Almost there

Done today

• I fixed some bugs in the transformation profiles
• I completed the transformation profiles support for saxon6.x , saxon 8.x and stx and fop.
• The outline plugin is behaving well
• I took the time to test most parts of the program
• I started programming the XML signature module
  

Remaining issues

• Code completion
• The code completion doesn't work well with nested schemas
• The code completion doesn't store a cache on the user's hard drive
• I need to add built-in completion for XSL, Apache ant files, xml schemas, HTML

• Virtual file system
• The implementation is not complete(ssh, ftp, smb)
• Sometimes I hardcoded the files path. I should always use a filesystem abstraction so that remote files can be found.
  
• Users settings
• In the previous versions of xpontus there was no plugins so I had a big panel with a tree to browse and set the preferences
• Now most parts of the software are plugins and some of them can/should be customized by the user(font color, encoding, etc.)
• I should write a settings importer for the old versions of XPontus
• Using a relational database to store user settings might not be a good idea. If I upgrade XPontus and there are some major changes in the API, then all the users settings are gone unless I keep lots of old code in the program. I might revert to XStream which serialize objects in XML. If XStream API changes, I can still parse the data.
  
• Plugin management
• Because there are plugins the plugins browser must be at least able to install, list, remove plugins. I will think about the upgrade later.
• The themes and icons modules are not ready yet. I need a settings initializer because the theme and icon set must be known before loading the program
• Expression evaluator
• The expression evaluator is not working well when evaluating xpath expressions. If the result is a boolean value, then no navigation link should be provided to look at the result. Sometimes it seems that the xpath expression is correct but no result is returned...
• The xquery expression evaluator should have it's own output window to display the result. Sometimes the result is a node, but sometimes it will be plain text or whatever else.