Features for DoDo release

For the purposes of this milestone, we will take an article from PLoS and upload it via a web interface into the repository. The PLoS article will consist of a zip archive. The format of the archive is TBD. It can be like the CDs are now, or we can separate the articles and supporting figures into directories.

The zip archive will be uploaded via a simple web form to the application server. The application will unzip the archive and iterate over the files. It will parse the files and add them to the repository. If it is an article, metadata will be read during the parse and will be set on the article object.

1. User goes to administration page and is presented with a login form
2. User logs in to administration site with admin ID
3. Upon successful login, user is presented with a page that allows him to browse for a local zip file containing a single article and associated files
4. File is uploaded to server
5. Server takes file, parses it for meta data, and inserts it into the Fedora repository
6. Upon successful completion of the ingestion, the user is presented with a success message.

Assumptions: There will be java classes which represent the article, can parse the uploaded files for metadata, make a best guess on Mime types for the figures, etc. Webservice APIs from Topaz will need to support:

• topaz.article.add
• topaz.object.add
• topaz.auth.login?? (somewhere to authenticate against for the application)

Article ingestion takes place inside of a separate web application which will require a separate user from the PloS One application.

Article Display/UserLogin

1. User goes to site and is presented with an (optional) login form. You should be able to read the site without logging in, but for the purposes of this milestone, we want to be able to register/login a user and know who he is.
2. User logs in to site with user credentials
3. Upon login, user is presented with a listing of articles in the repository
4. User can click on any of the article links
5. The user is presented with the article in HTML format

Assumptions:

• There should be a way for the user to register via the website. Though registration/login is not necessarily needed to view articles, this is something we want to be able to test
• There is an API to authenticate against for the user/application
• List articles in the repository and provide a way to reference it
• Given a reference, retrieve the article from the repository and present it back in some format (XML, HTML, PDF, etc)

Closed as the feature list has been moved to description of milestone itself, a more natural place.

Milestone dodo deleted