for numpy, pandas, and scipy (in requirements.pip):

on Arch Linux: # pacman -S blas lapack gcc-fortran

on Debian based: # apt-get install gfortran libatlas-base-dev

Installation

$ pip install bamboo-data

Python pip package for bamboo.

Usage

import bamboo as bm
from bamboo.lib.io import create_dataset_from_url

bf = bm.BambooFrame([{'date': '2012-12-21'}])
bff = bf.recognize_dates()
bff.to_json()

>>> '[{"date": {"$date": 1356048000000}}]'

# Turn asyncronous processing off
bm.set_async(False)

url = 'http://formhub.org/mberg/forms/good_eats/data.csv'
dataset = create_dataset_from_url(url)
dataset.schema

>>> {u'_gps_altitude': {u'cardinality': 14, u'label': u'_gps_altitude', ...

On the remote server

running the example basic commands

$ ./scripts/commands.sh

using bamboo.JS

var dataset = new bamboo.Dataset({url: 'http://bitly.com/ZfzBwP'});
bamboo.dataset_exists('nonexistentdataset_id');
dataset.query_info();
...

using pybamboo

from pybamboo import PyBamboo
pybamboo = PyBamboo()
response = pybamboo.store_csv_file('http://formhub.org/mberg/forms/good_eats/data.csv')
dataset_id = response['id']
...

posting a dataset

$ curl -X POST -d "url=http://formhub.org/mberg/forms/good_eats/data.csv" http://bamboo.io/datasets

Viewing Documentation

The latest docs are available at http://bamboo.io/