NULLIF misinterpreted when a comma is set as a decimal point (DB2 z/OS) ----------------------------------------------------------------------- Key: HHH-5137 URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-5137 Project: Hibernate Core Issue Type: Bug Components: core Affects Versions: 3.3.2 Environment: Hibernate 3.3.2.GA, DB2 9 for z/OS Reporter: Leonardo Priority: Minor If we set in DB2 a comma (instead a period) as a separator for decimal point I get an error when Hibernate executes this query: DEBUG SQL:111 - select foo0_.id as id0_0_, foo0_.myBar_id as myBar1_3_0_, foo0_.clazz_ as clazz_0_ from ( select nullif(0,0) as myBar_id, id, 0 as clazz_ from Foo union all select nullif(0,0) as myBar_id, id, 1 as clazz_ from SubFoo2 union all select myBar_id, id, 2 as clazz_ from SubFoo1 ) foo0_ where foo0_.id=? com.ibm.db2.jcc.b.SqlException: DB2 SQL error: SQLCODE: -170, SQLSTATE: 42605, SQLERRMC: NULLIF In fact, DB2 interprets the argument of nullif(0,0) as a decimal number and not as two arguments (zero, zero). DB2 documentation [1] says that if the comma is set as a decimal point a comma intended as a separator must be followed by a space. so in that case, nullif(0, 0) (note space after ','). We are using DB2390Dialect and getSelectClauseNullString(int sqlType) method doesn't generate a correct nullif instruction in this case. we could circumvent this issue by extending DB2390Dialect class and placing spaces after comma like this: return "nullif(" + literal + ", " + literal + ')'. But isn't a proper solution for us because we must release an extra jar with custom dialect to our customers. [1] http://publib.boulder.ibm.com/infocenter/dzichelp/v2r2/index.jsp?topic=/c... -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira