Description: ------------ htmlspecialchars() does not properly escape URL key tokens like #, " " or % and you should use urlencode() instead. Index: Selenium.php =================================================================== RCS file: /repository/pear/Testing_Selenium/Selenium.php,v retrieving revision 1.5 diff -u -r1.5 Selenium.php --- Selenium.php 14 Oct 2006 17:44:33 -0000 1.5 +++ Selenium.php 21 Oct 2006 05:50:25 -0000 @@ -1184,7 +1184,7 @@ $url = sprintf('http://%s:%s/selenium-server/driver/?cmd=%s', $this->host, $this->port, htmlspecialchars($verb)); for ($i = 0; $i < count($args); $i++) { $argNum = strval($i + 1); - $url .= sprintf('&%s=%s', $argNum, htmlspecialchars($args[$i])); + $url .= sprintf('&%s=%s', $argNum, urlencode($args[$i])); } if (isset($this->sessionId)) { Test script: --------------- <?php require_once 'Testing/Selenium.php;' $selenium = new Testing_Selenium('*firefox', 'http://pear.php.net/index.php'); $selenium->start(); $selenium->open('http://pear.php.net/index.php'); echo $selenium->getText('table#head'); Expected result: ---------------- works Actual result: -------------- selenium server strips the #head class selector

Thank you for the bug report. This bug will be fixed in next release, 0.2.1.

Currently Selenium RC server does not support CSS locator. I think they will support it soon. But, the urlencode problem you pointed out will be fixed, and I will release it very soon. Thank you.

Thank you for your bug report. This issue has been fixed in the latest released version of the package, which you can download at http://pear.php.net/get/Testing_Selenium I fixed this bug in 0.2.1.