Software

eagle-i software is available under a BSD 3 license.

Currently, there are a two ways to install eagle-i: from scratch (source code) or from pre-packaged applications. To make deployment even easier in the future, we anticipate offering eagle-i as a pre-configured virtual machine and via a hosted SaaS (cloud) model. Wonder Lake Software offers technical support and software engineering services to organizations wishing to integrate with, customize or extend the eagle-i platform.

Contents

Getting started: for developers

Get the code

Subversion organization overview

Browse the Subversion repository:

https://open.med.harvard.edu/svn/eagle-i-dev/

Check out the software development trunk:

svn co https://open.med.harvard.edu/svn/eagle-i-dev/apps/trunk

Check out a release of the software, e.g.:

svn co https://open.med.harvard.edu/svn/eagle-i-dev/apps/branches/1.5-MS3.x

Check out the data model development trunk:

svn co https://open.med.harvard.edu/svn/eagle-i-dev/datamodel/trunk

Check out a data model release, e.g.:

svn co https://open.med.harvard.edu/svn/eagle-i-dev/datamodel/releases/0.8.2

Set up a development environment

The eagle-i source code is a multi-module maven project (see code organization). Our code is IDE-agnostic, though we've had a good experience using eclipse with the m2eclipse plugin (standard with the latest version of eclipse):

• After installing m2eclipse, open the eclipse subversion perspective and select "check out as maven project" at the trunk level.
• This will create an eclipse project per maven module
• eagle-i-base contains the root POM. mvn install at this level builds the entire tree.

Alternately, here are some notes on setting up eclipse for GWT debugging from one of our team members.

Browse the Javadocs

Latest release

http://search.eagle-i.net/javadoc

Latest snapshot

http://qa.search.eagle-i.net/javadoc

Get the eagle-i resource ontology

http://code.google.com/p/eagle-i/

Getting started: for system administrators, install from scratch

System requirements

The current eagle-i network deployment is a reference configuration. In this deployment, nine eagle-i institutional servers are VMs with the following configuration:

• RAM: 6 GB
• Disk: 100 GB
• CPU: 2 core
• Operating system: Fedora 14
• Java environment: Sun JDK v. 1.6.0_24, 64 bit
• Servlet container: Tomcat 6.0.26
• RDBMS: MySQL 5.1 (optional, used for logging search usage)
• Mail server: postfix (optional, used to enable resource owner contact feature in Search pages)

In addition, the eagle-i network uses a tenth server for hosting the central search application and a few other central services. The configuration of the central server is similar to that of the institutional servers. 

Download and install the applications

The first three packages (Repository, SWEET, Institutional search) constitute an eagle-i institutional stack. Central search, identity service and ontology browser are central applications (not necessary for an insitutional server).

How are we doing?

Is there anything that could be clearer in our documentation? We welcome your questions and feedback.