To curate a brand new dataset, there are two steps involved:

1. Create (to copy from a template) policy files -- luckily /usr/local/ecas/filemgr/policy/template has the files: elements.xml, product-type-map.xml and product-types.xml readily available for use.  You can use the web interface to update the metadata definitions.

2. For the most part, a copy and rewrite extractor is sufficient.  To do so, we need to replicate an existing extractor meta configuration file in the /usr/local/ecas/aux/met_def folder to local computer then upload it to the website during the ingestion stage.