models
Data integrated with Metl can be structured or unstructured data. When dealing with Structured data, the structured data can be defined by modeling the data. Metl Models allow you to describe that structured data. Models can either be relational or hierarchical in nature. All structured data must be modeled in Metl. For example, a delimited file may exist that looks as follows:
Smith,John,male,10/19/2000 Smith,Sally,female,5/4/2004 Smith,Sarah,female,2/1/2015
This file represents people including their last name, first name, gender and date of birth. In order to use this file in a structured way, you must model the data contents of the file. I.E. define an entity named something like Person, and then define attributes for that person (lastName, firstName, gender, dateOfBirth).
To create a new model, click on the "Models" folder in the navigation pane, and then click "New" - "Model" as shown below.
Once the model has been created, double click the model in the navigation pane in order to edit the model. From the content editor pane, there are buttons for adding and entities and attributes. Highlighting an entity and clicking the "Add Attribute" button will add an attribute to the given entity.
Similarly when dealing with data in a relational database, there is an inherent model defined in the relational database itself. I.E. tables and columns equate to entities and attributes. Metl models for relational databases can be automatically imported from the database structure itself, or they can be maintained manually. In order to import a model from a relational database, click the "Import…" button. The following screen will be displayed which displays all database resources. Drilling into a database resource allows the user to multi-select one or more tables in order for those tables to be imported into the model.
