Menu
▾
▴
[HtmlUnit] [ htmlunit-Bugs-3371164 ] WebClient.download() wrongly skips actual request
|
From: SourceForge.net <no...@so...> - 2011-07-25 15:52:47
|
Bugs item #3371164, was opened at 2011-07-19 12:25 Message generated for change (Comment added) made by torstenfeig You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=448266&aid=3371164&group_id=47038 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: 2.8 Status: Open Resolution: None Priority: 5 Private: No Submitted By: Torsten Feig (torstenfeig) Assigned to: Nobody/Anonymous (nobody) Summary: WebClient.download() wrongly skips actual request Initial Comment: If a page is loaded by an URI with an anchor appended (e.g. "mypage.jsp#jumpHere") and then a form on that page is submitted which reloads the page without the anchor (e.g. "mypage.jsp"), then WebClient.download() skips the request for loading the page, because it "thinks" that the browser is only to jump to another position on the current page. But this optimization logic ignores the fact that the page can be dynamic so that a new requests (with or without anchor in the URI) yields new content an the page, which is the case here. So I wonder if this "isHashJump" optimization relies on a generally wrong assumption by ignoring the often dynamic character of pages. Currently I had to patch this for my webtests to work. ---------------------------------------------------------------------- >Comment By: Torsten Feig (torstenfeig) Date: 2011-07-25 17:52 Message: @rbri: I had not tried it with the latest snapshot at first because when I looked at the respective code, I had the impression that the error would still be there. But when I tried it today, my tests worked, so the error is obviously fixed (it was a POST request by the way - and checking the request method in download() when setting justHashJump=true seems to be the fix concerning this bug) . @mguillem: I will therefore leave out a unit test, the bug can be closed (shall I do this or is this up to someone else?). ---------------------------------------------------------------------- Comment By: RBRi (rbri) Date: 2011-07-25 09:09 Message: Hi Torsten, do you use the latest snapshot. We did some changes after the 2.8 release in this field. Your form/submit does a GET or POST request? ---------------------------------------------------------------------- Comment By: Marc Guillemot (mguillem) Date: 2011-07-19 14:25 Message: Can you provide a unit test for what you describe? ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=448266&aid=3371164&group_id=47038 |