flink算子合并

最新推荐文章于 2024-09-11 16:51:16 发布

scan724

最新推荐文章于 2024-09-11 16:51:16 发布

阅读量815

点赞数 14

分类专栏： Flink实时计算文章标签： python 前端开发语言

本文链接：https://blog.csdn.net/zhaoyangjian724/article/details/138681550

版权

Flink实时计算专栏收录该内容

102 篇文章 1 订阅

订阅专栏

import json
import re
import logging
import sys
from collections import Counter

from pyflink.datastream import DataStream, StreamExecutionEnvironment
from pyflink.datastream.functions import RuntimeContext, FlatMapFunction, MapFunction
from pyflink.common.typeinfo import Types

s_env = StreamExecutionEnvironment.get_execution_environment()
data = DataStream(s_env._j_stream_execution_environment.socketTextStream('192.168.137.201', 8899))
s_env.set_parallelism(1)

# data.print()
def get_key():
return '999'

class LogEvent:
world = None

def __init__(self, world):
self.world = world
# self.count = count

def to_dict(self):
return {
"world": str(self.world)
# "count": str(self.count)

}

class MyMapFunction(FlatMapFunction):
def open(self, runtime_context: RuntimeContext):
pass

def flat_map(self, raw_message):
raw_message = raw_message.upper()
dict1 = {}
dict1['value'] = raw_message +'!!!'
yield dict1

data.flat_map(MyMapFunction(), output_type=Types.MAP(Types.STRING(), Types.STRING())).set_parallelism(4).print().set_parallelism(8)
s_env.execute('data')

//并行度一样算子合并