pyflink get_list_state

# -*- coding: utf-8 -*-
from pyflink.datastream import StreamExecutionEnvironment
from pyflink.datastream.functions import RuntimeContext, FlatMapFunction, MapFunction
import json
import re
import logging
import sys
from pyflink.datastream.state import ValueStateDescriptor, MapStateDescriptor,ListStateDescriptor
from pyflink.datastream.connectors.kafka import FlinkKafkaConsumer, TypeInformation
from pyflink.common.typeinfo import Types
from pyflink.datastream.connectors.elasticsearch import Elasticsearch7SinkBuilder, ElasticsearchEmitter, FlushBackoffType
from  pyflink.datastream.connectors import  DeliveryGuarantee
from pyflink.common.serialization import SimpleStringSchema
from datetime import datetime

logging.basicConfig(stream=sys.stdout, level=logging.INFO, format="%(asctime)s-%(levelname)s-%(message)s")
logger = logging.getLogger(__name__)

# 创建 StreamExecutionEnvironment 对象
env = StreamExecutionEnvironment.get_execution_environment()
env.set_parallelism(3)
#env.add_jars("file:///root/pyflink/flink-sql-connector-kafka_2.11-1.14.4.jar")

## kafka

TEST_KAFKA_SERVERS = "1.1.223.1:9092,1.1.223.2:9092,1.1.223.3:9092"
TEST_KAFKA_TOPIC = "new-clpf-gaps-topic"
TEST_GROUP_ID =  "clpf_gaps_group_test"

def get_kafka_customer_properties(kafka_servers: str, group_id: str):
    properties = {
        "bootstrap.servers": kafka_servers,
        "fetch.max.bytes": "67108864",
        "key.deserializer": "org.apache.kafka.common.serialization.StringDeserializer",
        "value.deserializer": "org.apache.kafka.common.serialization.StringDeserializer",
        "enable.auto.commit": "false",  # 关闭kafka 自动提交，此处不能传bool 类型会报错
        "group.id": group_id,
        "max.poll.records": "150"
    }
    return properties

properties = get_kafka_customer_properties(TEST_KAFKA_SERVERS, TEST_GROUP_ID)

class LogEvent:
    # id表示全局流水
    id = None
    # source ip
    source = None
    #进程名字
    fileTag= None
    #文件名字
    fileName = None
    #场景码
    serviceCode = None
    #系统名字
    appName= None
    #时间戳
    timestamp = None
    #偏移量
    offset = None

    def __init__(self, id,source, fileTag,fileName, serviceCode,appName,timestamp,offset,message,index_name):
        self.id=id
        self.source = source
        self.fileTag = fileTag
        self.fileName = fileName
        self.serviceCode = serviceCode
        self.appName = appName
        self.timestamp= timestamp
        self.offset = offset
        self.message = message
        self.index_name = index_name

    def to_dict(self):
        return {
            "id": str(self.id),
            "source": str(self.source),
            "fileTag": str(self.fileTag),
            "fileName":str(self.fileName),
            "serviceCode":str(self.serviceCode),
            "appName":str(self.appName),
            "timestamp":self.timestamp,
            "offset":str(self.offset),
            "message":self.message,
            "index_name": self.index_name
        }

    def get_source(self):
         return self.source

class MyMapFunction(FlatMapFunction):
    def open(self, runtime_context: RuntimeContext):
        self.process_id_to_bus_seq = runtime_context.get_map_state(MapStateDescriptor('process_id_map_bus_seq', Types.STRING(), Types.STRING()))
        self.gapslist=runtime_context.get_list_state(ListStateDescriptor('process_list', Types.LIST(Types.STRING())))

