I understand why you might want to use code blocks to create AJAX-like interfaces to XML data streams, but I'm wondering why you don't offer an option to process a pure XML stream that might have been pre-processed by an XSLT file.

I recently built some nice mashups of employee population data using SQLXML to extra the data from a SQL Server, passing the data through XSLT transforms to generate KML files for Google Earth. The generate KML included links to SQLXML queries so the user could "drill down" to retrieve additional data. I took full advantage of SQLXML template queries, and also created queries which generated KML folders containing parameterized queries (typcially one folder per district or state).

The entire thing was built with Visual Web Developer Express Edition without a line of traditional code. All logic was expressed in XSLT and T-SQL.

-- Peter

Re: Why is it necessary to create a code block?


You could create a Popfly block that parsed the output of the XSLT into Javascript objects which other blocks could interact with. If the XSLT outputed JSON that would be as simple as an eval(). Since Javascript has limited support for XSLTs we don't expose any way to make use of them.