def getDownloadPageUrl(self, detailsResponse):
'''
get download page url by details page
:param detailsResponse:
:return:
'''
try:
torrentResultClass = TorrentResultClass()
if detailsResponse is not None and detailsResponse.status_code == 200:
detailsResponse.encoding = "gbk"
soup = bs(detailsResponse.text, "lxml")
title = soup.head.title.text
aTagList = soup.find_all("a")
for a in aTagList:
downloadPageUrl = a.string
if downloadPageUrl:
if self.isDownloadPageUrl(downloadPageUrl):
torrentResultClass.setCategory(self.getCategoryFromDatailsPageUrl(str(detailsResponse.url)))
torrentResultClass.setCrawData(self.formatDate())
torrentResultClass.setDetailsPageUrl(str(detailsResponse.url))
torrentResultClass.setResponse(self.urlToResponse([downloadPageUrl.strip()])[0])
torrentResultClass.setTitle(self.clearTitle(title))
else:
continue
# return self.getTorrentDownloadUrl(torrentResultClass)
return torrentResultClass
except BaseException:
logger.error(traceback.format_exc())
return None
def computeDetailsImageCount(self, detailsResponse):
'''
compute details image count bt detailsResponse
:param detailsResponse:
:return:
'''
try:
if detailsResponse is not None and detailsResponse.status_code == 200:
detailsResponse.encoding = "gbk"
soup = bs(detailsResponse.text, "lxml")
inputTagList = soup.find_all("input")
imageCount = 0
for inputTag in inputTagList:
try:
imageDownloadUrl = inputTag['data-src']
imageCount += 1
except KeyError:
continue
return imageCount
else:
logger.error("detalisResponse is None or status code not 200 in the computeDetailsImageCount function")
return 0
except BaseException:
logger.error(traceback.format_exc())
return 0
def getCategoryFromDatailsPageUrl(cls, detailsPageUrl):
try:
return Category.categoryNumberToNameDict[str(
detailsPageUrl.split("/")[4])]
except BaseException:
logger.error(traceback.format_exc())
return "unknown"
def getDetailsPageUrl(self, responseList):
'''
get details page url by list page
input url to queue
:return:
'''
if not responseList:
logger.error("get response list is empty in details page function")
detailsPageUrlList = []
# get details page url from list page response
try:
for response in responseList:
if response is not None and response.status_code == 200:
response.encoding = "gbk"
soup = bs(response.text, "lxml")
trList = soup.find_all("tr")
for tr in trList:
tdList = tr.find_all("td")
if len(tdList) == 5:
urlTagList = tdList[1].select('a[href]')
if urlTagList:
url = urlTagList[0]["href"]
# title = urlTagList[0].string
if url not in self.excludeUrlList and not self.urlInExclued(url):
detailsPageUrlList.append(self.domains + url.strip())
# url to response
logger.info("get details page response ...")
detailsPageResponseList = self.urlToResponse(detailsPageUrlList)
self.totalDownload = len(detailsPageResponseList)
logger.info("present craw count:{0}".format(str(len(detailsPageResponseList))))
# put to queue
for detailsPageResponse in detailsPageResponseList:
self._taskQueue.put(detailsPageResponse)
except BaseException:
logger.error(traceback.format_exc())
def delete(self, tableName, position=None):
'''
delete row from table, if position is None, clean table
:param tableName:
:param position: position is a dictionary
:return:
'''
conn = self.pool.connection()
cursor = conn.cursor()
sql = ""
try:
if position is not None:
if len(position) == 1:
key = [key for key in position.keys()][0]
sql = "DELETE FROM " + str(tableName) + " WHERE " + str(
key) + " = " + "'" + position[key] + "'"
else:
keys = [key for key in position.keys()]
sql = "DELETE FROM " + str(tableName) + " WHERE "
for key in keys:
sql += str(
key) + " = " + "'" + position[key] + "'" + " and "
sql = sql[:len(sql) - 5]
else:
sql = "TRUNCATE TABLE {0}".format(tableName)
cursor.execute(sql)
conn.commit()
return True
except BaseException:
logger.error(traceback.format_exc())
logger.error("error sql ---> " + sql)
return False
finally:
cursor.close()
conn.close()
def insert(self, tableName, info):
'''
insert data to tabel
:param tableName:
:param info: a dictionary
:return:
'''
conn = self.pool.connection()
cursor = conn.cursor()
sql = ""
try:
columnList = [column for column in info.keys()]
valueList = [value for value in info.values()]
sql = "INSERT INTO {0} ({1}) VALUES ({2})".format(
tableName, ", ".join(columnList),
", ".join(["%s" for _ in range(len(valueList))]))
cursor.execute(sql, valueList)
conn.commit()
return True
except pymysql.err.IntegrityError:
return True
except BaseException:
logger.error(traceback.format_exc())
logger.error("error sql ---> " + sql)
return False
finally:
cursor.close()
conn.close()
def torrentResultClassPreprocessing(self, torrentResultClass):
'''
the preprocessing torrentResultClass is used to write to the database
:param torrentResultClass:
:return:
'''
if torrentResultClass is None:
logger.error("torrentResultClass is None in the torrentResultClassPreprocessign function")
