def run(argv=None):
known_args, pipeline_options, avro_schema = generate_args(
job_name='computronix-domi-permits',
bucket='{}_computronix'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='domi_permits_computronix')
with beam.Pipeline(options=pipeline_options) as p:
field_name_swaps = [('PERMITTYPEPERMITTYPE', 'PERMITTYPE'),
('TYPEOFWORKDESCRIPTION', 'WORKTYPE'),
('APPLICANTCUSTOMFORMATTEDNAME', 'APPLICANTNAME'),
('ALLCONTRACTORSNAME', 'CONTRACTORNAME'),
('SPECIALPERMITINSTRUCTIONS',
'SPECIALINSTRUCTIONS'),
('STATUSDESCRIPTION', 'STATUS')]
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (lines
| beam.ParDo(ParseNestedFields())
| beam.ParDo(SwapFieldNames(field_name_swaps))
| beam.ParDo(GeocodeAddress())
| WriteToAvro(known_args.avro_output,
schema=avro_schema,
file_name_suffix='.avro',
use_fastavro=True))
def run(argv=None):
"""
If you want to run just this file for rapid development, pass the arg '-r DirectRunner' and add
GCS paths for --input and --avro_output, e.g.
python qalert_requests_dataflow.py --input gs://pghpa_test_qalert/requests/2020/06/2020-06-17_requests.json
--avro_output gs://pghpa_test_qalert/requests/avro_output/2020/06/2020-06-17/ -r DirectRunner
"""
known_args, pipeline_options, avro_schema = generate_args(
job_name='wprdc-fire-dataflow',
bucket='{}_ems_fire'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='ems_calls')
with beam.Pipeline(options=pipeline_options) as p:
field_name_swaps = [("census_block_group_center__x", "long"),
("census_block_group_center__y", "lat")]
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (lines
| beam.ParDo(SwapFieldNames(field_name_swaps))
| WriteToAvro(known_args.avro_output,
schema=avro_schema,
file_name_suffix='.avro',
use_fastavro=True))
def run(argv=None):
"""
If you want to run just this file for rapid development, change runner to 'DirectRunner' and add
GCS paths for --input and --avro_output, e.g.
python qalert_requests_dataflow.py --input gs://pghpa_test_qalert/requests/2020/06/2020-06-17_requests.json
--avro_output gs://pghpa_test_qalert/requests/avro_output/2020/06/2020-06-17/
"""
known_args, pipeline_options, avro_schema = generate_args(
job_name='parking-meters-dataflow',
bucket='{}_parking'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='parking_meters'
)
with beam.Pipeline(options=pipeline_options) as p:
field_name_swaps = [("longitude", "long"), ("latitude", "lat")]
type_changes = [("long", "float"), ("lat", "float")]
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (
lines
| beam.ParDo(SwapFieldNames(field_name_swaps))
| beam.ParDo(ChangeDataTypes(type_changes))
| WriteToAvro(known_args.avro_output, schema=avro_schema, file_name_suffix='.avro', use_fastavro=True))
def run(argv=None):
known_args, pipeline_options, avro_schema = generate_args(
job_name='accela-permits',
bucket='{}_accela'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='accela_permits'
)
with beam.Pipeline(options=pipeline_options) as p:
exclude_fields = [
'module',
'serviceProviderCode',
'undistributedCost',
'totalJobCost',
'recordClass',
'reportedChannel',
'closedByDepartment',
'estimatedProductionUnit',
'actualProductionUnit',
'createdByCloning',
'closedByUser',
'trackingId',
'initiatedProduct',
'createdBy',
'value',
'balance',
'booking',
'infraction',
'misdemeanor',
'offenseWitnessed',
'defendantSignature',
'parcels',
'id',
'statusDate',
'jobValue',
'reportedDate'
]
address_field = 'address'
field_name_swaps = [
('customId', 'id'),
('totalPay', 'total_paid')
]
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (
lines
| beam.ParDo(FilterInvalidRecord())
| beam.ParDo(FilterFields(exclude_fields))
| beam.ParDo(ParseNestedFields())
| beam.ParDo(GeocodeAddress(address_field))
| beam.ParDo(SwapFieldNames(field_name_swaps))
| beam.ParDo(ColumnsCamelToSnakeCase())
| WriteToAvro(known_args.avro_output, schema=avro_schema, file_name_suffix='.avro', use_fastavro=True))
def run(argv=None):
known_args, pipeline_options, avro_schema = generate_args(
job_name='twilio-311-dataflow',
bucket='{}_twilio'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='twilio_reports'
)
with beam.Pipeline(options=pipeline_options) as p:
