def dealItem(self, title, pageTitle, memberInfo, descInfo, result):
# 翻译描述信息
if descInfo:
descInfoCN = mainTranslate(descInfo)
else:
descInfoCN = ""
item = {
"_id": result["part"] + "_" + result["url"],
"url": result["url"],
"keyWord": result["keyWord"],
"language": result["language"],
"name": result["name"],
"part": result["part"],
"station": result["station"],
"title": title,
"pageTitle": pageTitle,
"memberInfo": memberInfo,
"descInfo": descInfo,
"descInfoCN": descInfoCN,
"country": ""
}
self.insertItem(item)
def parsePageUser(self, response, url, part, name):
try:
responseBody = json.loads(response)
try:
responseText = responseBody[1]
except Exception as e:
return None
# 获取url
try:
urlList = jsonpath.jsonpath(responseBody, "$..ownerUrls")[0]
except Exception as e:
logging.error(e)
urlList = [url]
userurl = ""
for i in urlList:
if "www.youtube.com" not in i:
continue
userurl = "https://" + i.split("://")[-1]
if not userurl:
return None
# 判断是否在数据库中
result = collection.find_one({
"part": part,
"url": userurl,
"platId": platId
})
if result:
logging.warn("存在库中part:{},name:{},url:{}".format(
part, name, userurl))
return None
result = formeryoutubecollection.find_one({
"part": part,
"url": url,
"platId": platId
})
if result:
logging.warn("存在库中part:{},name:{},url:{}".format(
part, name, url))
return None
if part == "clothes":
# 判断是否在cmms中
domain = "http://cmms.gloapi.com/"
isExists = checkUrl(userurl, domain)
if isExists:
# 代表存在接口中
return None
elif part == "GB":
# 判断是否在mms中
domain = "http://mms.gloapi.com/"
isExists = checkUrl(userurl, domain)
if isExists:
# 代表存在接口中
return None
# 订阅者数量
subscriberCount = self.dealSubscriberCount(responseText)
# 观看人数
viewCount = self.dealViewCont(responseText)
# 评论:内容
description, descriptionLong = self.dealDescription(responseText)
isBlack = False
blackWord = ""
if part == "GB":
blackListall = blackList
else:
blackListall = clothesblackList
# 翻译成中文
if not description.strip():
descriptionChinese = ""
else:
descriptionChinese = mainTranslate(description)
# descriptionChinese = youdao(description)
blackWord = ""
blackWordCount = 0
for word in blackListall:
if word in description or word in descriptionChinese:
blackWord += word + ""
blackWordCount += 1
blackWord = blackWord.strip()
# emailAddress = re.findall("(\w+@\S+.\w+)", descriptionLong)
pattern = re.compile(
r'\b[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,4}\b')
try:
emailAddress = re.search(pattern, descriptionLong).group()
except Exception as e:
emailAddress = ""
# 国家
country = self.dealCountry(responseText)
# 标题title
upTitle = self.dealTitle(responseText)
# 链接
Facebook, Youtube, Instagram = self.dealLinks(responseText)
# 商务邮箱
businessEmail = self.dealMail(responseText)
# 相关频道
relateChannel = self.relateChannel(responseText)
UserItem = {
"subscriberCount": subscriberCount,
"description": descriptionChinese,
"descriptionUn": descriptionLong,
"country": country,
"viewCount": viewCount,
"upTitle": upTitle,
"Facebook": Facebook,
"Youtube": Youtube,
"Instagram": Instagram,
"emailAddress": emailAddress,
"isMail": businessEmail,
"relateChannel": relateChannel.strip(),
"url": userurl, # blackWord = "" blackWordCount = 0
"blackWord": blackWord,
"blackWordCount": blackWordCount
}
except Exception as e:
logging.error(traceback.format_exc())
UserItem = {}
return UserItem
def parsePageVideo(self, response, videoUrl, part, station, userUrl):
try:
response = json.loads(response)
titleList = jsonpath.jsonpath(
response, "$..gridRenderer.items..title..simpleText")
# 取8个
titleList = titleList[:self.videoNum]
lastUpdateTimeList = jsonpath.jsonpath(
response,
"$..gridRenderer.items..publishedTimeText.simpleText")
lastUpdateTimeList = lastUpdateTimeList[:self.videoNum]
viewCountTextList = jsonpath.jsonpath(
response, "$..gridRenderer.items..viewCountText.simpleText")
viewCountTextList = viewCountTextList[:self.videoNum]
totalViewCount = 0
viewCountList = []
