You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@camel.apache.org by "Claus Ibsen (JIRA)" <ji...@apache.org> on 2011/04/17 17:00:05 UTC
[jira] [Assigned] (CAMEL-3842) keys in cxf's protocol_headers must
be handled case-insenstively
[ https://issues.apache.org/jira/browse/CAMEL-3842?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Claus Ibsen reassigned CAMEL-3842:
----------------------------------
Assignee: Claus Ibsen
> keys in cxf's protocol_headers must be handled case-insenstively
> ----------------------------------------------------------------
>
> Key: CAMEL-3842
> URL: https://issues.apache.org/jira/browse/CAMEL-3842
> Project: Camel
> Issue Type: Bug
> Components: camel-cxf
> Affects Versions: 2.7.0
> Reporter: Aki Yoshida
> Assignee: Claus Ibsen
> Fix For: 2.8.0
>
> Attachments: patch-trunk-20110407.zip
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> There was an issue in CXF, where the protocol headers were not stored in a case-insensitive manor. This lead to the problem of not being able to find some headers or finding some headers in duplicates at some CXF interceptors. This issue was fixed in CXF-3367 for CXF 2.3.4 and 2.4. It was fixed by not using the plain HashMap for storing the protocol headers.
> Currently (in trunk), camel-cxf's CxfHeaderHelper and DefaultCxfBinding seem to be using HashMap to create a new CXF's message instance and this can lead to the same issue.
> In CXF, this issue was fixed by replacing the usage of HashMap<String, List<String>)() with that of TreeMap<String, List<String>>(String.CASE_INSENSITIVE_ORDER). This approach was chosen over an alternative approach that converts the header names into lowercase and store them a hashmap because there are typically a small number of entries and using a binary search in TreeMap seems appropriate in this use case.
> I have prepared a patch for a similar change in camel-cxf's CxfHeaderHelper and DefaultCxfBinding and also prepared some unit test cases to verify the behavior.
> Please take a look at this suggestion.
> Thanks.
> Regards, Aki
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira