geronimo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "David Jencks (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (YOKO-434) When reading a value object from a GIOP stream, yoko does not determine if a source value object class had a writeObject method, it uses the target value object class.
Date Mon, 01 Aug 2011 21:30:48 GMT

     [ https://issues.apache.org/jira/browse/YOKO-434?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

David Jencks updated YOKO-434:
------------------------------

    Attachment: YOKO-434-test.diff

rev 1152937 commits an unpleasant solution for this.  This attachment is a unit test for deserializing
various GIOP messages.  Since it  requires significant modification to yoko to allow access
to internals, it should not be comitted in its current form.

> When reading a value object from a GIOP stream, yoko does not determine if a source value
object class had a writeObject method, it uses the target value object class.
> -----------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: YOKO-434
>                 URL: https://issues.apache.org/jira/browse/YOKO-434
>             Project: Yoko - CORBA Server
>          Issue Type: Bug
>      Security Level: public(Regular issues) 
>          Components: orb core
>    Affects Versions: v1.1.0
>            Reporter: David Jencks
>            Assignee: David Jencks
>         Attachments: YOKO-434-test.diff
>
>
> When a value object is serialized into a GIOP message, the existence of a writeObject
method on the source value object class determines whether 2 bytes (stream format version
and whether defaultWriteFields were called) are written to the output stream, and the value
object fields start at the next 4 byte boundary.  However when the value object is read in,
the existence of a writeObject method on the target class is used to determine whether to
expect these 2 bytes.  There is no reason to expect that these writeObject methods' existence
will be the same on the source and target classes.  For instance the oracle BitSet class has
a writeObject method but the Harmony and IBM jdk BitSet classes do not.
> I have not yet found a way to determine from the stream whether the source class had
a writeObject method other than trying both possibilities and picking the first result that
works.  This requires adding mark capabilities to the yoko InputStream. 

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message