    def flat_map(self, raw_message):
        id = ''
        source =''
        fileTag =''
        fileName =''
        serviceCode =''
        appName =''
        timestamp =''
        process_id = ''
        offset =''
        message =''
        unique_key =''
        index_name = ''
        bus_seq = ''
        try:
           raw_message = raw_message.replace("\n", "")
           #print(raw_message)
           out=json.loads(raw_message)
           message = out['message']
           source = out['source']
           fileTag = out['file_tag']
           serviceCode='00000'
           appName=out['app_name']
           timestamp=str(out.get('time_nano'))
           offset=out.get('offset')
           fileName=out.get('file_name')
           # pattern = r".*?接收数据.*?\d{26}"
           # matchObj = re.match(pattern, message)
        except:
             #logger.info('1111111111111111111111111111111')
               return
        date_str = datetime.now().strftime("%Y%m%d")
        index_name = 'flink-log-clpf-gapstest-' + str(date_str)
        if '开始分级日志' in message:
            self.process_id_to_bus_seq.clear()
            self.gapslist.clear()
            # 记录加到缓存
            self.gapslist.add([str(message)])
            return
        self.has_start='0'
        for x in self.gapslist.get():
             if  '开始分级日志' in x[0]:
                 self.has_start='1'
                 break;
        #<BUSS_SEQ_NO>20601020230621010072249201</BUSS_SEQ_NO>
        #if  "</BUSS_SEQ_NO>" in message:
        if re.findall("\<BUSS_SEQ_NO\>\d+\<\/BUSS_SEQ_NO\>",message):
              pat = re.compile(r"\<BUSS_SEQ_NO\>(\d+)\<\/BUSS_SEQ_NO\>")
              bus_seq = pat.search(message).group(1)
              self.process_id_to_bus_seq.put('id', bus_seq)
              self.gapslist.add([str(message)])
              id=bus_seq
              for output_message in  self.gapslist.get():
                      log_event = LogEvent(id, source, fileTag, fileName, serviceCode, appName, timestamp, offset,output_message[0], index_name)
                      yield log_event.to_dict()
              self.gapslist.clear()
              self.has_start='0'
              #log_event = LogEvent(id, source, fileTag, fileName, serviceCode, appName, timestamp, offset,message, index_name)
              #yield log_event.to_dict()
              return
               
        if   self.has_start == '1':
             self.gapslist.add([str(message)])
             return
        id= self.process_id_to_bus_seq.get('id')
        if not id:
           id ='0'
        log_event = LogEvent(id, source, fileTag, fileName, serviceCode, appName, timestamp, offset, message, index_name)
        yield log_event.to_dict()

data_stream = env.add_source(
    FlinkKafkaConsumer(topics=TEST_KAFKA_TOPIC,
        properties=properties,
        deserialization_schema=SimpleStringSchema()) \
        .set_commit_offsets_on_checkpoints(True) \
        .set_start_from_latest()
).name(f"消费{TEST_KAFKA_TOPIC}主题数据")

#env.add_jars("file:///root/pyflink/flink-sql-connector-elasticsearch7-3.0.1-1.16.jar")

# .set_hosts(['1.1.101.32:9200','1.1.101.33:9200','1.1.101.38:9200']) \
es_sink = Elasticsearch7SinkBuilder() \
        .set_bulk_flush_backoff_strategy(FlushBackoffType.EXPONENTIAL, 5, 1000) \
        .set_emitter(ElasticsearchEmitter.dynamic_index('index_name')) \
        .set_hosts(['1.1.101.32:9200','1.1.101.33:9200','1.1.101.38:9200']) \
        .set_delivery_guarantee(DeliveryGuarantee.AT_LEAST_ONCE) \
        .set_bulk_flush_interval(1000) \
        .set_connection_request_timeout(30000) \
        .set_connection_timeout(31000) \
        .set_socket_timeout(32000) \
        .build()

def get_line_key(line):
    message = ''
    try:
        message = line.replace("\n", "")
        source = json.loads(message)['source']
    except:
        source = '999999'
    return source

data_stream.key_by(get_line_key).flat_map(MyMapFunction(),output_type=Types.MAP(Types.STRING(), Types.STRING())).sink_to(es_sink).set_parallelism(3)
#data_stream.key_by(get_line_key).flat_map(MyMapFunction(),output_type=Types.MAP(Types.STRING(), Types.STRING())).print()
#data_stream.key_by(get_line_key).flat_map(MyMapFunction()).print()

# 执行任务
env.execute('flink_clpf_gaps_test')