VASCAN portal and database are now open source

Monday, 12 December 2011

The Database of Vascular Plants of Canada (VASCAN) web portal and its database are now open source. The code is hosted on our Google Code site and can be used by anyone to create their own checklist portal. By doing so, we hope that the time and effort that went into developing VASCAN can benefit other developers. We also hope to encourage collaborative development, as anyone can now review the code or develop additional modules. The repository also serves as a backup of our code and its development history.

What is VASCAN?

You don’t know yet? No problem, you can read all about it here:

• VASCAN about page
• Brand new page describing VASCAN as software
• Blog post summarizing VASCAN’s functionalities

Where can I get the code?

• Source code for the webportal
• SQL dump of the MySQL database
• Installation guidelines

Challenges in making VASCAN open source

Even though we always wanted to open-source our code, it is of course developed to fit our needs and setup first. So, our configuration is entangled with the functionality and we have some idiosyncratic code here and there. The challenge was to adapt the current code to support different setups, while keeping most of its original structure. This is what we did:

• All configurations (database connection, data folder, base url, etc.) are now stored in a separate file
• All generated data (images, Darwin Core Archives, indexes, sitemap, etc.) are now stored outside the webapp
• Automatic generation of a “sitemap.xml” for improved indexing by search engines like Google. Access can now be controlled with “robots.txt”.
• Developed an installation function to build the application on your own server. You only need to edit 3 files and building the application is as simple as typing the command “ant war” to create a Java war file.
• Improved documentation and installation guidelines.