Databricks¶

The Databricks data provider allows ChartFactor to interact with Databricks.

To use the Databricks data provider you need include its library:

<script src="./cft-databricks-provider.min.js"></script>

The Provider JSON object requires a name, the provider type (e.g. 'databricks'), host, warehouse id, and token parameters. Example:

// define providers
var providers = [{
    name:'Databricks',
    provider:'databricks',
    host: 'DATABRICKS_HOST',
    warehouseId: 'DATABRICKS_WAREHOUSE_ID',
    token: 'DATABRICKS_TOKEN',
    metadata: {}
}]

Optionally, the configuration for Databricks allows to specify headers in case header elements need to be provided (e.g. User-Agent).

var providers = [{
    ...
    headers: { "User-Agent": "myapp/1.0.0"},
}]

Supported provider properties:

host: The Server Hostname value
warehouseId: The Databricks warehouse id value
token: The Databricks personal access token value
headers: The HTTP headers to be sent to the Databricks server, if the Authorization header is provided it will be overwritten by the token value
metadata: Please refer to the Custom Metadata section for a description on how to configure custom metadata

Finally, use the setProviders() method of ChartFactor to set your data provider definitions. Example:

cf.setProviders(providers);

Supported Aggregations Out-Of-The-Box¶

SUM¶

    var metric = cf.Metric("amount","sum");

AVG¶

    var metric = cf.Metric("amount","avg");

MIN¶

    var metric = cf.Metric("amount","min");

MAX¶

    var metric = cf.Metric("amount","max");

PERCENTILES¶

    var metric = cf.Metric('commission', 'percentiles');

COUNT DISTINCT¶

    var metric = cf.Metric("my_attribute","unique");

Dependencies¶

None