Closed Bug 231155 Opened 21 years ago Closed 19 years ago

browser does not accept new value from drop-down list

Categories

(Core :: Networking: HTTP, defect)

x86
All
defect
Not set
normal

Tracking

()

RESOLVED WORKSFORME

People

(Reporter: jiang_wq, Assigned: darin.moz)

References

()

Details

User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5) Gecko/20031007 Firebird/0.7 Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5) Gecko/20031007 Firebird/0.7 Follow the steps below to reproduce. When you select a new value from the drop-down list, the browser would immediately restore the old value (that is, current value) and load not a new page but the very same current page. This happens to Firebird 0.7, and also happens to the newly released Mozilla 1.6 (so I guess the bug will also show up in Firebird 0.8). Reproducible: Always Steps to Reproduce: 1.Go the the URL 2.find and click on the link "By Points Range" in the left 3.click on "Less than 3,000" and wait until the page is fully loaded 4.in the "point range" drop-down list, the current value should be "Less than 3,000". click on its drop-down button and select "3,000 - 6,000". 5.The page will load automatically. But you'll find it's still the same as previous page, and the drop-down list value is returned to "Less than 3,000" 6.select "3,000 - 6,000" again, and the new page will load. 7.try other new range, such as "6,000 - 12,000", you'll find the bug in first try, and the desired page show up in the second try.
I see this also on LInux 2004012608
OS: Windows 2000 → All
Confirmed with Windows build 2004022909
Assignee: general → general
Status: UNCONFIRMED → NEW
Component: Browser-General → DOM
Ever confirmed: true
QA Contact: general → ian
Not a DOM issue, as far as I can see. And confirming a bug like this without a minimal testcase is not something that should happen. The location is being set fine in the DOM; the problem is that the server sends a 302 redirect right back to the page you came from. The content it delivers will change if a certain cookie is set, but I bet we load it from cache. Excerpt from HTTP log for our handling of the 302 response: 16384[8073fb0]: nsHttpChannel::ProcessRedirection [this=8a846a0 type=302] 16384[8073fb0]: redirecting to: http://www.petro-points.com/eng/rewards/p_RewardList.asp [redirection-limit=20] ... 16384[8073fb0]: nsHTTPChannel::CheckCache [this=8b84950 entry=863c790] 16384[8073fb0]: nsHttpResponseHead::Parse [this=863c490] 16384[8073fb0]: nsHttpResponseHead::ParseVersion [version=HTTP/1.1 200 OK] 16384[8073fb0]: Have status line [version=11 status=200 statusText=OK] 16384[8073fb0]: nsHttpResponseHead::ParseContentType [type=text/html] 16384[8073fb0]: nsHttpResponseHead::MustValidate ?? 16384[8073fb0]: no mandatory validation requirement 16384[8073fb0]: Not validating based on expiration time 16384[8073fb0]: CheckCache [this=8b84950 doValidation=0] 16384[8073fb0]: nsHttpChannel::ReadFromCache [this=8b84950] Using cached copy of: http://www.petro-points.com/eng/rewards/p_RewardList.asp This is the part that bothers me. The previous response for http://www.petro-points.com/eng/rewards/p_RewardList.asp looked like: 16386[811da58]: http response [ 16386[811da58]: HTTP/1.1 200 OK 16386[811da58]: Server: Microsoft-IIS/5.0 16386[811da58]: Date: Sun, 11 Apr 2004 19:28:00 GMT 16386[811da58]: X-powered-by: ASP.NET 16386[811da58]: Content-Length: 36033 16386[811da58]: Content-Type: text/html 16386[811da58]: Expires: Sun, 11 Apr 2004 19:28:00 GMT 16386[811da58]: Cache-Control: private 16386[811da58]: Via: 1.1 S1PS 16386[811da58]: ] So based on the expiration time we _should_ in fact revalidate, no?
Assignee: general → darin
Component: DOM → Networking: HTTP
QA Contact: ian → httpqa
Do you mean the page expired the moment Mozilla received it, so it should not be reused, and should not even had been stored in the cache in the first place?
It should have been stored in the cache (for history purposes). It should not have been read from cache, though.
The bug is still alive in the latest release of firefox 1.0.2.
the web page has been redesigned and I can't reproduce the problem. Please cancel the bug if you have not identified the cause yet. Thanks.
Status: NEW → RESOLVED
Closed: 19 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.