Flask-BigTempo

Flask extension offering several utilities for creating bigtempo servers.

Installing

pip should do the job:

$ pip install flask-bigtempo

There is a requirements.txt file is you want to checkout the source code directly.

Datastore API

It is meant to store timeseries data.

Each timeseries is identified by the conjunction of an reference and a symbol. It is structured this way so that the source (or type) of the data can be declared as the reference. Example:

• While in the stockmarket context, the reference can be NASDAQ while symbol is left for the company stock.
• Storing country 'UN Human Development Index' the reference can be HDI while the symbol would take a country's name or code.

Here you can find:

• A Storage implementation that offers methods to save / update, retrieve and delete pandas dataframes
• A flask extension that exposes an REST API that handles data as json
• A REST client that can communicate with the REST API
• A command line script that enables shell usage of the REST API
• Some bigtempo datasources that allows easy integration, after all, store api was conceived exactly to serve data to bigtempo.

Storage implementation

For the moment the is only one implementation based on SQLAlchemy. You can find it at flask_bigtempo/store/storages.py. Example usage can be found flask_bigtempo/store/clients.py

The flask extension:

You can easily have your flask server expose bigtempo store api:

#!/usr/bin/env python


from flask import Flask
from flask.ext.sqlalchemy import SQLAlchemy
from flask.ext.bigtempo import DatastoreAPI


app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite://'

db = SQLAlchemy(app)

# The datastore api needs flask's app instance and a sqlalchemy engine
datastore = DatastoreAPI(app, db.engine)


@app.route('/')
def hello_world():
    return '''
           <h1>Welcome!</h1>
           The routes for datastore can be found at "/api/store/"<br/>
           '''


if __name__ == '__main__':
    app.run(debug=True)

The following methods are made available:

• Data retrieval: GET /api/store/{reference}/{symbol}
• Data insertion: PUT /api/store/{reference}/{symbol}
• Data deletion: DELETE /api/store/{reference}/{symbol}

Optionally, you can use aditional url parameters:

• json_format (eg.: ?json_format=index).
• date_format (eg.: ?date_format=iso). The formats available are the same provided by the pandas to_json and read_json methods.

REST Clients

You can find them at flask_bigtempo/store/clients.py:

• DFStoreRestClient works with Dataframes as input and output;
• JSONStoreRestClient works with JSON as input and output;

Using it should be as simple as:

import flask_bigtempo.store.clients as store_client

api = store_client.DFStoreRestClient()
dataframe = api.retrieve('HDI', 'Brazil')

CL Script

Its code is available at the scripts directory. As soon as you install this lib at your computer, store_api should be available on the PATH.

You can learn more about its usage by executing store_api -h

Bigtempo DataSources

Available at flask_bigtempo/store/datasources.py.

You can import it by:

import flask_bigtempo.store.datasources as datasources

ds = datasources.RESTStoreDatasource('example')

And all that is left is to register it to your bigtempo engine.