19005 2008-05-12 06:30:32 -0700 GET form items without name attribute are still submitted 2009-03-02 03:56:35 -0800 1 1 1 Unclassified WebKit Forms 528+ (Nightly build) Mac (Intel) OS X 10.5 RESOLVED DUPLICATE 20184 P2 Normal --- 1 work webkit-unassigned ap oldest_to_newest 80170 0 work 2008-05-12 06:30:32 -0700 When you have for example a select element in a GET form, then remove its name attribute after it has been loaded (using Javascript), then submit the form, WebKit will *still* append the value to the URL, but set no name attribute. E.g. http://foo.com/somepage?=value, where "value" is the selected value of the select. Notes: - I did not test if this also happens with POST forms. - I did not test if this also happens when you don't have a name attribute from the beginning. (I tested with a document that has a name attribute set in the beginning and removed it afterwards, using JS.) 80172 1 21078 ap 2008-05-12 06:39:46 -0700 Created attachment 21078 a test that doesn't fail I cannot reproduce the problem using this test case. 80173 2 21079 work 2008-05-12 06:48:08 -0700 Created attachment 21079 Test case that uses a select element, which does show the bug. When I run your test example, a question mark gets appended to the URL. I'm not sure if that's correct, but that seems like a bug to me? Anyway, I've updated your test case to use a select element, and now it does happen. I guess it only happens with select elements. 80174 3 ap 2008-05-12 06:53:56 -0700 Confirmed with r33044; not a regression from latest shipping WebKit. (In reply to comment #2) > Test case that uses a select element, which does show the bug. Sorry, I should have thought of testing with select, as you described originally. > When I run your test example, a question mark gets appended to the URL. I'm not > sure if that's correct, but that seems like a bug to me? We match Firefox in this regard. Also, appears to match the steps in paragraph 17.13.3 of HTML 4.01 spec. 80175 4 work 2008-05-12 06:57:32 -0700 > Confirmed with r33044; not a regression from latest shipping WebKit. Excellent. I don't think it'll be too hard to fix this, and while annoying, one can live with this. This is my first "contribution" to WebKit :) > Sorry, I should have thought of testing with select, as you described > originally. No problem :) I'm already amazed by the professionalism and speed of your reply! > > When I run your test example, a question mark gets appended to the URL. I'm not > > sure if that's correct, but that seems like a bug to me? > > We match Firefox in this regard. Also, appears to match the steps in paragraph > 17.13.3 of HTML 4.01 spec. Ok, I had a suspicion that wasn't a bug :) Keep on rocking, WebKit Team! :) 111803 5 ap 2009-03-02 03:56:35 -0800 This bug was filed earlier, but bug 20184 got a fix, marking as a duplicate. *** This bug has been marked as a duplicate of 20184 *** 21078 2008-05-12 06:39:46 -0700 2008-05-12 06:39:46 -0700 a test that doesn't fail form.html text/html 274 ap PGJvZHk+PGZvcm0gbWV0aG9kPSJnZXQiIGFjdGlvbj0iZm9vYmFyIj4KPGlucHV0IG5hbWU9YSB2 YWx1ZT12YWx1ZT48L2lucHV0Pgo8L2Zvcm0+CjxzY3JpcHQ+CmZ1bmN0aW9uIHRlc3QoKSB7CiAg ICBkb2N1bWVudC5nZXRFbGVtZW50c0J5VGFnTmFtZSgiaW5wdXQiKVswXS5yZW1vdmVBdHRyaWJ1 dGUoIm5hbWUiKTsKICAgIGRvY3VtZW50LmZvcm1zWzBdLnN1Ym1pdCgpOwp9Cjwvc2NyaXB0Pgo8 YnV0dG9uIG9uY2xpY2s9InRlc3QoKSI+VGVzdDwvYnV0dG9uPgo8L2JvZHk+Cg== 21079 2008-05-12 06:48:08 -0700 2008-05-12 06:48:08 -0700 Test case that uses a select element, which does show the bug. testcase.html text/html 328 work PGJvZHk+PGZvcm0gbWV0aG9kPSJnZXQiIGFjdGlvbj0idGVzdGNhc2UuaHRtbCI+CjxzZWxlY3Qg bmFtZT1hPjxvcHRpb24gdmFsdWU9ImZvb2JhciIgc2VsZWN0ZWQ9InNlbGVjdGVkIj5mb29iYXI8 L3ZhbHVlPjwvc2VsZWN0Pgo8L2Zvcm0+CjxzY3JpcHQ+CmZ1bmN0aW9uIHRlc3QoKSB7CiAgICBk b2N1bWVudC5nZXRFbGVtZW50c0J5VGFnTmFtZSgic2VsZWN0IilbMF0ucmVtb3ZlQXR0cmlidXRl KCJuYW1lIik7CiAgICBkb2N1bWVudC5mb3Jtc1swXS5zdWJtaXQoKTsKfQo8L3NjcmlwdD4KPGJ1 dHRvbiBvbmNsaWNrPSJ0ZXN0KCkiPlRlc3Q8L2J1dHRvbj4KPC9ib2R5Pg==