def createAnalyticsReport():
#The real code that initalized the client
credentials = client.GoogleCredentials(access_token=ACCESS_TOKEN,
refresh_token=REFRESH_TOKEN,
client_id=CLIENT_ID,
client_secret=CLIENT_SECRET,
token_uri=TOKEN_URI,
token_expiry=TOKEN_EXPIRY,
user_agent=USER_AGENT)
#Initialize Http Protocol
http = lib2.Http()
#Authorize client
authorized = credentials.authorize(http)
#API Name and Verison, these don't change until
#they release a new API version for us to play with.
api_name = 'analyticsreporting'
api_version = 'v4'
#Let's build the client
analytics = google_build(serviceName=api_name,
version=api_version,
http=authorized)
return analytics
def create_table(project_id, table_id, dataset_id, debug=False):
"""Create a BigQuery table using a schema compatible with Parse.ly events
:param project_id: The BigQuery project ID to write to
:type project_id: str
:param table_id: The BigQuery table ID to write to
:type table_id: str
:param dataset_id: The BigQuery dataset ID to write to
:type dataset_id: str
"""
fields = mk_bigquery_schema()
schema = {
"description": "Parse.ly Data Pipeline",
"schema": {"fields": fields},
"tableReference": {
"projectId": project_id,
"tableId": table_id,
"datasetId": dataset_id
}
}
if debug:
print("Running the following BigQuery JSON table insert:")
print(json.dumps(schema, indent=4, sort_keys=True))
credentials = GoogleCredentials.get_application_default()
bigquery = google_build('bigquery', 'v2', credentials=credentials)
bigquery.tables().insert(projectId=project_id,
datasetId=dataset_id,
body=schema).execute()
def __init__(self): super().__init__() self.svc = google_build( 'customsearch', 'v1', developerKey='redacted' )
def copy_from_s3(network,
s3_prefix="",
access_key_id="",
secret_access_key="",
region_name="us-east-1",
project_id=None,
dataset_id=None,
table_id=None,
dry_run=False):
"""Load events from S3 to BigQuery using the BQ streaming insert API.
:param network: The Parse.ly network for which to perform writes (eg
"parsely-blog")
:type network: str
:param s3_prefix: The S3 timestamp directory prefix from which to fetch data
batches, formatted as YYYY/MM/DD
:type s3_prefix: str
:param access_key_id: The AWS access key to use when fetching data batches
:type access_key_id: str
:param secret_access_key: The AWS secret key to use when fetching data batches
:type secret_access_key: str
:param region_name: The AWS region in which to perform fetches
:type region_name: str
:param project_id: The BigQuery project ID to write to
:type project_id: str
:param dataset_id: The BigQuery dataset ID to write to
:type dataset_id: str
:param table_id: The BigQuery table ID to write to
:type table_id: str
:param dry_run: If True, don't perform BigQuery writes
:type dry_run: bool
"""
bq_conn = None
if not dry_run:
bq_conn = google_build(
'bigquery', 'v2',
credentials=GoogleCredentials.get_application_default())
s3_stream = events_s3(network, prefix=s3_prefix, access_key_id=access_key_id,
secret_access_key=secret_access_key,
region_name=region_name)
schema_compliant_fields = [column['name'] for column in mk_bigquery_schema()]
def schema_compliant(jsonline):
return {k: jsonline.get(k, None) for k in schema_compliant_fields}
def chunked(seq, chunk_size):
chunk = []
for item in seq:
chunk.append(schema_compliant(item))
if len(chunk) >= chunk_size:
yield chunk
chunk = []
if chunk:
yield chunk
for events in chunked(s3_stream, 500):
streaming_insert_bigquery(events, bq_conn=bq_conn, project_id=project_id,
dataset_id=dataset_id, table_id=table_id)
def load_batch_bigquery(network,
s3_prefix="",
access_key_id="",
secret_access_key="",
region_name="us-east-1",
project_id=None,
dataset_id=None,
table_id=None,
dry_run=False):
"""Load a batch of events from S3 to BigQuery
:param network: The Parse.ly network for which to perform writes (eg
"parsely-blog")
:type network: str
:param s3_prefix: The S3 timestamp directory prefix from which to fetch data
batches, formatted as YYYY/MM/DD
:type s3_prefix: str
:param access_key_id: The AWS access key to use when fetching data batches
:type access_key_id: str
:param secret_access_key: The AWS secret key to use when fetching data batches
:type secret_access_key: str
:param region_name: The AWS region in which to perform fetches
:type region_name: str
:param project_id: The BigQuery project ID to write to
:type project_id: str
:param dataset_id: The BigQuery dataset ID to write to
:type dataset_id: str
:param table_id: The BigQuery table ID to write to
:type table_id: str
:param dry_run: If True, don't perform BigQuery writes
:type dry_run: bool
"""
bq_conn = None
if not dry_run:
bq_conn = google_build(
'bigquery',
'v2',
credentials=GoogleCredentials.get_application_default())
s3_stream = events_s3(network,
prefix=s3_prefix,
access_key_id=access_key_id,
