def parseWebsite():
log = open('../logs/parseweb.log', 'a')
# 每次取5000条
count = TempLz.select().count()
pagesize = 5000
pagecount = int(math.ceil(float(count) / float(pagesize)))
for i in range(pagecount):
datas = TempLz.select().where(TempLz.id>18879).order_by(TempLz.id).paginate(i + 1, pagesize)
if datas is not None:
for d in datas:
data = d.lzpage
if data is not None:
parseData = parserCompanyAndWeb(data)
com = buildCompany(parseData['company'])
web = buildWebsite(parseData['web'])
if com is not None and web is not None:
c = Company.getOne(Company.coname == com.coname)
if c is not None:
web.regID = c
impWebsite(web)
else:
impCompanyInfo(com)
tempCom = Company.getOne(Company.regID == com.regID)
web.regID = tempCom
impWebsite(web)
log.write(str(d.id)+ "\n")
print d.id
log.flush()
log.close()
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 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)