chemistry-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Florian Müller (Commented) (JIRA) <j...@apache.org>
Subject [jira] [Commented] (CMIS-444) The InputStream returned by "ContentStream.getStream()" is read ONE byte at a time!
Date Sat, 08 Oct 2011 13:26:29 GMT

    [ https://issues.apache.org/jira/browse/CMIS-444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13123496#comment-13123496
] 

Florian Müller commented on CMIS-444:
-------------------------------------

The content has to be transformed to Base64 and embedded into XML when the AtomPub binding
is used. This transformation only works byte by byte. An application can, of course, provide
a buffered stream to improve the performance.

An even faster way is to create an empty document and set the content afterwards. In this
case the content is not encoded at all and directly streamed to the repository.

We should improve the documentation but I'm not so sure we should always buffer. The application
should be in charge and control.
                
> The InputStream returned by "ContentStream.getStream()" is read ONE byte at a time!
> -----------------------------------------------------------------------------------
>
>                 Key: CMIS-444
>                 URL: https://issues.apache.org/jira/browse/CMIS-444
>             Project: Chemistry
>          Issue Type: Improvement
>          Components: opencmis-client
>    Affects Versions: OpenCMIS 0.5.0
>         Environment: java version "1.6.0_27", Linux Fedora 15, latest version of openCMIS
(0.5.0)
>            Reporter: Ioan LUPU
>         Attachments: testOpenCmis_01.zip
>
>
> - The InputStream "org.apache.chemistry.opencmis.commons.data.ContentStream.getStream()"
is read one byte at a time. The other read(byte[], ...) methods are not called.
> - This affects the performance of the client when uploading files to the DMS server (by
the help of openCMIS) if the code that provides the InputStream doesn't uses a buffer! The
code can be improved to use the methods that read more than one byte a time or the documentation
should be improved in order to inform the developer of this way of reading the InputStream

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

       

Mime
View raw message