def intervalDelayTask(taskResultId):
# 获取celery中当前已经正在进行的任务数
nowCount = TaskInfo.select().order_by(TaskInfo.id).where((TaskInfo.state == '6'))
if nowCount >= 0:
# 当celery中正在做的任务数量少于指定的数量时，向celery添加需要执行的任务
if nowCount <= config.celeryMaxCount:
taskCount = TaskInfo.select().where(
(TaskInfo.state == '1') & (TaskInfo.taskResultId == taskResultId)).count()
print 'taskCount:', taskCount
if taskCount == 0:
singalCheck(taskResultId)
# 查询该任务设置的延迟时间开启下一次需要检查的任务
interval = Task.getOne(Task.taskId == (
TaskResult.select(TaskResult.taskId).where(TaskResult.taskResultId == taskResultId))).intervalDay
print 'interval:', interval
if interval != "":
# 生成需要轮巡的新主任务结果记录
taskResult = TaskResult()
lastTaskResult = TaskResult.getOne(TaskResult.taskResultId == taskResultId)
taskResult.taskId = lastTaskResult.taskId
taskResult.state = '1'
taskResult.save()
#将上一次的任务结果编号所对应的webId指定给新的任务结果
psql_db.transaction()
try:
query = (TaskInfo
.insert_from(
fields=[TaskInfo.webId],
query=TaskInfo.select(TaskInfo.webId).where(TaskInfo.taskResultId == lastTaskResult)))
query.execute()
q = TaskInfo.update(taskResultId=taskResult).where(TaskInfo.taskResultId.is_null())
q.execute()
except Exception, e:
print e
psql_db.rollback()
# 获取当前时间
ctime = datetime.datetime.now()
delay_time = int(interval)
stime = ctime + datetime.timedelta(seconds=delay_time)
scheduler.add_job(intervalDelayTask, "date", next_run_time=stime, args=[taskResult.taskResultId],
jobstore="default", id=taskResult.taskResultId)
else:
tasks = TaskInfo.select().order_by(TaskInfo.id).paginate(0, config.sendCeleryCount).where(
(TaskInfo.taskResultId == taskResultId) & (TaskInfo.state == '1'))
for subTask in tasks:
subtaskId = subTask.id
fetchCycle.apply_async((subtaskId,), queue="celery")
# 更新taskinfo状态为已发送
q = TaskInfo.update(state='6').where(TaskInfo.id == subtaskId)
q.execute()
def fetchWebsite(companyName, bigTaskId, subTaskId, url):
mainTask = Task.getOne(Task.taskId == bigTaskId)
subTask = TaskInfo.getOne(TaskInfo.id == subTaskId)
# 每一个新的任务存放首页的目录格式为:网站域名+任务编号+任务结果编号+企业首页信息------------->如:www.yummy77.com/123456789/1/www.yummy77.com.html
dirPath = os.path.abspath(
str.format(os.path.join('./data/{}/{}/{}/'), url, bigTaskId, subTask.taskResultId.taskResultId))
if not os.path.exists(dirPath):
os.makedirs(dirPath)
filePath = dirPath + '/' + url + '.html'
# 如果url中没有http://，则对其进行添加
if url.find('http://') == -1:
url = "http://%s" % url
# 进行抓取操作,然后对其进行亮照结果筛选
status = downloadByPhantom('/usr/bin/phantomjs', os.path.abspath('./phantomjs/fetch.js'),
url, filePath, 'utf-8', str(config.request_timeout), str(config.timeout), '', '')
if not os.path.exists(filePath):
#logger.debug('主页无法访问:', url)
# 无法访问
noAccess(dirPath, filePath, mainTask, url, subTask)
else:
# 检查网站首页是否包含工商亮照标识
webContent = open(filePath, 'r').read()
# 返回的结果类似于<html><head></head><body></body></html>
noContent = re.match('.*<body></body>', webContent)
# 返回空白结果
blankContent = len(webContent)
# 返回纠错网址结果
errorSite = webContent.find('网址纠错')
# 网站主页无法访问
if noContent is not None or blankContent == 0 or errorSite != -1:
# 按照无法访问处理
noAccess(dirPath, filePath, mainTask, url, subTask)
else:
# 对网站首页进行解析并检查
makeWeb(companyName, filePath, mainTask, url, subTask)
