You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@openjpa.apache.org by "Catalina Wei (JIRA)" <ji...@apache.org> on 2008/12/24 00:34:44 UTC

[jira] Resolved: (OPENJPA-833) An improved test case for XML column mapping

     [ https://issues.apache.org/jira/browse/OPENJPA-833?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Catalina Wei resolved OPENJPA-833.
----------------------------------

       Resolution: Fixed
    Fix Version/s: 2.0.0
                   1.3.0

> An improved test case for XML column mapping
> --------------------------------------------
>
>                 Key: OPENJPA-833
>                 URL: https://issues.apache.org/jira/browse/OPENJPA-833
>             Project: OpenJPA
>          Issue Type: Test
>          Components: sql
>    Affects Versions: 1.3.0, 2.0.0
>            Reporter: Milosz Tylenda
>            Priority: Minor
>             Fix For: 1.3.0, 2.0.0
>
>         Attachments: OPENJPA-833.patch, TestXMLCustomerOrder.java, TestXMLCustomerOrder.postgres
>
>
> 1. The current version of TestXMLCustomerOrder does not use assert methods
>   requiring programmer to watch the results on the screen. It also fails to
>   correctly determine whether some dictionaries support XML column.
> 2. The patched version does roughly the same tests but uses asserts methods
>   like other JUnit tests.
> 3. In order to correctly determine XML support, an EnitityManager is created
>   so that the DBDictionary executes its connectedConfiguration method. Let me
>   know if there is a better method for doing this.
> 4. I know the original idea with the current version was to catch the generated
>   SQL and compare it with one saved in a file. It was not done, only the SQL
>   is logged to a file and there are three files in the repository with
>   database specific SQL. I abandoned this approach as I find comparing SQL
>   strings too vulnerable to subtle changes in other areas of OpenJPA (e.g
>   column ordering, alias names). On the other hand I found the generated SQL
>   files very helpful to get familiar with how the XML support is done. So, I
>   removed the SQL logging but the patch contains a file with some queries
>   executed against PostgreSQL. Let me know if this removing of SQL logging or
>   adding new file is not desired.
> 5. I have found out that filling a List property in entity does not work. I
>   have annotated the respective test with @AllowFailure. I will file a
>   separate issue for this.
> 6. As in the current version, there are three queries that are expected to
>   fail.
> 7. Since running the test requires an JAXB implementation (not only the JAXB
>   API), I have added Sun's JAXB implementation to openjpa-jdbc/pom.xml
>   with runtime scope and also to the main pom.xml to specify the version.
>   I chose 2.0.5 as I found it the newset from the 2.0 line.
> 8. Since XML column mapping requires a newer PostgreSQL JDBC driver I have
>   upgraded the version in openjpa-jdbc/pom.xml and
>   openjpa-persistence-jdbc/pom.xml.
> 9. I welcome comments and suggestions.

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