for viewCountText in viewCountTextList:
try:
viewCount = int(
viewCountText.replace("次观看",
"").replace("人正在观看", "").replace(
",", "").strip())
viewCountList.append(viewCount)
totalViewCount += viewCount
except Exception as e:
continue
videoTittle = ""
for title, lastUpdateTime, viewCount in zip(
titleList, lastUpdateTimeList, viewCountList):
videoTittle += title + "\n"
# videoTittle = youdao(videoTittle)
if not videoTittle.strip():
videoTittleChinese = ""
else:
videoTittleChinese = mainTranslate(videoTittle)
VideoTitleCount = 0
whiteWord = ""
if part == "GB":
whiteListall = whiteList
else:
whiteListall = clotheswhiteList
if station == "Zaful":
whiteListall = zafulWhiltList
for word in whiteListall:
if word.lower() in videoTittle.lower() or word.lower(
) in videoTittleChinese.lower():
VideoTitleCount += 1
word_new = word + " "
whiteWord += word_new
logging.error("part:{},匹配度等于{}分,videoUrl:{},匹配单词:{}".format(
part, VideoTitleCount, userUrl, whiteWord.strip()))
try:
titleFirst = videoTittleChinese.split("\n")[0]
except Exception as e:
titleFirst = ""
try:
viewCountFirst = int(viewCountTextList[0].replace(
"次观看", "").replace("人正在观看", "").replace(",", "").strip())
except Exception as e:
viewCountFirst = 0
item = {
"videoTittle": videoTittleChinese,
"videotitleUn": videoTittle,
"viewCountAvg": int(totalViewCount / len(titleList)),
"titleLastUpdateTime": lastUpdateTimeList[0],
"whiteWord": whiteWord.strip(),
"VideoTitleCount": VideoTitleCount,
"titleFirst": titleFirst,
"viewCountFirst": viewCountFirst
}
except Exception as e:
item = {}
return item
def dealHeaderFooterInfo(selector):
textList = selector.xpath("//header//text()")
if not textList:
node = selector.xpath(
"//*[contains(@class,'header')]") + selector.xpath(
"//*[contains(@id,'header')]") + selector.xpath(
"//*[contains(@class,'Header')]") + selector.xpath(
"//*[contains(@id,'Header')]")
if node:
textList = node[0].xpath(".//text()")
else:
node = selector.xpath("//head")
if node:
textList = node[0].xpath(".//text()")
else:
textList = []
if not textList:
headerStr = ""
else:
textStr = ""
for text in list(set(textList)):
text = text.replace("\n", "").strip()
if not text:
continue
text = text + ","
textStr += text
headerStr = textStr.strip()[:-1]
textList = selector.xpath("//footer//text()")
if not textList:
node = selector.xpath(
"//*[contains(@class,'footer')]") + selector.xpath(
"//*[contains(@id,'footer')]") + selector.xpath(
"//*[contains(@class,'Footer')]") + selector.xpath(
"//*[contains(@id,'Footer')]")
if node:
textList = node[0].xpath(".//text()")
else:
textList = []
if not textList:
footerStr = ""
else:
textStr = ""
for text in list(set(textList)):
text = text.replace("\n", "").strip()
if not text:
continue
text = text + ","
textStr += text
footerStr = textStr.strip()[:-1]
if not footerStr and not headerStr:
headerZH = ""
footerZH = ""
else: # 两者至少有一个
if not footerStr:
headerZH = mainTranslate(headerStr[:4000])
footerZH = ""
elif not headerStr:
footerZH = mainTranslate(footerStr[:4000])
headerZH = ""
else:
# 合并翻译
headerZH = mainTranslate(headerStr[:4000])
footerZH = mainTranslate(footerStr[:4000])
fhBlackWord = ""
fhBlackWordCount = 0
for word in blackWordList:
if word in footerStr or word in headerZH or word in footerZH or word in headerStr:
fhBlackWord += word + " "
fhBlackWordCount += 1
fhBlackWord = fhBlackWord.strip()
return headerStr, footerStr, headerZH, footerZH, fhBlackWord, fhBlackWordCount
def dealResponse(responseBody, mongoUrl):
whiteNum = 0
blackNum = 0
blackStr = ""
whiteStr = ""
headerStr, footerStr = "", ""
headerZH, footerZH = "", ""
fhBlackWord, fhBlackWordCount = "", 0
facebook, instagram, youtube, twitter, title, desc, titleChinese, emailStr = "", "", "", "", "", "", "", ""
try:
selector = etree.HTML(responseBody)
except Exception as e:
logging.error(e)
selector = etree.HTML(responseBody.decode())
# 获取邮箱信息
emailStr = getMailPage(responseBody, selector)
# 获取标题
try:
