Official Python client for the Delighted API.
pip install --upgrade delighted
or
easy_install --upgrade delighted
If you previously used the python package named delighted-python, please note that the package name is now just delighted.
To get started, you need to configure the client with your secret API key.
import delighted
delighted.api_key = 'YOUR_API_KEY'For further options, read the advanced configuration section.
Note: Your API key is secret, and you should treat it like a password. You can find your API key in your Delighted account, under Settings > API.
Adding/updating people and scheduling surveys:
# Add a new person, and schedule a survey immediately
person1 = delighted.Person.create(email='[email protected]')
# Add a new person, and schedule a survey after 1 minute (60 seconds)
person2 = delighted.Person.create(email='[email protected]', delay=60)
# Add a new person, but do not schedule a survey
person3 = delighted.Person.create(email='[email protected]', send=False)
# Add a new person with full set of attributes, including a custom question
# product name, and schedule a survey with a 30 second delay
person4 = delighted.Person.create(
email='[email protected]',
name='Joe Bloggs',
properties={'customer_id': 123, 'country': 'USA',
'question_product_name': 'The London Trench'},
delay=30)
# Update an existing person (identified by email), adding a name, without
# scheduling a survey
updated_person1 = delighted.Person.create(email='[email protected]',
name='James Scott', send=False)Unsubscribing people:
# Unsubscribe an existing person
delighted.Unsubscribe.create(person_email='[email protected]')Listing people:
# List all people, auto pagination
# Note: Make sure to handle the possible rate limits error
people = delighted.Person.list()
while True:
try:
for person in people.auto_paging_iter():
# Do something with person
except TooManyRequestsError as e:
# Indicates how long to wait (in seconds) before making this request again
e.retry_after
continue
# For convenience, this method can use a sleep to automatically handle rate limits
people = delighted.Person.list(auto_handle_rate_limits=True)
for person in people.auto_paging_iter():
# Do something with personListing people who have unsubscribed:
# List all people who have unsubscribed, 20 per page, first 2 pages
delighted.Unsubscribe.all()
delighted.Unsubscribe.all(page=2)Listing people whose emails have bounced:
# List all people whose emails have bounced, 20 per page, first 2 pages
delighted.Bounce.all()
delighted.Bounce.all(page=2)Deleting a person and all of the data associated with them:
# Delete by person id
delighted.Person.delete(id=42)
# Delete by email address
delighted.Person.delete(email='[email protected]')
# Delete by phone number (must be E.164 format)
delighted.Person.delete(phone_number='+14155551212')Deleting pending survey requests
# Delete all pending (scheduled but unsent) survey requests for a person, by email.
delighted.SurveyRequest.delete_pending(person_email='[email protected]')Adding survey responses:
# Add a survey response, score only
survey_response1 = delighted.SurveyResponse.create(person=person1.id,
score=10)
# Add *another* survey response (for the same person), score and comment
survey_response2 = delighted.SurveyResponse.create(person=person1.id,
score=5,
comment='Really nice.')Retrieving a survey response:
# Retrieve an existing survey response
survey_response3 = delighted.SurveyResponse.retrieve('123')Updating survey responses:
# Update a survey response score
survey_response4 = delighted.SurveyResponse.retrieve('234')
survey_response4.score = 10
survey_response4.save()
# <delighted.SurveyResponse object at 0xabc123>
# Update (or add) survey response properties
survey_response4.person_properties = {'segment': 'Online'}
survey_response4.save()
# <delighted.SurveyResponse object at 0xabc123>
# Update person who recorded the survey response
survey_response4.person = '321'
survey_response4.save()
# <delighted.SurveyResponse object at 0xabc123>Listing survey responses:
# List all survey responses, 20 per page, first 2 pages
survey_responses_page1 = delighted.SurveyResponse.all()
survey_responses_page2 = delighted.SurveyResponse.all(page=2)
# List all survey responses, 20 per page, expanding person object
survey_responses_page1_expanded = delighted.SurveyResponse.all(expand=['person'])
survey_responses_page1_expanded[0].person
