python-kafka实现produce与consumer

1.python-kafka：

api送上：https://kafka-python.readthedocs.io/en/latest/apidoc/KafkaConsumer.html

2.实现一个broker、topic可配置的生产者与消费者：

#coding=utf-8

import time
import logging
import sys
import json
import etc.config as conf
sys.path.append('***********/kafka-python-1.3.3')
from kafka import KafkaProducer
from kafka import KafkaConsumer
from kafka.errors import KafkaError
from kafka import TopicPartition


def log_name():
    base_name = conf.kafka_logDir
    date = time.strftime('%Y%m%d',time.localtime(time.time())) + '.log'
    return base_name + date

logging.basicConfig(level=logging.DEBUG,
        format='%(asctime)-15s %(filename)s[line:%(lineno)d] %(levelname)s %(message)s',
        datefmt='%Y-%m-%d %H:%M:%S',
        filename=log_name(),
        filemode='a'
        )
console = logging.StreamHandler()
console.setLevel(logging.INFO)
logging.getLogger('').addHandler(console)


class kfkProducer(object):

    # producer = None

    def __init__(self, broker, kafkaPort, kafkaTopic=''):
        self._broker = broker
        self._kafkaPort = kafkaPort
        self._kafkaTopic = kafkaTopic

    def __str__(self):
        logging.info("--------------------------------")
        logging.info("kafka-producer params ...")
        logging.info("[KAFKA-BROKER]:%s" %self._broker)
        logging.info("[KAFKA-PORT]:%s" %self._kafkaPort)
        logging.info("[KAFKA-TOPIC]:%s" %self._kafkaTopic)
        logging.info("--------------------------------")

    def registerKfkProducer(self):
        try:
            producer = KafkaProducer(bootstrap_servers = '{kafka_host}:{kafka_port}'.format(
                kafka_host=self._broker,
                kafka_port=self._kafkaPort
                ))
        except KafkaError as e:
            logging.info(e)
        return producer

    def produceMsg(self, topic, msg, partition=0):
        # 自动将输入字符串转化为json格式，产出消息
        if(topic in ('', None)):
            logging.error("topic is None, plz check!")
        else:
            try:
                # parmas_message = json.dumps(msg)#转化为json格式
                producer = self.registerKfkProducer()
                producer.send(topic, value=msg, partition=partition)
                producer.flush()
                # time.sleep(1)
            except KafkaError as e:
                logging.info(e)



class kfkConsumer(object):

    # consumer = None

    def __init__(self, broker, kafkaPort, kafkaTopic=''):
        self._broker = broker
        self._kafkaPort = kafkaPort
        self._kafkaTopic = kafkaTopic

    def __str__(self):
        logging.info("--------------------------------")
        logging.info("kafka-consumer params ...")
        logging.info("[KAFKA-BROKER]:%s" %self._broker)
        logging.info("[KAFKA-PORT]:%s" %self._kafkaPort)
        logging.info("[KAFKA-TOPIC]:%s" %self._kafkaTopic)
        logging.info("--------------------------------")

    def registerConsumer(self):
        try:
            consumer = KafkaConsumer(
                bootstrap_servers=[self._broker+':'+self._kafkaPort],
                auto_offset_reset='earliest')
        except KafkaError as e:
            logging.info(e)
        return consumer

    def consumerMsg(self, topic, partition=0):
        if(topic in ('', None)):
            logging.error("topic is None, plz check!")
        else:
            try:
                v_consumer = self.registerConsumer()
                v_consumer.assign([TopicPartition(topic,partition)])
                # self.registerConsumer.subscribe([self._kafkaTopic])
                for message in v_consumer:
                    # message value and key are raw bytes -- decode if necessary!
                    # e.g., for unicode: `message.value.decode('utf-8')
                    logging.info("%s:%d:%d: msg=%s" % (message.topic, message.partition,
                                                            message.offset, message.value.decode('utf-8')))
            except KafkaError as e:
                logging.info(e)

3.实现命令行输入topic和partition，即可生产消息：

#coding=utf-8

import os 
import sys
import json
import etc.config as conf
from PykafkaMgr import kfkProducer

#从json文件获取消息
def getMsgFromJsonfile(filePath):
    if(not os.path.isfile(filePath)):
        print(u"[%s] 输入的json文件路径有误,请检查..." %filePath)
    else:
        with open(filePath) as json_file:
            return json.load(json_file)

def except4v():
    if(len(sys.argv) <= 1):
        print(u"未输入topic和partition！
你可以--help查看具体使用方法...")
    elif(sys.argv[1].startswith("--")):
        option = sys.argv[1][2:]
        # print(option)
        if(option in ("version", "Version")):
            print("Version 1.0 
Python 2.7.3 (default, Nov  6 2015, 14:11:14) 
                    
[GCC 4.4.7 20120313 (Red Hat 4.4.7-4)] on linux2")
        elif(option == "help"):
            print(u"produceMsg.py 接收两个参数, 第一个是topic, 第二个是partition 
eg:python produceMsg.py test 0 
向topic名为test第0分区生产消息")

def calcMsg(jsonMsg):
    sumMsg, sumAcct = 0, 0
    msgNum = len(jsonMsg)
    print("------------------------------------------")
    for i in range(msgNum):
        acct_num = len(jsonMsg[i]["MSGBODY"])
        print(u"第[%d]条消息，包含ACCT_ID账户数:[%d]个"%(i+1, acct_num))
        sumMsg = i+1
        sumAcct += acct_num
        acct_num = 0
    print(u"本次生产消息总共[%d]条, 总共账户数：[%d]个"%(sumMsg, sumAcct))
    print("------------------------------------------")

if __name__ == '__main__':

    except4v()

    if(len(sys.argv) == 3):
        topic = sys.argv[1]
        partition = int(sys.argv[2])
        produce = kfkProducer(conf.kafka_mgr["broker"], conf.kafka_mgr["port"], topic)
        produce.__str__()
        jsonMsg = getMsgFromJsonfile(conf.kafka_produce)
        for i in range(len(jsonMsg)):
            produce.produceMsg(topic, ('%s'%jsonMsg[i]).encode('utf-8'), partition)
        calcMsg(jsonMsg)

4.设置两个配置文件：

第一个是config.py

#coding=utf-8

#broker配置还有一种方式是：kafka_mgr={"broker":'ip1:port,ip2:port,...,ipn:port'},就是改为kafka集群，不过代码要稍微作调整（参数列表改下就行了）。当然配置两种，通过一个开关去控制也可以。自选
kafka_mgr = {
    "broker" : '10.***.***.***',
    "port" : 6667,
}

kafka_logDir = r"/*******/log/****"

#生产者输入json文件
kafka_produce = r"/**********/data/input/produceMsg.json"

生产者输入json文件：
produceMsg.json

json文件附上说明，具体可以按照说明配置

hi, welcome here~

produceMsg.json
=================================
输入json格式数据，作为生产者消息的输入。
1.支持多条json数据输入。格式如下：
    [
    json1,
    json2,
    ...,
    jsonN
    ]
总体结构是：[  ,  ]

2.此json文件不能加注释，因为会破坏json文件格式，导致无法解析
3.输入只要是json格式，不需要关注是不是一行或多行，多换行、空格等都不影响解析

消费者也是利用以上两个配置文件去实现即可。此处代码略