XML Grinder

 

Notes (Note)

 

Quick Tour of XML Grinder

XML Grinder is a fairly primitive tool for handling Minimal XML. These instructions should allow you to get a feel for what it can do.

  • follow this link: http://www.xpdl.org/wu/selectfile.jsp
  • Click on "Load Pasted Text"
  • Into the edit box, past a chunk of valid XML - the longer the better, but remember it should be well formed XML.
  • Click "Parse Value"

At this point, the XML value that you have pasted is entered into the system under a particular name, probably "File 1" unless you changed it on the paste window. There are a few options you can do to get a sense of the system.

  • Click on "Reformat" - this will indent the tags properly, and it will also sort the tags alphabetically within a container, so that the file is in a canonical form that can be easily "diff"ed.
  • Click on "Data View" which offers a view more convenent for casual browsing.
  • Click on "Return to XML View" to get back to previous options.
  • Click on "Generate Schema". This will create a "Min Schema", a non standardized format that describes the XML in terms of containers and data elements. Note that a Min-Schema (maybe MinSch for short) is also a Minimal XML file itself.
  • Click on "Change File" to see that you now have two files in memory, and can switch between them. You can of course load more files into memory.
  • Click on "Load Web Resource" to has the system read an XML file that is available on the web. You need to enter a valid URL to an XML resource, such as an RSS feed, etc. Finish with "Load from URL" button and see the XML displayed.
  • Click on "Modify" and see that you get a text editor that allows you to make direct modifications to the currently selected file.

Future Planned Enhancements

  • There is already a Validate button, which validates according to Min-Schema rules (which are minimalistic at this point). Note that the output is also a Minimal XML file.
  • You have to associate a schema with a file, and this is currently cumbersome user interface. Could be better.
  • XSLT Transforms - load an XSLT program in as a file, and apply it to another file. This could be an easy way to test validity.
  • File format recognition: given a set of schemas, it might be possible to find the "best fit" schema, in effect automatically recognizing the schema for a particular file. Thus you would load an XPDL file into memory, and it would immediately recognize it as such.
  • Additional means of specifying validation tests in Java.
  • Persistent files: if you log in and have an account, you coudl save files on the server.