secret_access_key=secret_access_key,
region_name=region_name)
def chunked(seq, chunk_size):
chunk = []
for item in seq:
chunk.append(item)
if len(chunk) >= chunk_size:
yield chunk
chunk = []
for events in chunked(s3_stream, 500):
write_events_bigquery(events,
bq_conn=bq_conn,
project_id=project_id,
dataset_id=dataset_id,
table_id=table_id)
refresh_token = config.refresh_token
client_id = config.client_id
client_secret = config.client_secret
credentials = google.oauth2.credentials.Credentials(
None,
refresh_token=refresh_token,
token_uri=token_uri,
client_id=client_id,
client_secret=client_secret)
api_name = 'analyticsreporting'
api_version = 'v4'
api_client = google_build(serviceName=api_name,
version=api_version,
credentials=credentials)
sample_request = {
'viewId':
viewId,
'dateRanges': {
'startDate': theepoch,
'endDate': theepoch
},
'dimensions': [{
'name': 'ga:campaign'
}, {
'name': 'ga:devicecategory'
}, {
'name': 'ga:city'
def pull_data(last_date):
access_token = new_access_code()
credentials = client.GoogleCredentials(
access_token=access_token,
refresh_token=refresh_token,
client_id=client_id,
client_secret=CLIENT_SECRETS_FILE,
token_uri="https://www.googleapis.com/oauth2/v4/token",
token_expiry=token_expiry,
user_agent=user_agent)
http = lib2.Http()
#Authorize client
authorized = credentials.authorize(http)
#Let's build the client
api_client = google_build(serviceName=API_SERVICE_NAME,
version=API_VERSION,
http=authorized)
#Specify which data you want to pull from Google Analytics
sample_request = {
'viewId':
'83705367',
'dateRanges': {
'startDate':
last_date,
'endDate':
datetime.strftime(datetime.now() - timedelta(days=1), '%Y-%m-%d')
},
'dimensions': [{
'name': 'ga:adContent'
}, {
'name': 'ga:date'
}, {
'name': 'ga:campaign'
}],
'metrics': [{
'expression': 'ga:users'
}, {
'expression': 'ga:newUsers'
}, {
'expression': 'ga:sessions'
}, {
'expression': 'ga:bounceRate'
}, {
'expression': 'ga:pageviewsPerSession'
}, {
'expression': 'ga:avgSessionDuration'
}, {
'expression': 'ga:goal1ConversionRate'
}, {
'expression': 'ga:goal1Completions'
}, {
'expression': 'ga:goal1Value'
}],
'filtersExpression':
'ga:adContent=@_;ga:adContent!@GDN;ga:campaign!=id',
'orderBys': {
'fieldName': 'ga:date',
'sortOrder': 'ASCENDING'
},
'pageSize':
100000,
'includeEmptyRows':
True
}
response = api_client.reports().batchGet(body={
'reportRequests': sample_request
}).execute()
return response
def create_bigquery_table(project_id, table_id, dataset_id):
"""Create a BigQuery table using a schema compatible with Parse.ly events
:param project_id: The BigQuery project ID to write to
:type project_id: str
:param table_id: The BigQuery table ID to write to
:type table_id: str
:param dataset_id: The BigQuery dataset ID to write to
:type dataset_id: str
"""
schema = {
"description": "Parse.ly event data",
"schema": {
"fields": [{
"name": "url",
"mode": "REQUIRED",
"type": "STRING"
}, {
"name": "apikey",
"mode": "REQUIRED",
"type": "STRING"
}, {
"name": "action",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "display_avail_height",
"mode": "NULLABLE",
"type": "INTEGER"
}, {
"name": "display_avail_width",
"mode": "NULLABLE",
"type": "INTEGER"
}, {
"name": "display_pixel_depth",
"mode": "NULLABLE",
"type": "INTEGER"
}, {
"name": "display_total_height",
"mode": "NULLABLE",
"type": "INTEGER"
}, {
"name": "display_total_width",
"mode": "NULLABLE",
"type": "INTEGER"
}, {
"name": "engaged_time_inc",
"mode": "NULLABLE",
"type": "INTEGER"
}, {
"name": "extra_data",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "referrer",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "session_id",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "session_initial_referrer",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "session_initial_url",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "session_last_session_timestamp",
"mode": "NULLABLE",
"type": "TIMESTAMP"
}, {
"name": "session_timestamp",
"mode": "NULLABLE",
"type": "TIMESTAMP"
}, {
"name": "timestamp_info_nginx_ms",
"mode": "NULLABLE",
"type": "TIMESTAMP"
}, {
"name": "timestamp_info_override_ms",
"mode": "NULLABLE",
"type": "TIMESTAMP"
}, {
"name": "timestamp_info_pixel_ms",
"mode": "NULLABLE",
"type": "TIMESTAMP"
}, {
"name": "user_agent",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "visitor_ip",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "visitor_network_id",
"mode": "NULLABLE",
"type": "STRING"
}, {
"name": "visitor_site_id",
"mode": "NULLABLE",
"type": "STRING"
}]
},
"tableReference": {
"projectId": project_id,
"tableId": table_id,
"datasetId": dataset_id
}
}
credentials = GoogleCredentials.get_application_default()
bigquery = google_build('bigquery', 'v2', credentials=credentials)
bigquery.tables().insert(projectId=project_id,
datasetId=dataset_id,
tableId=table_id,
body=schema).execute()
