Hans,
I looked into the SQL problem and consulted with the JDBC specification lead, Jonathan Bruce. Jonathan said that what JSTL is doing is correct: when setObject(index, null) is passed in a null value this should be converted by the driver to an SQL null.
This behavior is in fact enforced as part of the J2EE compatibility in the CTS. The JDBC Driver Test Suite is publicly accessible and can be used to weed out the JDBC drivers that are not compatible.
Thanks Justyna! So, it sounds like the case is closed. A JDBC driver that doesn't handle setObject(index, null) the way JSTL expects isn't spec compliant.
Hans
Hans Bergsten wrote:
Wolfgang Röckelein wrote:
Hi,
at JDBC level there are two different possibilities to set a parameter value to null: with setNull and setting to null. Depending on the driver sometimes only on of these methods work, and when it does not work, you see the "java.sql.SQLException: Invalid column type" error you see.
I think this was already changed or discussed sometime during the standard taglib development.
Right. I was looking at the code for JSTL in the CVS archive, and it calls setObject(index, null) when passed a null value, and there's a comment that this should be converted by the driver to an SQL null. Browsing through the JDBC JavaDocs and the JDBC spec, there seems to be some support for this claim, but it's not 100% clear. It's possible that the driver Keith is using doesn't handle it, and maybe it would be better if JSTL used setNull(). The reason it doesn't is that it would required additional type info for the <sql:param> case.
Pierre, this may be something to look at again for JSTL.next.
Keith, a work-around for this would be:
<fmt:parseDate value="${param.dob}" var="parsed_dob" pattern="dd-MM-yyyy" />
<c:choose> <c:when test="${!empty parsed_dob}"> <sql:update> INSERT INTO resource_registry (resource_id, dob) VALUES (res_id_seq.NEXTVAL, ? ) <sql:dateParam value="${parsed_dob}" type="date"/> </sql:update> </c:when> <c:otherwise> INSERT INTO resource_registry (resource_id) VALUES (res_id_seq.NEXTVAL) </sql:update> </c:otherwise> </c:choose>
If the real case involves many parameters that may be null, this gets ugly, but if it's just this one, it may be okay.
Hans
-- Hans Bergsten <[EMAIL PROTECTED]> Gefion Software <http://www.gefionsoftware.com/> Author of O'Reilly's "JavaServer Pages", covering JSP 2.0 and JSTL 1.1 Author of O'Reilly's "JavaServer Faces", covering JSF 1.0 Details at <http://TheJSPBook.com/>
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
