[
https://issues.apache.org/jira/browse/WICKET-1886?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12651996#action_12651996
]
Jörn Zaefferer commented on WICKET-1886:
----------------------------------------
Thanks Timo, but that doesn't fix it.
I think the problem is setupRequestAndResponse isn't called "often enough". The
flow seems to be like this:
setup
add cookie to response
try to read previous cookie from request -> missing, therefore write new cookie
setup
So between rendering a page and submit a form and processing the submit,
setupRequestAndResponse isn't called.
So in that respect, SecureForm needs to read a cookie from the request just
after it was written to the response.
The WicketTesterTest is insufficient, as it never reads a cookie from the
request after setting it to the response...
> WicketTester Cookie handling
> ----------------------------
>
> Key: WICKET-1886
> URL: https://issues.apache.org/jira/browse/WICKET-1886
> Project: Wicket
> Issue Type: Bug
> Affects Versions: 1.3.5, 1.4-RC1
> Reporter: Jörn Zaefferer
> Assignee: Timo Rantalaiho
> Fix For: 1.3.6, 1.4-RC2
>
> Attachments: CookieTest.java, patch-WICKET-1886.diff,
> WICKET-1886__SecureForm_and_failing_test.patch
>
>
> While trying to test my SecureForm implementation
> (https://issues.apache.org/jira/browse/WICKET-1885) with WicketTester I ran
> into this issue: A cookie set in the response never shows up in the "next"
> request, because both have their own lists of cookies that aren't shared.
> Afaik both should share the same List instance to handle cookies. That way
> its possible to set a cookie in the response and read it from the request.
> A simple testcase is attached.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
