def user_pull():
"""
Pull app creators Mixpanel information.
Parameters
----------
Global Variables
----------
api_creator_secret: Mixpanel Creator Project API key.
Use to make API calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe containing app creator information from Mixpanel.
"""
#generate JQL query
query_user = JQL(api_creator_secret,
people=People({'user_selectors': [{}]
})).group_by(keys=[
"e.properties.$email",
"e.properties.$username",
"e.properties.$distinct_id",
"e.properties.hs_mrr"
],
accumulator=Reducer.count())
#initialize list to track emails, user IDs, distinct IDs, and mrr
email_list = []
user_id_list = []
distinct_id_list = []
hs_mrr_list = []
#process query results
for row in query_user.send():
email_list.append(row['key'][0])
user_id_list.append(row['key'][1])
distinct_id_list.append(row['key'][2])
hs_mrr_list.append(row['key'][3])
#create dataframe
data = {
'email': email_list,
'user_id': user_id_list,
'distinct_id': distinct_id_list,
'hs_mrr': hs_mrr_list
}
df_users = pd.DataFrame(data=data)
return df_users
def test_people_user_selectors(self):
def good_params():
return {
'user_selectors': [{
'selector': 'my selector',
}]
}
# Test valid
People(good_params())
# Bad key types
bad_params = good_params()
bad_params['user_selectors'][0]['selector'] = 3
e = self._try_invalid_people(bad_params)
self.assertEqual(
str(e),
"people_params['user_selectors'][0].selector must be a string")
# Bad key
bad_params = good_params()
bad_params['user_selectors'][0]['mrao'] = 3
e = self._try_invalid_people(bad_params)
self.assertEqual(
str(e),
"'mrao' is not a valid key in people_params['user_selectors'][0]")
def get_new_signup(yesterday):
"""
Get yesterday new signup.
Parameters
----------
yesterday: date
Yesterday's date.
Global Variables
----------
api_creator_secret: Mixpanel Creator Project API key.
Use to make API calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe containing user IDs and emails of creators who signed up yesterday.
"""
# New Signup Web
query = JQL(api_creator_secret,
events=Events({
'event_selectors': [{
'event': "New Signup Web"
}],
'from_date': yesterday,
'to_date': yesterday
}),
people=People({'user_selectors': []}),
join_params={
'type': 'full',
'selectors': [{
'event': "New Signup Web"
}]
})
#store email, user id, and sign up events
email_list = []
userid_list = []
for row in query.send():
if 'user' in list(row.keys()):
if '$username' in list(row['user']['properties'].keys()):
userid_list.append(row['user']['properties']['$username'])
if '$email' in list(row['user']['properties'].keys()):
email_list.append(row['user']['properties']['$email'])
#create dataframe
data = {'user_id': userid_list, 'email': email_list}
df_new_users = pd.DataFrame(data=data)
return df_new_users
def _try_invalid_join(self, params):
try:
JQL(api_secret="asas",
events=Events(),
people=People(),
join_params=params)
self.fail("Expected Events syntax error with params: %s" % params)
except JQLSyntaxError as e:
return e
def creator_pull():
"""
Pull app creators' info in Mixpanel.
Parameters
----------
Global Variables
----------
api_creator_secret: str
Client secret used to make calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe app creators info in Mixpanel.
"""
#generate JQL query
query = JQL(api_creator_secret,
people=People({'user_selectors':
[{}]})).group_by(keys=[
"e.properties.$email",
"e.properties.$username",
"e.properties.hs_owner"
],
accumulator=Reducer.count())
#store emails, user IDs, and user journey stages in lists
email_list = []
user_id_list = []
hs_owner_list = []
for row in query.send():
if row['key'][0] is not None:
email_list.append(row['key'][0])
user_id_list.append(int(row['key'][1]))
hs_owner_list.append(row['key'][2])
#create dataframe
data = {
'email': email_list,
'user_id': user_id_list,
'hs_owner': hs_owner_list
}
df_creators = pd.DataFrame(data=data)
#only keep users with HubSpot owner field. Indicating that it exists in hubspot
df_creators = df_creators[~df_creators.hs_owner.isnull()]
return df_creators
def app_creator_pull():
"""
Pull app creators' info from Mixpanel.
Parameters
----------
Global Variables
----------
api_creator_secret: str
Client secret used to make calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe app creators info in Mixpanel.
"""
#generate JQL query
query_user = JQL(api_creator_secret,
people=People({'user_selectors': [{}]
})).group_by(keys=[
"e.properties.$email",
"e.properties.$username",
"e.properties.$distinct_id",
"e.properties.active_milestone"
],
accumulator=Reducer.count())
#store emails, user IDs, distinct_id, active_milestone in lists
email_list = []
user_id_list = []
distinct_id_list = []
active_milestone_list = []
for row in query_user.send():
if row['key'][1] is not None:
email_list.append(row['key'][0])
user_id_list.append(int(row['key'][1]))
distinct_id_list.append(row['key'][2])
active_milestone_list.append(row['key'][3])
#create dataframe
data = {
'email': email_list,
'app_owner_id': user_id_list,
'distinct_id': distinct_id_list,
'active_milestone': active_milestone_list
}
df_creators = pd.DataFrame(data=data)
return df_creators
def user_pull():
"""
Pull app creators' info in Mixpanel.
