the problem is as follows: We're using hibernate with annotations as O/R Mapper.
Some @Column annotations look like:
@Column(columnDefinition = "longblob", name = "binaryData", nullable = true)
or
@Column(columnDefinition = "mediumtext", name = "remark", nullable = true)
with the columnDefinition attributes being mysql specific
on postgres for example, the columnDefinition values should be "bytea" and "varchar(999999)"
and on oracle probably something else.
Problems arise currently at the time of Schema Export, e.g. when creating the DDL statements.
Possible workarounds that I can think of are - Hack some JDBC driver that does a text replace (e.g. longblob->bytea) for the DDL statements. That's ugly but will work somehow - Use hibernate xml configuration instead of annotations. That will probably work but I prefer annotations
Does anybody know any alternatives? Hibernate specific workarounds are ok, e.g. if the columnDefinition attribute can contain dialect specific values like
@Column(columnDefinition = "mysql->mediumtext, postgres->varchar(999999)", name = "remark", nullable = true)
Thanks Holger
Why don't you use the database-agnostic annotations like:
@Lob
(on a byte[]
or a String
property)@Column(length=90000)
(on a String
property)and see what columns will be generated in the database. They will most likely be of the types you want them to be.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With