->.

sneakyimp

Guldstrand;11044517 wrote:
That didn't work.
I got:

You should also check to make sure that you are actually loading the correct data.

Another approach (which is trickier) is to use a pattern-matching expression like [man]preg_match[/man]. The advantage of this approach is that it can work on poorly structured documents. If your documents are orderly and well-structured then you should probably use DOM.

Guldstrand

As you can see in my first post, the data isn't well-structured.
I have a few pages where i want/need to parse the fuelprices.

Greatfull for all help.

NogDog

If using DOM is impractical, then...

preg_match_all('#<td>(.*?)</td>#is', $theTextMarkup, $matches);
print_r($matches[1]);

Guldstrand

I've managed to parse 3 different sites for the simular info (with the exact same regexp), and now i'm wondering how i can "merge" all the data and save it to mysql?

Exampel:

$data = file_get_contents('url to site 1');
preg_match_all('#<td>(.*?)</td>#is', $data, $matches);

print("<b>" .$matches[1][29]. "</b> Blyfri 95 " .$matches[1][27]. "kr<br />"); 
print("<b>" .$matches[1][32]. "</b> Blyfri 98 " .$matches[1][30]. "kr<br />"); 
print("<b>" .$matches[1][35]. "</b> Diesel " .$matches[1][33]. "kr<br />"); 
print("<b>" .$matches[1][38]. "</b> Etanol " .$matches[1][36]. "kr<br />"); 

$data = file_get_contents('url to site 2');
preg_match_all('#<td>(.*?)</td>#is', $data, $matches);

print("<b>" .$matches[1][4]. "</b> " .$matches[1][0]. " - " .$matches[1][1]. " (" .$matches[1][2]. ")<br />"); 
print("<b>" .$matches[1][9]. "</b> " .$matches[1][5]. " - " .$matches[1][6]. " (" .$matches[1][7]. ")<br />"); 
print("<b>" .$matches[1][14]. "</b> " .$matches[1][10]. " - " .$matches[1][11]. " (" .$matches[1][12]. ")<br />"); 
print("<b>" .$matches[1][19]. "</b> " .$matches[1][15]. " - " .$matches[1][16]. " (" .$matches[1][17]. ")<br />");

sneakyimp

Not sure what you mean by 'merge' the data. Also, it looks like there isn't much of a pattern to the values you want...29, 32, 35, 38, etc. You'll need to be clearer about what you want to do.

Guldstrand

sneakyimp;11044589 wrote:
Not sure what you mean by 'merge' the data. Also, it looks like there isn't much of a pattern to the values you want...29, 32, 35, 38, etc. You'll need to be clearer about what you want to do.

I want to merge all data from the file_get_contents-function, since i'm scraping more then one site.

sneakyimp

Still not sure what you mean by 'merge' the data. A 'merge' to me typically refers to taking data from a spreadsheet, for example, and 'merging' it into a form letter so I can do a mass mailing or something.

Please be more specific. You could even try a little code.

NogDog

Maybe this is what you need?

$data = file_get_contents('url to site 1');
preg_match_all('#<td>(.*?)</td>#is', $data, $matches);
$result = $matches[1];
$data = file_get_contents('url to site 2');
preg_match_all('#<td>(.*?)</td>#is', $data, $matches);
$result = array_merge($result, $matches[1]);

You might want to then apply array_unique() to $result if you want to get rid of any duplicates.

sneakyimp

NogDog;11044597 wrote:
Maybe this is what you need?
$data = file_get_contents('url to site 1');
preg_match_all('#<td>(.*?)</td>#is', $data, $matches);
$result = $matches[1];
$data = file_get_contents('url to site 2');
preg_match_all('#<td>(.*?)</td>#is', $data, $matches);
$result = array_merge($result, $matches[1]);
You might want to then apply array_unique() to $result if you want to get rid of any duplicates.

The problem with that is that it will include any data at all between <td> and </td> tags. It looks from one of his posts like he is only interested in particular cells.

NogDog

If that's the case, then just add them to a separate array variable whose purpose is to hold those values of interest, I suppose, then work with that when you've collected all the values you want -- though it's horribly non-robust, as it would be very sensitive to any changes in the source structure/content (which would be somewhat less of an issue using DOM).