openjpa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Fernando (JIRA)" <j...@apache.org>
Subject [jira] Updated: (OPENJPA-798) slices does can not work with sequences properly (no adding objects?)
Date Sun, 30 Nov 2008 23:04:44 GMT

     [ https://issues.apache.org/jira/browse/OPENJPA-798?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Fernando updated OPENJPA-798:
-----------------------------

    Attachment: slices-resultset.diff

big patch to fix up badly broken DistributedResultSet.

Because it was a big re-write I created a DistributedResultSet2 and DelegatingResultSet2.
 Then changed DistributedTemplate and DistributedPreparedStatement to use the new DistributedResultSet2.

It works now, I can now persist objects.

> slices does can not work with sequences properly (no adding objects?)
> ---------------------------------------------------------------------
>
>                 Key: OPENJPA-798
>                 URL: https://issues.apache.org/jira/browse/OPENJPA-798
>             Project: OpenJPA
>          Issue Type: Bug
>          Components: kernel
>    Affects Versions: 2.0.0
>            Reporter: Fernando
>            Priority: Critical
>         Attachments: slices-resultset.diff
>
>
> I can't believe that this was a bug that was existent.. but essentially once I got slices
fixed up (see many preceding bugs), it all seems to work fine, except for I can't persist
new objects.  Essentially the DistributedResultSet is very very buggy, and not implemented
properly, so TableJDBCSeq can't do it's job properly ( since it's using the DistributedConnection/PreparedStatement/RestulSet
).
> I just wanted to file the bug to get the word out.. but I basically have to re-write
the whole class, So I guess I'll be submitting a patch tomorrow.
> The first bug is that it will drop out the first row from every ResultSet (one per db).
 It does this because as ResultSets are added, it calls "first", but then most people (and
TableJDBCSeq), assume that it's "beforeFirst", so they call "next" expecting it to move to
"first".  So in TableJDBCSeq, it goes beyond the first row, returning null.. and TableJDBCSeq
thinks that the query actually returned an empty RestultSet and tries to act accordingly...
> and as you review DistributedResultSet, you see lots and lots of bad logic, missing code,
etc.  It's a pretty hacked together implementation of a "Distributed ResultSet".... need a
proper version..

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message