I used middelgen before and is happy with its feature for generating hbm config file and POJO classes. Now with updating to Hibernate tool, what I have done so far is set up class path in <taskdef/>, not sure what else I should do. I read some articles, it says to config <jdbcconfiguration/>. To achieve my simple tasks, what can I do in that <jdbcconfiguration/> setting? My general purpose is to provide table name through user's entry after promt, then based on this table name tools will generate hbm config file and POJO. Thanks!

Firstly the documentation is the place to start looking:
http://www.hibernate.org/hib_docs/tools/reference/en/html/

Secondly the Ant tasks are different in organisation from H2 but are much more powerful. See this basic example for just generating the POJOs from HBMs:


To generate the artifacts from a difference source you use a different <configuration> tags, eg <jdbcconfiguration> and then you select the output generator(s) such as <hbm2hbmxml>.

Hope this helps get you started.

I did went through the doc, but failed to find explanation or sample code for generating hbm from db. Can anyone post an example for such purpose? Thanks!

Also, how can I specify the table name to be reverse engineered in hibernate tool? Which ant task will handle it?

One of the goals for the rewrite of the tools was to allow greater matching or sources and sinks. Meaning different inout sources and Output sinks. Inputs can be JDBC, hbm.xml, annotation classes etc See http://www.hibernate.org/hib_docs/tools/reference/en/html/ant.html#d0e710. Outputs can be POJOs, hbm.xml, DAOs, HTML, etc. See http://www.hibernate.org/hib_docs/tools/reference/en/html/ant.html#d0e796.

We do this through a standard meta data datastructure. Fortunately, hibernate has one it uses internally so we have used it for the tools (hence why it more critcal to use the correct tools built to the hibernate built but Im going off topic).

So what does this mean for you. Notice in the example I provided that you have an input stream which is a configuration. In your case you need to use the <jdbcconfiguration> since the input stream is the db meta data. Then you set the output stream. For example you use the <hbm2hbmxml> exporter to generate the hbm.xml files. You can use multiple exporters.

In the case of controlling the <jdbcconfiguration> you can set properties in the hibernate.reveng.xml file. See http://www.hibernate.org/hib_docs/tools/reference/en/html/reverseengineering.html. This allows you to control the process of tables etc to read and many other aspects. The section of the docs I have linked to will explain it all.

Hope this helps.