Parameters
----------
Global Variables
----------
api_creator_secret: str
Client secret used to make calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe contains app creators info in Mixpanel.
"""
#generate JQL query
query = JQL(api_creator_secret,
people=People({'user_selectors':
[{}]})).group_by(keys=[
"e.properties.$email",
"e.properties.company_domain",
"e.properties.$distinct_id"
],
accumulator=Reducer.count())
#initiate list to store emails, user IDs, and company domains
email_list = []
company_domain_list = []
distinct_id_list = []
#process query results
for row in query.send():
email_list.append(row['key'][0])
company_domain_list.append(row['key'][1])
distinct_id_list.append(row['key'][2])
#create dataframe
data = {
'email': email_list,
'company_domain': company_domain_list,
'distinct_id': distinct_id_list
}
df_users = pd.DataFrame(data=data)
return df_users
def creator_pull():
"""
Pull app creators' info in Mixpanel.
Parameters
----------
Global Variables
----------
api_creator_secret: str
Client secret used to make calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe contains app creators info in Mixpanel.
"""
#generate JQL query
query = JQL(api_creator_secret,
people=People({'user_selectors':
[{}]})).group_by(keys=[
"e.properties.$email",
"e.properties.$username",
"e.properties.user_journey_stage"
],
accumulator=Reducer.count())
#initialize lists to store user IDs, email, and journey stage
email_list = []
user_id_list = []
user_journey_stage_list = []
#process query results
for row in query.send():
email_list.append(row['key'][0])
user_id_list.append(row['key'][1])
user_journey_stage_list.append(row['key'][2])
#create dataframe
data = {
'email': email_list,
'creator_user_id': user_id_list,
'user_journey_stage': user_journey_stage_list
}
df_creators = pd.DataFrame(data=data)
return df_creators
def creator_pull():
"""
Pull app creators' info in Mixpanel.
Parameters
----------
Global Variables
----------
api_creator_secret: str
Client secret used to make calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe app creators info in Mixpanel.
"""
#generate JQL query
query_user = JQL(
api_creator_secret,
people=People({
'user_selectors': [{
}
]
})
).group_by(
keys=[
"e.properties.$distinct_id",
"e.properties.$username"],
accumulator=Reducer.count()
)
#store emails, user IDs, and user journey stages in lists
distinct_id_list = []
user_id_list = []
for row in query_user.send():
distinct_id_list.append(row['key'][0])
user_id_list.append(row['key'][1])
#create dataframe
data = {'distinct_id': distinct_id_list, 'user_id': user_id_list}
df_creators = pd.DataFrame(data=data)
return df_creators
def test_join_types(self):
# Good types
for t in ('full', 'left', 'right', 'inner'):
JQL('some_key',
events=Events(),
people=People(),
join_params={'type': t})
# Bad type
e = self._try_invalid_join({'type': 'mew'})
self.assertEqual(
'"mew" is not a valid join type (valid types: full, left, right, inner)',
str(e))
def pull_creators():
"""
Pull app creators Mixpanel information.
Parameters
----------
Global Variables
----------
api_creator_secret: Mixpanel Creator Project API key
Use to make API calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe containing app creator information from Mixpanel.
"""
#generate JQL query
query_category = JQL(
api_creator_secret, people=People({'user_selectors': [{}]})).group_by(
keys=["e.properties.$email", "e.properties.$distinct_id"],
accumulator=Reducer.count())
#initialize lists to store emails and distinct IDs
email_list = []
distinct_id_list = []
#process query response
for row in query_category.send():
email_list.append(row['key'][0])
distinct_id_list.append(row['key'][1])
#create dataframe
data = {'email': email_list, 'mixpanel_distinct_id': distinct_id_list}
df_creators_mixpanel = pd.DataFrame(data=data)
#remove creators with missing information
df_creators_mixpanel = df_creators_mixpanel.dropna()
return df_creators_mixpanel
def user_pull():
"""
Pull app creators Mixpanel information.
Parameters
----------
Global Variables
----------
api_creator_secret: Mixpanel Creator Project API key.
Use to make API calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe containing app creator information from Mixpanel.
"""
#generate JQL query
query_user = JQL(api_creator_secret,
people=People({'user_selectors': [{}]
})).group_by(keys=[
"e.properties.$email",
"e.properties.$username"
],
accumulator=Reducer.count())
#initialize lists to record app creator information
email_list = []
user_id_list = []
for row in query_user.send():
email_list.append(row['key'][0])
user_id_list.append(row['key'][1])
#create dataframe
data = {'email': email_list, 'user_id': user_id_list}
df_users = pd.DataFrame(data=data)
return df_users
def creator_pull():
"""
Pull app creators' info in Mixpanel.
