带有不同标签的preg_match [关闭]

Question

I'm in need of some assistance. I'm trying to scrape some specify data from a website.

<tbody>
    <tr style="mso-yfti-irow: 1;">
        <td style="width: 184.4pt; border: none; border-left: solid windowtext 1.5pt; padding: 0cm 5.4pt 0cm 5.4pt;" valign="top" width="307">
            <p class="MsoNormal" style="margin-bottom: .0001pt; line-height: normal;">Certifikat springer 1000m</p>
        </td>

        <td style="width: 44.7pt; border: none; border-right: solid windowtext 1.5pt; padding: 0cm 5.4pt 0cm 5.4pt;" valign="top" width="75">
            <p class="MsoNormal" style="margin-bottom: .0001pt; text-align: right; line-height: normal;" align="right">90,-</p>
        </td>
    </tr>

    <tr style="mso-yfti-irow: 2;">
        <td style="width: 184.4pt; border: none; border-left: solid windowtext 1.5pt; padding: 0cm 5.4pt 0cm 5.4pt;" valign="top" width="307">
            <p class="MsoNormal" style="margin-bottom: .0001pt; line-height: normal;">Certifikat springer 1200m</p>
        </td>

        <td style="width: 44.7pt; border: none; border-right: solid windowtext 1.5pt; padding: 0cm 5.4pt 0cm 5.4pt;" valign="top" width="75">
            <p class="MsoNormal" style="margin-bottom: .0001pt; text-align: right; line-height: normal;" align="right">100,-</p>
        </td>
    </tr>   
</tbody>

what I want is to get the "Certifikat springer 1000" from mos-yfti-irow1 and the 90,- from the next TD. but I don't want to get the data from mos-yfti-irow2 in this output.

I'm want to build something where people can compare prices on some activities on our sports group with different clubs. I'm not really sure how to.

This is what I have for now, but can't really get it to work

    <?php 

    $file_string = file_get_contents('http://www.mfkviborg.dk/index.php?    option=com_content&view=article&id=21&Itemid=151');

    preg_match_all('/<p class="MsoNormal" style="margin-bottom: .0001pt;(.*)">(.*)<\/p>/i', $file_string, $links);

    ?>

    <p><strong>Links:</strong> <em>(Name - Link)</em><br />
    <?php
    echo '<ol>';
    for($i = 0; $i < count($links[1]); $i++) {
        echo '<li>' . $links[2][$i] . ' - ' . $links[1][$i] . '</li>';
    }
    echo '</ol>';
    ?>
</p>

Any clues?

Answer 1

A few issues:

• The . does not match with newlines, unless you specify the s modifier at the end of your regex. So that should be added.

• The .* is greedy, so it will match as much as possible including some intermediate </p>. It should not do that, so add a ? (in both cases)

Less of a problem, but still worth changing:

• The first capture group probably does not give you useful information, so remove the parentheses there.

• The . in .0001 is taken as any character, so you should escape it. One way is to put it as [.]

This gives you this line of code:

preg_match_all('/<p class="MsoNormal" style="margin-bottom: [.]0001pt;.*?">(.*?)<\/p>/is', 
             $file_string, $links);

Use DOM parser

Note that if your source HTML only changes slightly (with extra spacing or changing double to single quotes, or swaps the position of attributes ...) you will bump into issues, and be called to adapt the code.

It is much better to use the DOMDocument interface together with a DOMXPath query. Here is how that could work:

$doc = new DOMDocument();
libxml_use_internal_errors(true);
    $doc->loadHTML($file_string, LIBXML_NOCDATA | LIBXML_NOWARNING | LIBXML_NOERROR );
libxml_use_internal_errors(false);
$xpath = new DOMXPath($doc);
$nodes = $xpath->query("//p[contains(@class, 'MsoNormal') and contains(@style, 'margin-bottom: .0001pt')]");
foreach ($nodes as $node) {
    echo $node->textContent . "
";
}

Instead of the loadHTML method you can also use the load method, and pass the URL as first argument.

Follow-up

You asked in comments to further filter the output by tr with mso-yfti-irow in the style attribute:

$nodes = $xpath->query("//tr[contains(@style, 'mso-yfti-irow')]//p[contains(@class, 'MsoNormal') and contains(@style, 'margin-bottom: .0001pt')]");