Online PHP editor | output for WuPSt

<?php // Define a test HTML document containing two tables. Note that the first table // has a row that combined <th> and <td> cells, while the second one only has // <th> cells. $xhtml = <<<XHTML <html> <head> <title>Table test page</title> </head> <body> <table> <tr> <th>Cell 1</th> <td>Cell 2</td> </tr> </table> <table> <tr> <th>Cell 1</th> <th>Cell 2</th> </tr> </table> </body> </html> XHTML; $document = new DOMDocument(); $document->loadXML($xhtml); $xpath = new DOMXPath($document); $xpath = function ($query) use ($xpath) { echo $query . "\n"; $result = $xpath->query($query); if (method_exists($result, 'count')) { $count = $result->count(); } else { // Fallback for older PHP versions that don't have DomNodeList::count(). $result_array = iterator_to_array($result); // Filter out any bogus results that equal `NULL`. $result_array = array_filter($result_array); $count = count($result_array); } echo $count . ' result' . ($count != 1 ? 's' : '') . ($count ? ':' : '') . "\n"; foreach ($result as $element) { $document = new DOMDocument(); $document->appendChild($document->importNode($element->cloneNode(TRUE), TRUE)); echo $document->saveHTML(); } echo "\n"; }; echo "I am using the following XPath expression to select table cells from\n"; echo "the first row of the first table, by combining the results for <th>\n"; echo "and <td>. This works fine, I get both cells back.\n"; echo "Note that the first table combines <th> and <td> in a single row.\n"; $xpath('(((descendant-or-self::html/descendant-or-self::table)[1]//tr)[1]/td|((descendant-or-self::html/descendant-or-self::table)[1]//tr)[1]/th)'); echo "When I try to get the first cell by appending [1] to the expression\n"; echo "I get the second cell back instead of the first. Why?\n"; echo "The expected result of this expression is '<th>Cell 1</th>'\n"; $xpath('(((descendant-or-self::html/descendant-or-self::table)[1]//tr)[1]/td|((descendant-or-self::html/descendant-or-self::table)[1]//tr)[1]/th)[1]'); echo "Appending [2] yields the second cell successfully. But why not the first?\n"; $xpath('(((descendant-or-self::html/descendant-or-self::table)[1]//tr)[1]/td|((descendant-or-self::html/descendant-or-self::table)[1]//tr)[1]/th)[2]'); echo "Similarly, for the second table I can get all cells with this expression.\n"; echo "The second table only contains <th> cells in the first row.\n"; $xpath('(((descendant-or-self::html/descendant-or-self::table)[2]//tr)[1]/td|((descendant-or-self::html/descendant-or-self::table)[2]//tr)[1]/th)'); echo "However, when I append [1] to get only the first cell back, I get an empty result.\n"; echo "The expected result for this expression is '<th>Cell 1</th>'\n"; $xpath('(((descendant-or-self::html/descendant-or-self::table)[2]//tr)[1]/td|((descendant-or-self::html/descendant-or-self::table)[2]//tr)[1]/th)[1]'); echo "The second cell can be retrieved successfully by appending [2].\n"; echo "But why not the first?\n"; $xpath('(((descendant-or-self::html/descendant-or-self::table)[2]//tr)[1]/td|((descendant-or-self::html/descendant-or-self::table)[2]//tr)[1]/th)[2]'); echo "Note that by replacing the first 'descendant-or-self' with '//' the expression\n"; echo "seems to work as expected in all cases for both the first and second table:\n"; $xpath('(((//html/descendant-or-self::table)[1]//tr)[1]/td|((//html/descendant-or-self::table)[1]//tr)[1]/th)'); $xpath('(((//html/descendant-or-self::table)[1]//tr)[1]/td|((//html/descendant-or-self::table)[1]//tr)[1]/th)[1]'); $xpath('(((//html/descendant-or-self::table)[1]//tr)[1]/td|((//html/descendant-or-self::table)[1]//tr)[1]/th)[2]'); $xpath('(((//html/descendant-or-self::table)[2]//tr)[1]/td|((//html/descendant-or-self::table)[2]//tr)[1]/th)'); $xpath('(((//html/descendant-or-self::table)[2]//tr)[1]/td|((//html/descendant-or-self::table)[2]//tr)[1]/th)[1]'); $xpath('(((//html/descendant-or-self::table)[2]//tr)[1]/td|((//html/descendant-or-self::table)[2]//tr)[1]/th)[2]');