Parameters
----------
Global Variables
----------
api_creator_secret: str
Client secret used to make calls to Mixpanel Creator Project.
Returns
----------
dataframe
Dataframe app creators info in Mixpanel.
"""
#generate JQL query
query_user = JQL(api_creator_secret,
people=People({'user_selectors': [{}]
})).group_by(keys=[
"e.properties.$email",
"e.properties.$unsubscribed"
],
accumulator=Reducer.count())
#store emails, user IDs, and user journey stages in lists
email_list = []
unsubscribed_list = []
for row in query_user.send():
if ((row['key'][0] is not None) & (row['key'][1] is not None)
): #only keep accounts with both email and unsubscribe status
email_list.append(row['key'][0])
unsubscribed_list.append(row['key'][1])
#create dataframe
data = {'email': email_list, 'unsubscribed': unsubscribed_list}
df_creators = pd.DataFrame(data=data)
return df_creators
def app_user_pull():
"""
Pull app users' information in Mixpanel.
Parameters
----------
Global Variables
----------
api_user_secret: str
Client secret used to make calls to Mixpanel User Project.
Returns
----------
dataframe
Dataframe contains app users' information in Mixpanel.
"""
#generate JQL query
query_user = JQL(api_user_secret, people=People({
'user_selectors': [{}]
})).group_by(keys=["e.properties.$email", "e.properties.$username"],
accumulator=Reducer.count())
#initialize lists to record email and user ID
email_list = []
user_id_list = []
#process query results
for row in query_user.send():
email_list.append(row['key'][0])
user_id_list.append(row['key'][1])
#create dataframe
data = {'user_email': email_list, 'app_user_id': user_id_list}
df_app_users = pd.DataFrame(data=data)
return df_app_users
def creator_pull():
#generate JQL query
query = JQL(api_creator_secret,
people=People({'user_selectors': [{}]})).group_by(
keys=["e.properties.$email", "e.properties.$distinct_id"],
accumulator=Reducer.count())
#store emails, user IDs, and user journey stages in lists
email_list = []
distinct_id_list = []
for row in query.send():
if row['key'][0] is not None:
email_list.append(row['key'][0])
distinct_id_list.append(row['key'][1])
#create dataframe
data = {'email': email_list, 'distinct_id': distinct_id_list}
df_creators = pd.DataFrame(data=data)
return df_creators
def test_join_selectors(self):
def good_params():
return {
'selectors': [{
'event': 'my_event',
'selector': 'my selector'
}]
}
# Test valid
JQL('some_api_key',
events=Events(),
people=People(),
join_params=good_params())
# Bad array
bad_params = good_params()
bad_params['selectors'] = 3
e = self._try_invalid_join(bad_params)
self.assertEqual(str(e), "join_params['selectors'] must be iterable")
# Bad key types
for key in ('event', 'selector'):
bad_params = good_params()
bad_params['selectors'][0][key] = 3
e = self._try_invalid_join(bad_params)
self.assertEqual(
str(e),
"join_params['selectors'][0].%s must be a string" % key)
# Bad key
bad_params = good_params()
bad_params['selectors'][0]['mrao'] = 3
e = self._try_invalid_join(bad_params)
self.assertEqual(
str(e), "'mrao' is not a valid key in join_params['selectors'][0]")
def test_query_plan(self):
with warnings.catch_warnings(record=True) as w:
q = JQL('key', events=Events(), people=People())
q.query_plan()
self.assertIs(w[-1].category, DeprecationWarning)
self.assertIn('query_plan', str(w[-1].message))
def _try_invalid_people(self, params):
try:
People(params)
self.fail("Expected People syntax error with params: %s" % params)
except JQLSyntaxError as e:
return e
def app_user_pull():
"""
Pull app users' info in Mixpanel.
Parameters
----------
Global Variables
----------
api_user_secret: str
Client secret used to make calls to Mixpanel User Project.
Returns
----------
dataframe
Dataframe contains app users info in Mixpanel.
"""
#generate JQL query
query = JQL(api_user_secret, people=People(
{'user_selectors': [{}]})).group_by(
keys=[
"e.properties.$email",
"e.properties.$username",
"e.properties.$distinct_id",
"e.properties.creator", #pulling current creator status
"e.properties.active_user"
], #pulling current active user status
accumulator=Reducer.count())
#initialize lists to record email, user ID, creator status, active user status, and distinct ID
email_list = []
user_id_list = []
creator_list = []
active_user_list = []
distinct_id_list = []
#process query results
for row in query.send():
email_list.append(row['key'][0])
user_id_list.append(row['key'][1])
distinct_id_list.append(row['key'][2])
creator_list.append(row['key'][3])
active_user_list.append(row['key'][4])
#create dataframe
data = {
'email': email_list,
'app_user_id': user_id_list,
'distinct_id': distinct_id_list,
'creator_current': creator_list,
'active_user_current': active_user_list
}
df_app_users = pd.DataFrame(data=data)
df_app_users = df_app_users.dropna()
df_app_users.app_user_id = df_app_users.app_user_id.astype(int)
return df_app_users
