Reverse engineering uses hibernate type="int" in hbm.xml instead of "integer" for primary key column ---------------------------------------------------------------------------------------------------- Key: HBX-617 URL: http://opensource.atlassian.com/projects/hibernate/browse/HBX-617 Project: Hibernate Tools Issue Type: Bug Affects Versions: 3.1beta4 Environment: Hibernate 3.0, SQL Server 2000, jtds-1.2 driver Reporter: Stefanie Liepmann Priority: Minor Fix For: 3.1.beta5 I have a table in SQL Server 2000 with the following properties: Name: Case_ID DataType: int Length: 4 Allow Nulls: No It is the primary key on the table and it uses the identity generator. I created a hibernate.reveng.xml file and edit it in the Hibernate Reverse Engineering Editor. The code for the file looks some like this: <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE hibernate-reverse-engineering PUBLIC "-//Hibernate/Hibernate Reverse Engineering DTD 3.0//EN" "http://hibernate.sourceforge.net/hibernate-reverse-engineering-3.0.dtd" > <hibernate-reverse-engineering> <table-filter match-catalog="DBName" match-schema="dbo" match-name="Study" /> <table catalog="DBName" schema="dbo" name="Study"> <primary-key> <generator class="identity"></generator> </primary-key> <column name="Column2" jdbc-type='' /> <!-- more columns here --> </table> </hibernate-reverse-engineering> When it generates the hbm.xml file the code looks like this: <id name="caseId" type="int"> <column name="Case_ID" /> <generator class="identity"></generator> </id> Therefore when I generate the .java code, the Java datatype is a primitive "int". This is not desired since the primary key column should be an object (java.lang.Integer in this case) and not a primitive. Shouldn't the type in the hbm.xml file default to "integer". I know that I can "fix" this behaviour by mapping the datatypes in the "Type Mappings" tab, but I wonder if this is what is intended and if it will frustrate other users. I also noticed a similar problem with the short datatype in that I had to explicitly use "java.lang.Short" when the column allows nulls. Otherwise the hbm2java did not map it correctly. If I am wrong about this, please let me know.