we do not want to give developers direct access to query our data in production. For java developers we are simply returning a ref cursor. We have some other people who want to go into production with Toad. Returning them a REF cursor requires them to write a bunch of code to parse the output and print it to the screen. So one option was to use a pipeline function. The developer then queries the pipeline function. All the examples I see involve some pl/sql loops that are piped back such as here: http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:4447489221109 do I need to select bulk collect into the object loop through the object and pipe it back row by row? Is there a better way? -- //www.freelists.org/webpage/oracle-l