geckoboard-python

A Python client library for Geckoboard datasets API.

Installation

Install the package with pip

pip install geckoboard.py

Usage

The latest documentation and user guide can be found on the Geckoboard developer docs

Getting started

Install the python client from PIP

pip install geckoboard.py

Import the Geckoboard package and create an instance of the client using your API key:

import geckoboard

client = geckoboard.client(API_KEY)

Note: You can find your API key by logging into the Geckoboard application and visiting the Account section.

Authentication

Verify that your API key is valid and that you can reach the Geckoboard API with the ping method.

client.ping()

Example:

client('good-api-key').ping() # => true
client('bad-api-key').ping() # => raises Exception

Finding or creating a dataset

Find and verify an existing dataset or create a new dataset with the find_or_create method. unique_by is an optional list of one or more field names whose values will be unique across all your records.

client.datasets.find_or_create(dataset_id, fields, unique_by)

Params:

dataset_id [required] <str>
fields [required] <dict>
- type [required] <str>
- name [required] <str>
- optional <boolean>
unique_by <list>

Example:

dataset = client.datasets.find_or_create('sales.by_night', {
  'amount': { 'type': 'number', 'name': 'Amount', 'optional': False },
  'timestamp': { 'type': 'datetime', 'name': 'Time' }
}, ['timestamp'])

The full list of available field types is described at the top of this page.

Replacing all data in a dataset

Replace all data in the dataset by calling the put method.

dataset.put(items)

Params:

items [required] <list>

Example:

dataset.put([
  { 'timestamp': '2016-01-01T12:00:00Z', 'amount': 819 },
  { 'timestamp': '2016-01-02T12:00:00Z', 'amount': 409 },
  { 'timestamp': '2016-01-03T12:00:00Z', 'amount': 164 }
])

Appending data to a dataset

Append records to a dataset by calling the post method.

Should the number of records in your dataset exceed the limit following a post old records will be discarded.

dataset.post(items, delete_by)

Params:

items [required] <list>
delete_by <str>

Example:

dataset.post([
  { 'timestamp': '2016-01-03T12:00:00Z', 'amount': 312 },
  { 'timestamp': '2016-01-04T12:00:00Z', 'amount': 665 },
  { 'timestamp': '2016-01-05T12:00:00Z', 'amount': 453 }
], 'timestamp')

Deleting a dataset

Delete the dataset and all data with the given id.

client.datasets.delete(dataset_id)

Params:

dataset_id [required] <str>

Example:

client.datasets.delete('sales.gross') # => true

You can also delete a dataset by calling the delete method on a dataset.

dataset = client.datasets.find_or_create(...)
dataset.delete() # => true

Timeout

To enable timeouts for requests package

import geckoboard

client = geckoboard.client(API_KEY, timeout=60)
# or as positional parameter
client = geckoboard.client(API_KEY, 60)

Development

Clone this repo

git clone https://github.com/geckoboard/geckoboard-python && cd geckoboard-python

Install package dependencies

python setup.py develop

Run tests

nosetests