Python+SparkStreaming+kafka+写入本地文件案例（可执行）

从kafka中读取指定的topic，根据中间内容的不同，写入不同的文件中。

文件按照日期区分。

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time    : 2018/4/9 11:49
# @Author  : baoshan
# @Site    :
# @File    : readTraceFromKafkaStreamingToJson.py
# @Software: PyCharm Community Edition

from pyspark import SparkContext
from pyspark.streaming import StreamingContext
from pyspark.streaming.kafka import KafkaUtils
import datetime
import json
import time
from collections import defaultdict

import subprocess

class KafkaMessageParse:

    def extractFromKafka(self, kafkainfo):
        if type(kafkainfo) is tuple and len(kafkainfo) == 2:
            return kafkainfo[1]

    def lineFromLines(self, lines):
        if lines is not None and len(lines) > 0:
            return lines.strip().split("\n")

    def messageFromLine(self, line):
        if line is not None and "message" in line.keys():
            return line.get("message")

    def extractFromMessage(self, message):
        try:
            jline = json.loads(message)
            result = defaultdict()
            name = jline.get("name")

            if "speech" in name:
                trace_id = jline.get("trace_id")
                parent_id = jline.get("parent_id")
                span_id = jline.get("span_id")
                sa = jline.get("sa")
                sr = jline.get("sr")
                ss = jline.get("ss")
                ret = jline.get("ret")

                result['trace_id'] = trace_id
                result['parent_id'] = parent_id
                result['span_id'] = span_id
                result['name'] = name
                result['sa'] = sa
                result['sr'] = sr
                result['ss'] = ss
                result['ret'] = ret

                annotation = jline.get("annotation")
                try:
                    for anno in annotation:
                        if anno.get("name") == "nlp":
                            debug_log = anno.get("debug_log")
                            debug_log_anno = debug_log[0]
                            asr = debug_log_anno.get("asr")  # asr
                            nlp = debug_log_anno.get("nlp")

                            action = debug_log_anno.get("action")
                            jaction = json.loads(action)
                            response = jaction.get("response")
                            tts = response.get("action").get("directives")[0].get("item").get("tts")
                            result['tts'] = tts

                            jnlp = json.loads(nlp)
                            intent = jnlp.get('intent')
                            app_id = jnlp.get('appId')
                            cloud = jnlp.get("cloud")
                            slots = jnlp.get("slots")

                            result['app_id'] = app_id
                            result['intent'] = intent
                            result['cloud'] = cloud
                            result['asr'] = asr
                            result['nlp'] = nlp
                            result['slots'] = slots
                    debug_log = jline.get("debug_log")
                    debug_log0 = debug_log[0]
                    session_id = debug_log0.get("session_id")
                    codec = debug_log0.get("codec")
                    if not session_id:
                        session_id = ""  # 超级无敌重要
                    wavfile = session_id + ".wav"
                    codecfile = session_id + "." + codec

                    asrtimestr = session_id.split("-")[-1]
                    try:
                        st = time.localtime(float(asrtimestr))
                    except:
                        st = time.localtime()
                    asrtime = time.strftime("%Y-%m-%d %H:%M:%S", st)
                    asrthedate = time.strftime("%Y%m%d", st)

                    asrdeviceid = debug_log0.get("device_id")
                    asrdevicetype = debug_log0.get("device_type")
                    asrdevicekey = debug_log0.get("device_key")

                    result['session_id'] = session_id
                    result['device_id'] = asrdeviceid
                    result['device_key'] = asrdevicekey
                    result['device_type'] = asrdevicetype
                    result['thedate'] = asrtime
                    result['wavfile'] = wavfile
                    result['codecfile'] = codecfile
                    result['asrthedate'] = asrthedate

                    strmessage = json.dumps(result, ensure_ascii=False)

                    return strmessage
                except:
                    return strmessage

            elif "tts" in name: # tts
                try:
                    trace_id = jline.get("trace_id")
                    parent_id = jline.get("parent_id")
                    span_id = jline.get("span_id")
                    name = jline.get("name")
                    sa = jline.get("sa")
                    sr = jline.get("sr")
                    ss = jline.get("ss")
                    ret = jline.get("ret")

                    result['trace_id'] = trace_id
                    result['parent_id'] = parent_id
                    result['span_id'] = span_id
                    result['name'] = name
                    result['sa'] = sa
                    result['sr'] = sr
                    result['ss'] = ss
                    result['ret'] = ret

                    debug_log = jline.get("debug_log")
                    debug_log_tts = debug_log[0]
                    text = debug_log_tts.get("text")
                    codec = debug_log_tts.get("codec")
                    declaimer = debug_log_tts.get("declaimer")
                    logs = debug_log_tts.get("logs")
                    params = debug_log_tts.get("params")

                    result['text'] = text
                    result['codec'] = codec
                    result['declaimer'] = declaimer
                    result['logs'] = logs
                    result['params'] = params

                    strresult = json.dumps(result, ensure_ascii=False)

                    return strresult
                except:
                    return None
        except:
            return ''

def tpprint(val, num=10000):
    """
    Print the first num elements of each RDD generated in this DStream.
    @param num: the number of elements from the first will be printed.
    """
    def takeAndPrint(time, rdd):
        taken = rdd.take(num + 1)
        print("########################")
        print("Time: %s" % time)
        print("########################")
        DATEFORMAT = '%Y%m%d'
        today = datetime.datetime.now().strftime(DATEFORMAT)

        speechfile = open("/mnt/data/trace/trace.rt.speech." + today, "a")
        ttsfile = open("/mnt/data/trace/trace.rt.tts." + today, "a")
        otherfile = open("/mnt/data/trace/trace.rt.other." + today, "a")

        for record in taken[:num]:
            if record is not None and len(record) > 2: # None 不打印
                print(record)
                jrecord = json.loads(record)
                name = jrecord.get("name")
                if "speech" in name:
                    speechfile.write(str(record) + "\n")
                elif "tts" in name:
                    ttsfile.write(str(record) + "\n")
                else:
                    otherfile.write(str(record) + "\n")

        speechfile.close()
        ttsfile.close()
        otherfile.close()

        if len(taken) > num:
            print("...")

    val.foreachRDD(takeAndPrint)

if __name__ == '__main__':
    zkQuorum = 'datacollect-1:2181,datacollect-2:2181,datacollect-3:2181'
    topic = {'trace-open-gw-5': 1, 'trace-open-gw-6': 1, 'trace-open-gw-7': 1, 'trace-open-gw-8': 1, 'trace-open-gw-9': 1}
    groupid = "rokid-trace-rt"
    master = "local[*]"
    appName = "SparkStreamingRokidTrace"
    timecell = 5

    sc = SparkContext(master=master, appName=appName)
    ssc = StreamingContext(sc, timecell)

    kvs = KafkaUtils.createStream(ssc, zkQuorum, groupid, topic)
    kmp = KafkaMessageParse()
    lines = kvs.map(lambda x: kmp.extractFromKafka(x))
    lines1 = lines.flatMap(lambda x: kmp.lineFromLines(x))
    valuedict = lines1.map(lambda x: eval(x))
    message = valuedict.map(lambda x: kmp.messageFromLine(x))
    rdd2 = message.map(lambda x: kmp.extractFromMessage(x))  # result is a json str

    tpprint(rdd2)

    ssc.start()
    ssc.awaitTermination()

还请各位大仙不吝赐教！