Open Source XSLerator: A Visual Transformation Tool

XSLerator: A Visual Transformation Tool

Coming up with effective XSLT scripts is not a trivial task specially when
complex transformations are involved and the source document has a complicated
structure. For certain transformations XSLT offers more than one solution
further complicating the task of writing XSLT scripts as issues like performance
and efficiency need to be considered. Fundamentally, the problem of
transformation is not anything new. For decades, tools have been transforming
data from one format to another. One of the areas where such transformations
have been common is databases where data mapping tools have been helping
designers visualize the changes. After some fine-tuning, these tools are able to
generate the necessary scripts.

Since XSLT transformations address the same fundamental issue (although the
data is presumed to be in XML format), it was just a matter of time before new
tools brought about the same visual approach. Tools like XML Spy from Altova (www.xmlspy.com),
and Stylus from Excelon (www.exceloncorp.com) are some examples, which provide
an XML editing environment along with support for XSLT.

Another tool that exclusively focuses on XSLT transformations is XSLerator
from IBM Alphaworks. It can be downloaded from http://www.alphaworks.ibm.com/tech/xslerator.
Although it is still under development and it doesn’t support the latest version
of XSLT specification, I found the tool quite practical. You actually need very
little knowledge of XSLT syntax to use the tool. You start be importing a source
file and a target file. These could be raw XML files or generated by connecting
to a database and extracting the appropriate fields. The tool uses JDBC to
connect to a data source. The tool will generate a tree structure representing
the structure of both the source and target data. The "transformation"
is then achieved by visually creating a series of mappings. A mapping consists
of a source element, a conversion function and a target element.

The default mapping is to simply copy the source into target. You have a
number of functions such as concat(), translate(), substring(), contains(),
min(), and max(). You can also use XSLT constructs for conditions, loops
(for-each) and sorting. As you complete the mapping you can see the XSLT script
that is being constructed allowing you to experiment on the fly with various
approaches. Upon completion, you can generate the entire XSLT script and execute
the transformation from within the tool. It uses Apache Xalan’s implementation
of XSLT. The tool ships with examples and a complete
tutorial.

If your main usage of XSLT is to transform XML into HTML or XHTML for browser
viewing, XSLerate may not be the right tool for you because it doesn’t treat
HTML in any special way. On the other hand, I think XSLerate would be a great
tool for typical B2B applications where documents from one application need to
be mapped into another structure to be consumed by the receiving application.
Typically in these cases, the source and target structures are known and the
problem is coming up with the "right" XSLT script. XSLerate provides a
visual environment to capture the appropriate mapping and then generates the
XSLT script.

About the Author

Piroz Mohseni is president of Bita
Technologies, focusing on business improvement through the effective use of
technology. His areas of interest include enterprise Java, XML, and e-commerce
applications.

Latest Posts

Related Stories