XSite

Description

XSite is a program that combines a directory tree of XML files with an XSLT template to create a web site. This program is only suitable for those who know XML and XSL.

Features

• Will transform an entire directory tree from XML.
• Can apply different templates for different files or directories, or use the same template for everything.
• A "site map" is inserted into each XML document before it is processed. This allows production of navigation aids and similar to be automated (eg whenever a new file is added to a directory, all the other pages in that directory get a link to the new file automatically).
• The "sitemap" contains the title of each file, plus an optional description and a place for other data that might be needed outside the file itself.
• Each file is given its modification date and URI (filename/path).
• For development purposes, selected files can be transformed while leaving most files as they are.
• Outputs to HTML, XML or text files.
• This version has source code, documentation and examples (basic first versions).
• GUI or command line versions.

Issues/Bugs

This version is fairly stable, no serious problems were found in the previous version (some bugs turned out to be correct XSLT operation). The core code to this version is much the same. The GUI interface is mostly rewritten.

• Java doesn't guarrentee modification dates are accurate. On my computer (MS JView) they are only accurate to day resolution. I don't know why this is.
• Slow when whole website is processed.
• Ignored Chapters still have their files updated, this is pointless and prevents non-XSite XML files from being placed there.

Legal Stuff

This software is free for personal use. It may not be modified or distributed in any form.

No warrenty is made about fitness for purpose of this program. Read the legal page before proceeding

Update Log

• 0.1: First Version.
• 0.1.1: Added GUI interface for convenience.
• 0.2: GUI interface much more useable. Command line version made more sensible. Files can name the template they want to use. Meta data copied whole rather than being reduced to strings. Ouput in HTML, XML and text. Processing instruction moved to Content tag rather than Meta tag (they shouldn't have been there in the first place). Documentation, source and examples added to distribution. Standard error stream redirected to Text box (since some XSL errors go there rather than the standard output). GUI interface allows only selected files to be processed (not command line though).