Australian National Corpus Wiki

An ongoing project to collate and provide access to language data

Brought to you by: gweis, mfallu, rebollor, shirren, stevecassidy

Home

Welcome to the AUSNC sourceforge site - this site contains an archive of the source code for the AUSNC site (http://www.ausnc.org.au/) at the time of it's first release. End users seeking more information about how to use the ausnc website should go to - https://www.ausnc.org.au/help-1

Developers seeking to extend, customise or make a copy of the existing AusNC system can use the guidelines below.

Install and Deployment:
The ausnc system utilises buildout for automated deployment to linux based systems.

After installing buildout - http://www.buildout.org/install.html you will then need to do an initial check out of the core AusNC code.

Perform the following command:
git clone ssh://mfallu@git.code.sf.net/p/ausnc/ausncplone ausnc-ausncplone

This will copy the core AusNC code to your local system.

Then follow the instructions at:
http://sourceforge.net/p/ausnc/ausncplone/ci/9ac772bf472e20b05236914a62b9f401091ad14d/tree/README.txt

This will automatically download and install the remaining components.

These components include 3rd party dependencies and AusNC customised packages.

Code Documentation:
All code is internally documented in the form of readme.txt files and internal code comments. It is recommended that developers familiarise themselves with the documentation of the relevant 3rd party dependencies before attempting to customise the code.

The major 3rd party products include:
Plone: http://plone.org/
Lucene: http://lucene.apache.org/core/
SOLR: http://lucene.apache.org/solr/
ORDF: http://ordf.org/
Seasame: http://www.openrdf.org/
JQuery: http://jquery.com/
NLTK: http://nltk.org/

Discussion

mfallu - 2012-07-02

Latest AusNC Architecture Diagram

Things to note:

Seasame, Solr/Lucene, and OAI-Cat are all Java applications that will operate from within a tomcat container

Shibboleth will be implemented within Apache and will pass credential/attributes to plone (but would be available to other components if needed)

Plone, ORDF, RDFLib, and the ETL ingest tools are all written in Python

This the latest AusNC architecture diagram Showing the loosely coupled architecture of the AusNC platform.

Note:

Apache is used in front of all components - including implementing shibboleth.

Specific products are identified to satisfy functional requirements

Relevenat existing Internal components/apis of products are identified, but only some are development targets

ausnc_architecture.jpg
If you would like to refer to this comment somewhere else in this project, copy and paste the following link: