[Hibernate-issues] [Hibernate-JIRA] Closed: (HB-583) Subclass.java and MSSQL JDBC driver cause a reread exception.

[<leg...@at...>]

Message:

   The following issue has been closed.

   Resolver: Max Rydahl Andersen
       Date: Sat, 27 Dec 2003 11:42 AM

duplicate af HB-484
---------------------------------------------------------------------
View the issue:

  http://opensource.atlassian.com/projects/hibernate/secure/ViewIssue.jspa?key=HB-583


Here is an overview of the issue:
---------------------------------------------------------------------
        Key: HB-583
    Summary: Subclass.java and MSSQL JDBC driver cause a reread exception.
       Type: Bug

     Status: Closed
   Priority: Minor
 Resolution: WON'T FIX

    Project: Hibernate2
 Components: 
             core
   Versions:
             2.1.1

   Assignee: 
   Reporter: Eric Pugh

    Created: Sat, 27 Dec 2003 11:25 AM
    Updated: Sat, 27 Dec 2003 11:42 AM
Environment: Hibernate 2.1.1 and also the nightly build.  Using MS JDBC driver, both the  Service Pack 1 and Service Pack 2 versions.

Description:
Per my forum posting about MSSQL Throwing "ResultSet can not re-read row data for column 51.".

The MSSQL JDBC Driver throws an exception if you don't read left to right if you are quering columns that contain types IMAGE, NTEXT, or TEXT.  According to the JDBC API this actually okay:

For maximum portability, columns within a row should be read 
in left-to-right order, and each column should only be read once. This reflects implementation limitations in some underlying database protocols

The way around this is to use proxies and lazy loading to not load all the objects at once.  However, for joined-subclasses you have to.  The fix is to swap the order of the propertys are read  in Subclass.java:
public Iterator getPropertyClosureIterator() { 
return new JoinedIterator( new Iterator[] { 
getSuperclass().getPropertyClosureIterator(), 
getPropertyIterator()} 
); 
} 

However, I think that a better approch (but beyond me) is to use indexes instead of columns names.  Which I also believe is supposed to be faster for most JDBC drivers.

Eric Pugh



---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.

If you think it was sent incorrectly contact one of the administrators:
   http://opensource.atlassian.com/projects/hibernate/secure/Administrators.jspa

If you want more information on JIRA, or have a bug to report see:
   http://www.atlassian.com/software/jira