405 2003-12-09 16:14:18 +0000 validator fails on rewritten tomcat URLs (;jsessionid=... dropped after semicolon) 2004-05-16 08:43:06 +0000 1 1 1 Unclassified Validator check 0.7.0 Other other RESOLVED WONTFIX http://validator.w3.org/check?uri=https://bugs.privasphere.com:8443/listinbox_en.jsp;jsessionid=6E5EC0F1A3F7C21344191EE1C421D56C P2 enhancement --- 1 hauser link oldest_to_newest 933 0 hauser 2003-12-09 16:14:18 +0000 My application rewrites the jsessionIDs always in the URL. I would assume, this is easier for the validator to cope with than if it were to use cookies. But still, for the above URL, I get << I got the following unexpected response when trying to retrieve <https://bugs.privasphere.com:8443/listinbox_en.jsp>: 500 Internal Server Error Please make sure you have entered the URI correctly. >> So, my interpretation is that it chops off anything after the semicolon, i.e. it looses the sesson. Therefore, it appears impossible with the current version to validate pages where a user is logged in into a tomcat/struts session? 945 1 hauser 2003-12-11 07:55:04 +0000 as per w3validator@puoli.net (Samuel), the URL must be encoded e.g. http://validator.w3.org/check?uri=https://bugs.privasphere.com:8443/index_en.jsp;jsessionid=69F4FB9E833C78B32C832A745790A1B9 must be encoded (the uri attribute you're sending to validator). The correct URL would be http://validator.w3.org/check?uri=https%3A%2F%2Fbugs.privasphere.com%3A8443%2Findex_en.jsp%3Bjsessionid%3D69F4FB9E833C78B32C832A745790A1B9 but just replacing the semicolon by %3B appears to be enough. Conclusion: 1) please warn the user, if a semicolon is encountered! 2) perhaps even amend the warning with some hints how to encode a URI (e.g. which java library class to use?) 1838 2 link 2004-05-16 04:43:06 +0000 When you enter this URL into the form at validator.w3.org, the Validator will correctly deal with the semi-colon. The problem will only crop up when you try to construct the URL manually. We might conceivably detect this situation and emit a warning about it (including the suggested fix), but given the amount of overhead involved and the relative obscurity of the problem, I don't think we'll do anything about this in the forseeable future. Resolving WONTFIX, but feel free to reopen if I'm missing something.