db-ojb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Armin Waibel <arm...@apache.org>
Subject Re: Postponing 1.0 until Weds, Jun 23
Date Wed, 23 Jun 2004 11:50:34 GMT
Hi,

 >
 > With OJB268 (already in CVS), OJB269, OJB270 and OJB271 applied all
 > JUnit tests pass
 > on Oracle9i.

This are good news!

I have some question relating to OJB269, OJB270 and OJB271:

---------------
OJB269:
"Incorrect JDBC type mappings in Torque schema for JUnit test table 
BLOB_TEST cause BlobTest to fail when using Oracle"
----------------
If I apply your patch the BlobTest fails with sapDB/maxDB. With current 
version the test pass. The problem seems to be the datatype used by 
Torque to create the blob column

[torque-sql-exec] Our new url -> jdbc:sapdb://localhost/OJB
[torque-sql-exec] Executing file: 
E:\dev\ojb-head\db-ojb\target\test\sql\ojbtest-schema.sql
[torque-sql-exec] Failed to execute: CREATE TABLE BLOB_TEST ( ID INTEGER 
NOT NULL, BLOB_VALUE_ LONG BYTE, CLOB_VALUE_ LONG UNICODE )
[torque-sql-exec] com.sap.dbtech.jdbc.exceptions.DatabaseException: 
[-3002] (at 87): Invalid datatype

Seems Torque doesn't handle BOLB/CLOB type for sapDB in right way. Does 
the BlobTest pass against MySQL and PostgreSQL out of the box?



--------------
OJB270:
"OJB DB-profile and build system support for ConnectionPool validation 
query"
--------------
I agree in pass the @validationQuery@ attribute, but why should we pass 
@TEST_ON_BORROW@ and @TEST_ON_RETURN@ too? By default AFAIK 
@TEST_ON_BORROW@ is true and why do we need to test on return?


--------------
OJB271:
" JdbcTypesHelper handling of LONGVARBINARY not compatible with Oracle9i"
--------------
All tests pass against sapDB after change. I don't know anything about 
the JDBC spec, so please excuse my question. Is this patch only needed 
for LONGVARBINARY columns? What about VARBINARY/BINARY?


regards,
Armin


Martin Kalén wrote:

> Martin Kalén wrote:
> 
>>> With these issues out of the way, these are the results running
>>> ant junit
>>> against db-ojb HEAD from CVS as of June 21 2004 12:30 UTC
>>> with Sun JDK 1.4.2_04 on Windows XP SP 1, Oracle9i Personal rel 
>>> 9.2.0.1.0
>>> (with OJB.skip.issues=true in build.properties):
>>>     [junit] Running org.apache.ojb.odmg.AllTests
>>>     [junit] Tests run: 196, Failures: 0, Errors: 1
> 
> 
> The ODMG test failure was due to Oracle thin driver truncating the 
> LONGVARBINARY column
> used as OID_ for OJB_NRM-table, causing the conversion class trying to 
> read the data
> as a gzip InputStream to spit exceptions about incompatible format 
> (which I can
> understand when the gzipped output stream written was byte[403], but 
> read back as
> byte[255]).
> 
> After applying OJB271 locally, patching JdbcTypesHelper to stream data 
> from LONGVARBINARY,
> this test pass.
> 
> 
> With OJB268 (already in CVS), OJB269, OJB270 and OJB271 applied all 
> JUnit tests pass
> on Oracle9i.
> "
>     [junit] Running org.apache.ojb.broker.AllTests
>     [junit] Tests run: 402, Failures: 0, Errors: 0, Time elapsed: 92,672 
> sec
>     [junit] Running org.apache.ojb.odmg.AllTests
>     [junit] [BOOT] INFO: Loading OJB's properties from file [snip]
>     [junit] [BOOT] INFO: No value for key 'InterceptorClass'
>     [junit] Tests run: 196, Failures: 0, Errors: 0, Time elapsed: 21,813 
> sec
>     [junit] Running org.apache.ojb.soda.AllTests
>     [junit] Tests run: 3, Failures: 0, Errors: 0, Time elapsed: 3,469 sec
>     [junit] Running org.apache.ojb.otm.AllTests
>     [junit] Tests run: 79, Failures: 0, Errors: 0, Time elapsed: 17,218 sec
> "
> 
> Neat. :-)
> 
>  /Martin
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: ojb-dev-unsubscribe@db.apache.org
For additional commands, e-mail: ojb-dev-help@db.apache.org


Mime
View raw message