EBIHARA, Yuichiro wrote on 22.06.2007 06:09:
It seems like PG JDBC driver CANNOT handle 'bytea' as BLOB nor 'text' as CLOB.
getBlob()/setBlob()/getClob()/setClob() can work with only Large Objects (at
least with
postgresql-8.1-405.jdbc3.jar).
org.postgresql.util.PSQLException: Bad Integer Z\273\330x\336\335\226\243
at
org.postgresql.jdbc1.AbstractJdbc1ResultSet.toInt(AbstractJdbc1ResultSet.java:862)
at
org.postgresql.jdbc1.AbstractJdbc1ResultSet.getInt(AbstractJdbc1ResultSet.java:287)
at org.postgresql.jdbc3.Jdbc3ResultSet.getBlob(Jdbc3ResultSet.java:42)
at PgTest.main(PgTest.java:33)
The same exception occurs when using getClob() against a text column.
Using Large Objects may solve my issue but I have to note that a large object
is not automatically
deleted when the record referring to it is deleted.
I found that using getBinaryStream(), setBinaryStream(), getCharacterStream()
and setCharacterStream() to handle LOBs across different DBMS is much more
portable (and reliably) than using the Clob()/Blob() methods.
The Postgres JDBC driver handles the stream/writer methods just fine to read and
write text and bytea columns.
Thomas
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
