IGetDBWiki/GettingStarted

Getting Started

This HowTo explains briefly how to get started with developing code for IGetDB project.

DB Access

All IGetDB projects are backed by the projectmanagement system GPMS. To be able to work with IGetDB you need an account.

If you are a developer send an email to: MailTo(krunte DASH devel AT cebitec DOT uni DASH bielefeld DOT de) to get a developer account.

Version Control

IGetDB has dependencies on several CVS-controlled projects (IGetDB itself is in a Subversion repository, see below).

CVS

• create a working direktory for CVS in your home named 'CVS'
• cd to that directory and get the projects needed to run emma:

cvs checkout bioinfo/<name of project>

You need to checkout the following projects:

• projectmanagement
• O2DBI

Subversion

• change into the directory CVS/bioinfo and execute

svn co http://svn.cebitec.uni-bielefeld.de/svn/igetdb/trunk igetdb

API Documentation

The API documentation of IGetDB is not checked in as this causes more trouble then it's worth. To generate your own copy of the API documentation, simply run ant doc in igetdb directory.

Web-Server

IGetDB should be able to run in any servlet-container, although it has been tested with Tomcat 5.5 only.

Compilation and Creation of .war File

Simply run:

ant war

This creates the igetdb.war file in share/webserver/webapps/. The target "war" has a dependency on the target "build": If any classes need to be compiled, this happens automatically.

Documentation

---

As developers we are also responsible for documentation. Please let's try the following conventions:

• document your code
• use the specific documentation tools of the domain
• use JavaDoc comments for documenting Java classes
• when using Eclipse use the Java formatter profile "Java Conventions"
• use POD for documenting Perl module
• use perlmodule Pod::Usage for documenting perl-scripts
• each script needs a usage message (see above)
• O2DBI-documentation (will create POD-doku for server classes)

---

The other main source is the perldoc documentation. The generated O2DBI-servermodules are pod documented. You can read them with:

perldoc module.pm

Other modules may also have pod-documentation but this is not guaranteed. The documentation found can also be incomplete or rather outdated. If you wish to generate the documentation for all perl modules you can do so be executing the shell script:

CVS/bioinfo/igetdb/share/exec/create_html_docs.pl

It generates an HTML version of the perldoc documentation in the directory 'CVS/bioinfo/igetdb/share/doc/html'.

Additional documentation can be found in CVS/bioinfo/igetdb/share/doc.

Directory Structure

The directory `CVS/bioinfo/igetdb/` contains the specific components of the igetdb software:

• `/bin` start wrapper(s) for executable scripts in `/share/exec`.

To run a script in share exec type: `gendb_start scriptname.pl -p project [options]`

• `/share` contains all shared files
  • `/exec` executable scripts, such as the importer scripts GenDB->IGetDB and SAMS->IGetDB
  • `/doc` contains written and generated (e.g. `ant doc`) documentation
• `/WebContent` contains web-server specific code and the document root
  • `/META-INF` contains a default version of the MANIFEST.MF file required for the `.war` file
  • `/WEB-INF` contains the servlet configuration file `web.xml`
    • `/lib` contains all the `.jar` files the application is depending on
• `/src` contains the database schema and source code
  • `/conf` contains all the configuration and resource files that are to be included into the `.war` file
  • `/java` contains the Java source code