[
https://issues.apache.org/jira/browse/CMIS-377?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13036810#comment-13036810
]
Florian Müller commented on CMIS-377:
-------------------------------------
I don't understand the problem. The Atom binding always returns UTF-8,
regardless of the encoding of the request. I can't find any hint in the AtomPub
spec that defines that the encoding of the request and the encoding of the
response must be the same. I think nowadays all XML parsers can handle UTF-8
just fine.
> Characters encoding issue when using Atom binding
> -------------------------------------------------
>
> Key: CMIS-377
> URL: https://issues.apache.org/jira/browse/CMIS-377
> Project: Chemistry
> Issue Type: Bug
> Components: opencmis-server
> Affects Versions: OpenCMIS 0.3.0
> Reporter: Cedric Moitrier
>
> Atom binding does not properly handle characters encoding in HTTP responses.
> Actually, the XMLOutputStream is created using the following piece of code:
> <pre>
> XMLOutputFactory factory = XMLOutputFactory.newInstance();
> fWriter = factory.createXMLStreamWriter(out);
> fWriter.writeStartDocument();
> </pre>
> which is located in
> org.apache.chemistry.opencmis.server.impl.atompub.XMLDocumentBase#startDocument(OutputStream).
> In order for the encoding in the XML header to match the actual characters
> encoding, this should be replaced by:
> <pre>
> XMLOutputFactory factory = XMLOutputFactory.newInstance();
> fWriter = factory.createXMLStreamWriter(out, encoding);
> fWriter.writeStartDocument(encoding, "1.0");
> </pre>
> The output encoding can be obtain from the HTTP request.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
