dataset_example.py

from moztelemetry.dataset import Dataset

# Let's start selecting the `telemetry` dataset.
# This will load all the metadata about available dimensions and file locations.
dataset = Dataset.from_source('telemetry')

#The list of dimensions is now available on the `schema` attribute.
assert dataset.schema == [
    u'submissionDate',
    u'sourceName',
    u'sourceVersion',
    u'docType',
    u'appName',
    u'appUpdateChannel',
    u'appVersion',
    u'appBuildId'
]

# Filter values can be either strings or python functions
dataset = (dataset.where(docType='main')
                  .where(appName='Firefox')
                  .where(appUpdateChannel='nightly')
                  .where(submissionDate=lambda x: x.startswith('2016091')))

# Retrieve 10% of the files
records = dataset.records(sc, sample=.1)