Data

Standards

mx mapping to TDWG Data Standards.

Legacy Data

Bulk/Batch uploading

Existing Methods

Several in-app methods exist for loading data in tab-delimited or plain text files directly into the database. These include individual terms for the parts (ontology) table, unparsed collecting events, and links to Morphbank images. The code for doing this is easily transferable to other aspects of the database.

You can also access the Rails environment directly, loading it and using it as you would a shell. This allows you to script-load your data. We currently have methods for the bulk loading of images, languages data, and some very customized reference loading code is also floating around there somewhere. Given the former examples it should be straightforward to write scripts, "tasks" in Rails speak, to get your data loaded. See /lib/tasks for rake examples.

We have in the past used ODBC connections to upload data to mx (e.g. Access). This is relatively straightforward once you understand the various meta-data that are stored in mx.

Exporting

Methods to export data are easily coded, for instance TNT, Nexus (Mesquite legal) etc. formated matrices are already available, as is prototype code for and OWL based ontology dump. Custom dumps for lot data and others have been created as needed, but aren't part of the present source.

Rails has lots of nice features for returning data based on different MIME-type request, .to_xml, .to_json, etc., and its built in RESTful features look to be very nice for future development.