def checkAllLz(filePath, taskResultId):
taskResult = TaskResult.getOne(TaskResult.taskResultId == taskResultId)
delayDay = taskResult.taskId.intervalDay
#根据当前任务编号目录获取该目录下的所有xml文件
print '目录路径:',filePath
fnames=os.listdir(filePath)
for name in fnames:
xmlName = filePath+'/'+name
print 'xmlName:',xmlName
data = parseXMLFile(xmlName, 'CheckItem')
print 'data:',data
# 将需要检查的信息入库
for d in data:
cname = d.get('cname')
url = d.get('url')
if url != '':
if url[-1] == '/':
url = url.replace('http://', '')[0:-1].replace(' ','')
if cname!='':
cname = cname.replace(' ','')
webArea = d.get('area')
webtype = d.get('WebType')
# 检查更新company
if Company.getOne(Company.coname == cname) is None:
c = Company()
c.coname = cname
c.save(force_insert=True)
# 检查更新website
if Website.getOne(Website.domain == url) is not None:
q = Website.update(domain=url, type=webtype, area=webArea).where(Website.domain == url)
q.execute()
else:
com = Company.getOne(Company.coname == cname)
w = Website()
w.regID = com
w.domain = url
w.area = webArea
w.type = webtype
w.save(force_insert=True)
updateWeb = Website.getOne(Website.domain == url)
subTask = TaskInfo()
subTask.taskResultId = taskResult
subTask.webId = updateWeb
subTask.state = '1'
subTask.save(force_insert=True)
taskResultId = str(taskResultId)
if delayDay > 0:
# 需要周期执行的任务
executeMultiTaskInfo(taskResultId)
else:
#logger.debug("开始调用单次任务")
# 单次执行的任务
executeSingleTaskInfo(taskResultId)
def checkLz(subtaskId):
try:
# 根据任务结果编号获取本次需要检查的任务记录
subTask = TaskInfo.getOne(TaskInfo.id == subtaskId)
checkTaskResult = subTask.taskResultId
bigTaskId = checkTaskResult.taskId.taskId
if subTask is not None:
companyName = subTask.webId.regID.coname
url = subTask.webId.domain
if (validateUrl(url)):
# 从任务表中获取是否已经进行检查过了(根据网站有没有更新时间进行判断)
isExistsTask = TaskInfo.getOne(
TaskInfo.webId == Website.getOne((Website.domain == url) & (Website.updateDate.is_null(False))))
subId = subTask.id
# 数据库中无此task对应的网站记录
if isExistsTask is None:
# 抓取检测
fetchWebsite(companyName, bigTaskId,subId, url)
else:
isExistsWebId = isExistsTask.webId.webId
# 如果当前网站更新时间小于过期时间,说明不用重新进行抓取并检查
expired = Configs.getOne(Configs.type == 'update').expired
# 获取当前时间
currentTime = datetime.datetime.now()
isAlreadyExistsWeb(companyName, currentTime, expired, url, subId,
url,
isExistsWebId)
except Exception:
q = TaskInfo.update(state='-1')
q.execute()
def isAlreadyExistsWeb(companyName, currentTime, expired, shortUrl, subTaskId, url, isExistsWebId):
subTask = TaskInfo.getOne(TaskInfo.id == subTaskId)
taskResult = subTask.taskResultId
bigTask = taskResult.taskId
bigTaskId = bigTask.taskId
# 获取当前网站上一次的更新时间
websiteResult = Website.getOne(Website.webId == isExistsWebId)
webUpdateTime = websiteResult.updateDate
if webUpdateTime is None:
diffDay = expired + 1
else:
diffDay = (currentTime - webUpdateTime).days
if diffDay > expired:
print '时间已经过期'
# 网站信息过期,需要重新抓取并检测
fetchWebsite(companyName, bigTaskId, subTaskId, url)
else:
# 判断是否有当前网站信息是否存在亮照编号
if websiteResult.licID != '':
#logger.debug('已经亮照:', shortUrl)
q = TaskInfo.update(state='2').where(TaskInfo.id == subTaskId)
q.execute()
else:
#logger.debug('未亮照:', shortUrl)
q = TaskInfo.update(state='3').where(TaskInfo.id == subTaskId)
q.execute()
def noAccess(dirPath, filePath, mainTask, shortUrl, subTask):
dt = format(datetime.datetime.now(), '%Y-%m-%d %H:%M:%S')
qw = Website.update(updateDate=dt).where(Website.webId == subTask.webId.webId)
qw.execute()
q = TaskInfo.update(state='7').where(TaskInfo.id == subTask.id)
q.execute()
# 删除此首页文件
if os.path.exists(filePath):
os.remove(filePath)
# 删除本次任务目录
if os.path.exists(dirPath):
os.rmdir(dirPath)
def judgeLzResult(com, web, shortUrl, subTask):
# 更新网站信息和公司信息
impCompanyInfo(com)
# impWebsite(web)
# nWeb = Website.getOne(Website.domain == web.domain)
# judgeWeb = Website.getOne(Website.domain ** str.format("%{}%", shortUrl))
# 如果查询网址与抓取亮照后的网址不匹配
if shortUrl != '':
shortUrl = shortUrl.replace('http://', '').replace(' ', '')
print 'shortUrl:', shortUrl
print "web.domain:", web.domain
com = Company.getOne(Company.coname == com.coname)
if shortUrl != web.domain:
# 当已经存在跳转关系记录时,不再操作
existsJumpWeb = Website.getOne((Website.domain == web.domain))
if existsJumpWeb is None:
dt = format(datetime.datetime.now(), '%Y-%m-%d %H:%M:%S')
web.updateDate = dt
web.regID = com
web.save(force_insert=True)
# 更新网站跳转地址
q = Website.update(jump=web).where(Website.webId == subTask.webId.webId)
q.execute()
else:
impWebsite(web)
# 更新网站跳转地址
q = Website.update(jump=existsJumpWeb).where(Website.webId == subTask.webId.webId)
q.execute()
else:
#更新网站信息
impWebsite(web)
count = Website.select().where(
(Website.licID != '') & (Website.regID == com) & (Website.domain == shortUrl)
).count()
if count == 0:
onlyCname = Website.select().where(
(Website.licID != '') & (Website.regID == com)).count()
onlyDomain = Website.select().where(
(Website.licID != '') & (Website.domain == shortUrl)).count()
if onlyCname > 0:
q = TaskInfo.update(state='9').where(TaskInfo.id == subTask.id)
q.execute()
elif onlyDomain > 0:
q = TaskInfo.update(state='8').where(TaskInfo.id == subTask.id)
q.execute()
else:
q = TaskInfo.update(state='4').where(TaskInfo.id == subTask.id)
q.execute()
else:
q = TaskInfo.update(state='2').where(TaskInfo.id == subTask.id)
q.execute()
def judgeNoLz(companyName, shortUrl, subTask):
dt = format(datetime.datetime.now(), '%Y-%m-%d %H:%M:%S')
qw = Website.update(updateDate=dt).where(Website.webId == subTask.webId.webId)
qw.execute()
q = TaskInfo.update(state='3').where(TaskInfo.id == subTask.id)
q.execute()