# Read the text file[pattern] into a PCollection.
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (
lines
| WriteToAvro(known_args.avro_output, schema=avro_schema, file_name_suffix='.avro', use_fastavro=True))
def run(argv=None):
known_args, pipeline_options, avro_schema = generate_args(
job_name='firearms-dataflow',
bucket='{}_firearm_seizures'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='firearm_seizures'
)
with beam.Pipeline(options=pipeline_options) as p:
# Read the text file[pattern] into a PCollection.
lines = p | ReadFromText(known_args.input, skip_header_lines=1)
load = (
lines
| beam.ParDo(ConvertToDicts())
| WriteToAvro(known_args.avro_output, schema=avro_schema, file_name_suffix='.avro', use_fastavro=True))
def run(argv=None):
known_args, pipeline_options, avro_schema = generate_args(
job_name='computronix-trades-dataflow',
bucket='{}_computronix'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='trade_licenses_computronix')
with beam.Pipeline(options=pipeline_options) as p:
# Read the text file[pattern] into a PCollection.
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (lines
| beam.ParDo(FormatColumnNames())
| beam.ParDo(ConvertTypes())
| WriteToAvro(known_args.avro_output,
schema=avro_schema,
file_name_suffix='.avro',
use_fastavro=True))
def run(argv=None):
known_args, pipeline_options, avro_schema = dataflow_utils.generate_args(
job_name='comm-ctr-attendance-dataflow',
bucket='{}_community_centers'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='community_center_attendance')
with beam.Pipeline(options=pipeline_options) as p:
# Read the text file[pattern] into a PCollection.
lines = p | ReadFromText(known_args.input,
coder=dataflow_utils.JsonCoder())
load = (lines
| beam.ParDo(ColumnsCamelToSnakeCase())
| WriteToAvro(known_args.avro_output,
schema=avro_schema,
file_name_suffix='.avro',
use_fastavro=True))
def run(argv=None):
known_args, pipeline_options, avro_schema = generate_args(
job_name='registered-businesses-dataflow',
bucket='{}_finance'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='registered_businesses')
with beam.Pipeline(options=pipeline_options) as p:
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (lines
| beam.ParDo(ColumnsToLowerCase())
| beam.ParDo(ParseAddress())
| beam.ParDo(
NormalizeAddress(StaticValueProvider(str, 'address_full')))
| WriteToAvro(known_args.avro_output,
schema=avro_schema,
file_name_suffix='.avro',
use_fastavro=True))
def run(argv=None):
"""
If you want to run just this file for rapid development, change runner to 'DirectRunner' and add
GCS paths for --input and --avro_output, e.g.
python qalert_requests_dataflow.py --input gs://pghpa_test_qalert/requests/2020/06/2020-06-17_requests.json
--avro_output gs://pghpa_test_qalert/requests/avro_output/2020/06/2020-06-17/
"""
known_args, pipeline_options, avro_schema = generate_args(
job_name='parking-transactions-dataflow',
bucket='{}_parking'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='parking_transactions'
)
with beam.Pipeline(options=pipeline_options) as p:
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (
lines
| WriteToAvro(known_args.avro_output, schema=avro_schema, file_name_suffix='.avro', use_fastavro=True))
def run(argv=None):
known_args, pipeline_options, avro_schema = generate_args(
job_name='police-blotter-30-day-dataflow',
bucket='{}_police'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='30_day_police_blotter')
with beam.Pipeline(options=pipeline_options) as p:
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
data_type_changes = [('CCR', 'int'), ('TRACT', 'int')]
field_name_swaps = [('PK', 'id')]
load = (lines
| beam.ParDo(CleanPKs())
| beam.ParDo(ChangeDataTypes(data_type_changes))
| beam.ParDo(SwapFieldNames(field_name_swaps))
| WriteToAvro(known_args.avro_output,
schema=avro_schema,
file_name_suffix='.avro',
use_fastavro=True))
def run(argv=None):
"""
If you want to run just this file for rapid development, add the arg '-r DirectRunner' and add
GCS paths for --input and --avro_output, e.g.
python qalert_requests_dataflow.py --input gs://pghpa_test_qalert/requests/2020/06/2020-06-17_requests.json
--avro_output gs://pghpa_test_qalert/requests/avro_output/2020/06/2020-06-17/ -r DirectRunner
"""
known_args, pipeline_options, avro_schema = generate_args(
job_name='qalert-requests-dataflow',
bucket='{}_qalert'.format(os.environ['GCS_PREFIX']),
argv=argv,
schema_name='City_of_Pittsburgh_QAlert_Requests')
with beam.Pipeline(options=pipeline_options) as p:
date_conversions = [('lastActionUnix', 'lastAction'),
('addDateUnix', 'createDate')]
field_name_swaps = [('addDateUnix', 'createDateUnix'),
('status', 'statusCode'), ('latitude', 'lat'),
('longitude', 'long'),
('master', 'masterRequestId'),
('typeId', 'requestTypeId'),
('typeName', 'requestType')]
lines = p | ReadFromText(known_args.input, coder=JsonCoder())
load = (lines
| beam.ParDo(GetDateStrings(date_conversions))
| beam.ParDo(SwapFieldNames(field_name_swaps))
| beam.ParDo(GetStatus())
| beam.ParDo(GetClosedDate())
| WriteToAvro(known_args.avro_output,
schema=avro_schema,
file_name_suffix='.avro',
use_fastavro=True))