# <delighted.Person object at 0xabc123>
# List all survey responses, 20 per page, for a specific trend (ID: 123)
survey_responses_page1_trend = delighted.SurveyResponse.all(trend='123')
# List all survey responses, 20 per page, in reverse chronological order (newest first)
survey_responses_page1_desc = delighted.SurveyResponse.all(order='desc')
# List all survey responses, 100 per page, page 5, with a time range
import pytz
timezone = pytz.timezone('America/Chicago')
filtered_survey_responses = delighted.SurveyResponse.all(
page=5,
per_page=100,
since=timezone.localize(datetime.datetime(2014, 3, 1)),
until=timezone.localize(datetime.datetime(2014, 4, 30))
)Retrieving metrics:
# Get current metrics, 30-day simple moving average, from most recent response
metrics = delighted.Metrics.retrieve()
# Get current metrics, 30-day simple moving average, from most recent response,
# for a specific trend (ID: 123)
metrics = delighted.Metrics.retrieve(trend='123')
# Get metrics, for given time range
import pytz
timezone = pytz.timezone('America/Chicago')
metrics = delighted.Metrics.retrieve(
since=timezone.localize(datetime.datetime(2013, 10, 1)),
until=timezone.localize(datetime.datetime(2013, 11, 1))
)Managing Autopilot:
# Get Autopilot configuration for the `email` platform
autopilot = delighted.AutopilotConfiguration.retrieve('email')
# List people in AutopilotMembership for the `email` platform
people_autopilot = delighted.AutopilotMembership.forEmail().list(auto_handle_rate_limits=True)
for person in people_autopilot.auto_paging_iter():
# Do something with person
# Add people to AutopilotMembership
autopilot = delighted.AutopilotMembership.forEmail().create(person_email='[email protected]')
# Add people to AutopilotMembership, with a full set of attributes
properties = {'customer_id': 123, 'country': 'USA', 'question_product_name': 'The London Trench'}
autopilot = delighted.AutopilotMembership.forSms().create(person_phone_number='+14155551212', properties=properties)
# Delete by person id
delighted.AutopilotMembership.forSms().delete(person_id=42)
# Delete by email address
delighted.AutopilotMembership.forEmail().delete(person_email='[email protected]')
# Delete by phone number (must be E.164 format)
delighted.AutopilotMembership.forSms().delete(person_phone_number='+14155551212')If a request is rate limited, a TooManyRequestsError exception is raised. You can rescue that exception to implement exponential backoff or retry strategies. The exception provides a retry_after attribute to tell you how many seconds you should wait before retrying. For example:
try:
metrics = delighted.Metrics.retrieve()
except delighted.errors.TooManyRequestsError as err:
retry_after_seconds = err.retry_after
# wait for retry_after_seconds before retrying
# add your retry strategy here ...The following options are configurable for the client:
delighted.api_key
delighted.api_base_url # default: 'https://api.delighted.com/v1/'
delighted.http_adapter # default: delighted.HTTPAdapterBy default, a shared instance of delighted.Client is created lazily in delighted.get_shared_client(). If you want to create your own client, perhaps for test or if you have multiple API keys, you can:
# Create an custom client instance, and pass as last argument to resource actions
import delighted
from delighted import Client
client = Client(api_key='API_KEY',
api_base_url='https://api.delighted.com/v1/',
http_adapter=HTTPAdapter())
metrics_from_custom_client = delighted.Metrics.retrieve(client=client)
# Or, you can set Delighted.shared_client yourself
delighted.shared_client = delighted.Client(
api_key='API_KEY',
api_base_url='https://api.delighted.com/v1/',
http_adapter=delighted.HTTPAdapter()
)
metrics_from_custom_shared_client = delighted.Metrics.retrieve()- 2.6+, 3.3+ (PyPy supported)
- Fork it
- Create your feature branch (
git checkout -b my-new-feature) - Run the tests (
tox) - Commit your changes (
git commit -am 'Add some feature') - Push to the branch (
git push origin my-new-feature) - Create new Pull Request
- Bump the version in
delighted/__init__.py. - Update the README and CHANGELOG as needed.
- Tag the commit for release.
- Create the distribution
python setup.py sdist - Update the package against PyPI's test server with twine
twine upload --repository-url https://test.pypi.org/legacy/ dist/TEST_PACKAGE_NAME. - If (4 and 5) work, repeat all steps, then push to PyPI's live servers with
twine upload dist/PACKAGE_NAME.
Originally by Jason Pearson. Graciously transfered and now officially maintained by Delighted.