def run(self):
'''
每隔1秒，循环读取tasks
交给Downloader
:return:
'''
#创建进程池
pool = Pool()
while True:
#获取一条待执行的Task,并置为doing状态
task = self.taskUtil.get_ready()
if task is not None and len(task) > 0 or True:
Log.i('-----------------------------')
#用进程池启动Downloader
pool.apply_async(self.run_downloader, args=(task, ))
#休眠n秒(从配置文件中读取)
items = ConfigUtil.getItems('scheduler')
interval_min = items['interval_min']
interval_max = items['interval_max']
seconds = random.randint(int(interval_min), int(interval_max))
Log.i('Start sleep ' + str(seconds) + ' seconds')
time.sleep(seconds)
pool.close()
pool.join()
log.i('All subprocesses done.')
def run(self):
'''
生产进程执行，每隔60*60*60*24秒，循环读取tasks
:return:
'''
Log.i('ProducerUrl.run() in {0}'.format(time.ctime()))
while True:
#监听数据
DictData = self.pipeDictData.recv()
if DictData is None:
continue
#源数据处理(实体类)
self.URL_inf.dict2class(DictData)
#检查Mongo
if self.mogodbControl is None:
self.mogodbControl = Mongodb_Operator(
DbdataCCGPDFZB["host"], DbdataCCGPDFZB["port"],
DbdataCCGPDFZB["db_name"],
DbdataCCGPDFZB["default_collection"])
#检查Kafka
if self.KafkaOperator is None:
self.KafkaOperator = localKafkaUrlinformation()
#查重
uuid = self.get_md5(self.URL_inf.Urlname, self.URL_inf.title)
item = {"uuid": uuid}
value = self.mogodbControl.findone(
item, self.__Sendcollection) # 查询到返回文档
# #TODO 插入数据库有问题
if value is not None:
continue
#获取首页内容
self.URL_inf = self.downLoadHtml()
if self.URL_inf is None:
continue
#异步保存数据
self.savedata(self.URL_inf)
# #休眠n秒(从配置文件中读取)
items = ConfigUtil.getItems('consumerScheduler')
interval_min = items['interval_min']
interval_max = items['interval_max']
seconds = random.randint(int(interval_min), int(interval_max))
Log.i('StartProducerUrl sleep ' + str(seconds) + ' seconds')
time.sleep(seconds)
def run(self):
'''
获取免费IP代理进程执行，循环读取tasks
:return:
'''
Log.i('proxyIpPool.run() in {0}'.format(time.ctime()))
while True:
#调用本地和远程的免费ip代理api并推进ip消息队列
proxyIpPool = getIpProxyPool()
#统一改成本地
proxyIpPoolFromeRemote = getIpProxyPool()
# proxyIpPoolFromeRemote = getIpProxyPoolFromeRemote()
if proxyIpPool is not None:
self.queueDictData.put(proxyIpPool)
if proxyIpPoolFromeRemote is not None:
self.queueDictData.put(proxyIpPoolFromeRemote)
# 休眠n秒(从配置文件中读取)
items = ConfigUtil.getItems('proxyIpScheduler')
interval_min = items['interval_min']
interval_max = items['interval_max']
seconds = random.randint(int(interval_min), int(interval_max))
Log.i('proxyIpPool sleep ' + str(seconds) + ' seconds')
time.sleep(seconds)
def run(self):
'''
生产进程执行，每隔60*60*60*24秒，循环读取tasks
:return:
'''
Log.i ('ProducerUrl.run() in {0}'.format(time.ctime()))
while True:
#生产URL
if USE_SOURCEURL_TYPE is True:
if USE_ASYNCTASK_TYPE is True:
urlInformationList = ConfigUtil.readSourceListByParams(self.begin, self.end)
else:
urlInformationList = ConfigUtil.readSourceList()
else:
urlInformationList = ConfigUtil.readTaskList()
if urlInformationList is None:
continue
for urlInfor in urlInformationList:
data = urlInfor.class2dict()
diststrjson = json.dumps(data)
Log.i(diststrjson)
KafkaOperator = kafkaUrlinformation()
KafkaOperator.producerUrl(diststrjson)
#日执行一次不用休眠了.使用crontab定时任务驱动
if self.crontab==1:
os._exit(0)
else:
# #休眠n秒(从配置文件中读取)
items=ConfigUtil.getItems('producerScheduler')
interval_min = items['interval_min']
interval_max = items['interval_max']
seconds=random.randint(int(interval_min),int(interval_max))
Log.i('StartProducerUrl sleep ' + str(seconds) + ' seconds')
time.sleep(seconds)
def simpleRun(self):
'''
生产进程执行，每隔60*60*60*24秒，循环读取tasks
:return:
'''
Log.i('ProducerUrl.run() in {0}'.format(time.ctime()))
while True:
#资源检查
# KafkaOperator = kafkaUrlinformation()
KafkaOperator = localKafkaUrlinformation()
# if self.mogodbControl is None:
# self.mogodbControl = Mongodb_Operator(Dbdata["host"], Dbdata["port"], Dbdata["db_name"],
# Dbdata["default_collection"])
#解析数据源
# if USE_SOURCEURL_TYPE is True:
# if USE_ASYNCTASK_TYPE is True:
# urlInformationList = ConfigUtil.readSourceListRealTime()
# else:
# urlInformationList = ConfigUtil.readSourceList()
# else:
# urlInformationList = ConfigUtil.readTaskList()
urlInformationList = ConfigUtil.readSourceListRealTime()
#爬取,解析子URL
if urlInformationList is None:
continue
for urlInfor in urlInformationList:
data = urlInfor.class2dict()
#获取首页内容
dowloadData = self.downLoadHtml(data)
if dowloadData is None:
continue
# 解析提取分页url
pageData = self.getPageNumFromHome(dowloadData)
if pageData is None:
continue
for pageIndex in pageData:
# 获取首页内容
dowloadPageData = self.downLoadHtml(pageIndex.class2dict())
if dowloadPageData is None:
continue
#提取子链接
# self.URL_inf.dict2class(pageIndex)
ccgpChildrenLink = self.getChildrenLink(dowloadPageData)
if ccgpChildrenLink is None:
continue
#KAFKA下发子链接
for link in ccgpChildrenLink:
# 检查Mongo
if self.mogodbControl is None:
self.mogodbControl = Mongodb_Operator(
DbdataCCGPDFZB["host"], DbdataCCGPDFZB["port"],
DbdataCCGPDFZB["db_name"],
DbdataCCGPDFZB["default_collection"])
# 查重,不重复发送到kafka节省资源
if link.title is None: #标题为空的不发送
continue
uuid = self.get_md5(link.Urlname, link.title)
item = {"uuid": uuid}
value = self.mogodbControl.findone(
item, self.__Sendcollection) # 查询到返回文档
# #TODO 插入数据库有问题
if value is not None: #数据库查重
continue
# 于浩说不要发父链接给他
if link.DeepNum >= 0:
producerData = json.dumps(link.class2dict())
Log.i("produce<<" + producerData)
KafkaOperator.producerUrl(producerData)
#日执行一次不用休眠了.使用crontab定时任务驱动
if self.crontab == 1:
os._exit(0)
else:
# #休眠n秒(从配置文件中读取)
items = ConfigUtil.getItems('producerScheduler')
interval_min = items['interval_min']
interval_max = items['interval_max']
seconds = random.randint(int(interval_min), int(interval_max))
Log.i('StartProducerUrl sleep ' + str(seconds) + ' seconds')
time.sleep(seconds)