title = selector.xpath('//title/text()')[0].replace('\n', '').replace(
' ', ' ').strip()
except:
logging.error("url:{}".format(mongoUrl))
title = ""
# 获取描述信息
try:
desc = selector.xpath(
'//meta[@name="description"]/@content')[0].replace('\n',
'').replace(
' ', ' ')
except:
try:
desc = selector.xpath(
'//meta[@name="Description"]/@content')[0].replace(
'\n', '').replace(' ', ' ')
except Exception as e:
desc = ""
# 标题和描述信息拼接
titkeDesc = title + "\n" + desc
try:
if titkeDesc:
titleChinese = mainTranslate(titkeDesc)
for bd in black:
if bd in titleChinese:
blackword = bd
blackNum += 1
blackStr += blackword + " "
logging.error("存在黑名单,word:{},url:{}".format(
blackword, mongoUrl))
# return blackNum, whiteNum, whiteStr, title, desc, titleChinese, emailStr, facebook, instagram, youtube, twitter
blackStr = blackStr.strip()
for td in white:
if td in titleChinese:
whiteNum += 1
whiteStr += td + " "
whiteStr = whiteStr.strip()
else:
titleChinese = ""
except:
titleChinese = ""
for lid in li:
urlds = selector.xpath(lid[0])
if len(urlds) > 0:
for keys in urlds:
if ('facebook' in str(keys)) and (facebook == ''):
facebook = lid[1] + str(keys)
elif ('instagram' in str(keys)) and (instagram == ''):
instagram = lid[1] + str(keys)
elif ('twitter' in str(keys)) and (twitter == ''):
twitter = lid[1] + str(keys)
elif ('youtube' in str(keys)) and (youtube == ''):
youtube = lid[1] + str(keys)
headerStr, footerStr, headerZH, footerZH, fhBlackWord, fhBlackWordCount = dealHeaderFooterInfo(
selector)
return fhBlackWord, fhBlackWordCount, blackStr, headerZH, footerZH, headerStr, footerStr, blackNum, whiteNum, whiteStr, title.strip(
), desc.strip(), titleChinese.strip(), emailStr, facebook.strip(
), instagram.strip(), youtube.strip(), twitter.strip(), blackStr.strip()
def dealHeaderFooterInfo(selector):
headerStr, footerStr, headerZH, footerZH, fhBlackWord, fhBlackWordCount = "", "", "", "", "", 0
try:
textList = selector.xpath("//header//text()")
node = selector.xpath(
"//*[contains(@class,'header')]") + selector.xpath(
"//*[contains(@id,'header')]") + selector.xpath(
"//*[contains(@class,'Header')]") + selector.xpath(
"//*[contains(@id,'Header')]")
if node:
for i in node:
try:
textList += i.xpath(".//text()")
except Exception as e:
textList = []
textList += selector.xpath("//head//text()")
if not textList:
try:
textList = selector.xpath("//html//text()")
except Exception as e:
textList = []
if not textList:
headerStr = ""
else:
textStr = ""
for text in list(set(textList)):
text = text.replace("\n", "").strip()
if not text:
continue
text = text + ","
textStr += text
headerStr = textStr.strip()[:-1]
try:
textList = selector.xpath("//footer//text()")
except Exception as e:
textList = []
node = selector.xpath(
"//*[contains(@class,'footer')]") + selector.xpath(
"//*[contains(@id,'footer')]") + selector.xpath(
"//*[contains(@class,'Footer')]") + selector.xpath(
"//*[contains(@id,'Footer')]")
if node:
for i in node:
try:
textList += i.xpath(".//text()")
except Exception as e:
textList = []
if not textList:
footerStr = headerStr
else:
textStr = ""
for text in list(set(textList)):
text = text.replace("\n", "").strip()
if not text:
continue
text = text + ","
textStr += text
footerStr = textStr.strip()[:-1]
if not footerStr and not headerStr:
headerZH = ""
footerZH = ""
else: # 两者至少有一个
if not footerStr:
headerZH = mainTranslate(headerStr[:4000])
footerZH = ""
elif not headerStr:
footerZH = mainTranslate(footerStr[-4000:])
headerZH = ""
else:
# 合并翻译
headerZH = mainTranslate(headerStr[:4000])
footerZH = mainTranslate(footerStr[-4000:])
fhBlackWord = ""
fhBlackWordCount = 0
for word in blackWordList:
if word in footerStr or word in headerZH or word in footerZH or word in headerStr:
fhBlackWord += word + " "
fhBlackWordCount += 1
fhBlackWord = fhBlackWord.strip()
except Exception as e:
logging.error(traceback.format_exc())
return headerStr, footerStr, headerZH, footerZH, fhBlackWord, fhBlackWordCount
