-
Notifications
You must be signed in to change notification settings - Fork 0
/
data_stream.py
86 lines (66 loc) · 2.65 KB
/
data_stream.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
import logging
import json
from pyspark.sql import SparkSession
from pyspark.sql.types import *
import pyspark.sql.functions as psf
schema = StructType([
StructField("crime_id", StringType(), True),
StructField("original_crime_type_name", StringType(), True),
StructField("report_date", StringType(), True),
StructField("call_date", StringType(), True),
StructField("offense_date", StringType(), True),
StructField("call_date_time", StringType(), True),
StructField("call_time", StringType(), True),
StructField("disposition", StringType(), True),
StructField("address", StringType(), True),
StructField("address_type", StringType(), True),
StructField("agency_id", StringType(), True),
StructField("city", StringType(), True),
StructField("state", StringType(), True),
StructField("common_location", StringType(), True),
])
def run_spark_job(spark):
df = spark \
.readStream \
.format("kafka") \
.option("kafka.bootstrap.servers", "localhost:9092") \
.option("subscribe", "org.sf.police.service-calls") \
.option("startingOffsets", "earliest") \
.option("maxOffsetsPerTrigger", 200) \
.option("stopGracefullyOnShutdown", "true") \
.load()
df.printSchema()
kafka_df = df.selectExpr("CAST(value AS STRING)")
service_table = kafka_df\
.select(psf.from_json(psf.col('value'), schema).alias("DF"))\
.select("DF.*")
distinct_table = service_table \
.select(psf.col('crime_id'),
psf.col('original_crime_type_name'),
psf.col('disposition')).distinct()
agg_df = distinct_table \
.groupBy(distinct_table.original_crime_type_name).count()
query = agg_df \
.writeStream \
.trigger(processingTime="10 seconds") \
.outputMode('complete') \
.format('console') \
.start()
query.awaitTermination()
radio_code_json_filepath = "./radio_code.json"
radio_code_df = spark.read.json(radio_code_json_filepath)
radio_code_df = radio_code_df.withColumnRenamed("disposition_code", "disposition")
join_query = agg_df.join(radio_code_df, agg_df.disposition==radio_code_df.disposition, "left_outer")
join_query.awaitTermination()
if __name__ == "__main__":
logger = logging.getLogger(__name__)
spark = SparkSession \
.builder \
.master("local[*]") \
.config("spark.ui.port", 3000) \
.appName("KafkaSparkStructuredStreaming") \
.getOrCreate()
spark.conf.set("spark.default.parallelism", "2")
spark.conf.set("spark.sql.shuffle.partitions", "1")
run_spark_job(spark)
spark.stop()