return None
try:
torrentResultDict = {}
torrentResultDict.setdefault("id", None)
torrentResultDict.setdefault("category", torrentResultClass.getCategory())
torrentResultDict.setdefault("title", torrentResultClass.getTitle())
torrentResultDict.setdefault("detailsPageUrl", torrentResultClass.getDetailsPageUrl())
torrentResultDict.setdefault("downloadPageUrl", torrentResultClass.getDownloadPageUrl())
torrentResultDict.setdefault("torrentDownloadUrl", torrentResultClass.getTorrentDownloadUrl())
torrentResultDict.setdefault("savePath", torrentResultClass.getSavePath())
torrentResultDict.setdefault("crawData", torrentResultClass.getCrawData())
torrentResultDict.setdefault("md5", torrentResultClass.getMd5())
torrentResultDict.setdefault("magnet", torrentResultClass.getMagnet())
torrentResultDict.setdefault("downloaded", torrentResultClass.getDownloaded())
return torrentResultDict
except BaseException:
logger.error(traceback.format_exc())
return None
def getTorrentDownloadUrl(self, torrentResultClass):
'''
get torrent download url by download information page
:param torrentResultClass:
:return:
'''
if torrentResultClass is None:
logger.error("get download page url failed, because torrentResultClass is None")
return None
try:
response = torrentResultClass.getResponse()
# i don't knonw why i might get a string here
if isinstance(response, str):
return None
if response is not None and response.status_code == 200:
response.encoding = "utf-8"
downloadUrl = self.torrentDownloadUrl(response.text)
torrentResultClass.setDownloadPageUrl(str(response.url))
torrentResultClass.setTorrentDownloadUrl(downloadUrl)
return torrentResultClass
else:
return None
except BaseException:
logger.error(traceback.format_exc())
return None
def clearTitle(cls, title):
try:
title = title.split(" - ")[0]
title = title.replace("\xa0", "")
return title
except BaseException:
logger.error(traceback.format_exc())
return title
def filterTorrentName(cls, torrentName):
try:
torrentName = re.sub(r'[？\\*|“<>:/]', '', torrentName)
torrentName = re.sub(r'[\/\\\:\*\?\"\<\>\|]', '', torrentName)
return torrentName
except BaseException:
logger.error(traceback.format_exc())
return torrentName
def existInDatabase(cls, tableName, position):
result = cls._database.queryIsExist(tableName, position)
try:
if result:
return True
else:
return False
except BaseException:
logger.error(traceback.format_exc())
return False
def update(self, tableName, info):
'''
指定表名,和更新数据更新表
更新数据为一个字典,例如:
{
"key_values":{"key1":"value1","key2":"value2"...},
"postions":{"key1":"value1","key2":"value2"...}
}
key_values接受多个参数,但是注意该表里是否有该keys
postions接受多个参数,但是目前判断条件只用and,如果需要or请重写代码
实例:
info = {
"key_values":{
"name":"naonao",
"age":"23",
"date":"940208",
"sex":"man"
},
"postions":{
"sex":"man",
"date":"940208"
}
}
返回: True
'''
conn = self.pool.connection()
cursor = conn.cursor()
sql = ""
try:
infoKeyList = [key for key in info['key_values'].keys()]
infoValueList = [value for value in info['key_values'].values()]
positionKeyList = [key for key in info['postions'].keys()]
positionValueList = [value for value in info['postions'].values()]
infoKeyList = list(map(lambda key: key + " = %s", infoKeyList))
positionKeyList = list(
map(lambda key: key + " = %s", positionKeyList))
sql = "UPDATE {tableName} SET {infoKey} WHERE {positionKey}".format(
tableName=tableName,
infoKey=", ".join(infoKeyList),
positionKey=" AND ".join(positionKeyList))
cursor.execute(sql, infoValueList + positionValueList)
conn.commit()
return True
except BaseException:
logger.error(traceback.format_exc())
logger.error("error sql ---> " + sql)
return False
finally:
cursor.close()
conn.close()
def executeCustomSQL(self, sql):
conn = self.pool.connection()
cursor = conn.cursor()
try:
cursor.execute(sql)
conn.commit()
return True
except BaseException:
logger.error(traceback.format_exc())
logger.error("error sql ---> " + sql)
return False
finally:
cursor.close()
conn.close()
def downloadTorrentFile(self, torrentResultClass):
'''
download torrent file
:param torrentResultClass:
:return:
'''
if torrentResultClass is None:
logger.error("torrentResultClass is None in the downloadTorrentFile function")
return None
torrentName = self.filterTorrentName(self.clearTitle(torrentResultClass.getTitle()))
torrentDownloadUrl = torrentResultClass.getTorrentDownloadUrl()
detailsPageUrl = torrentResultClass.getDetailsPageUrl()
if not torrentName:
logger.error("get torrent name failed")
return None
if not torrentDownloadUrl:
logger.error("get torrent download url failed")
return None
if not detailsPageUrl:
logger.error("get details page url failed")
return None
try:
categoryName = self.getCategoryFromDatailsPageUrl(detailsPageUrl)
torrentResponse = requests.get(torrentDownloadUrl, headers=self.headers)
torrentMd5 = self.computeMD5ByFile(torrentResponse.content)
# check save path is vailed
torrentPath = self.checkDirExist(
os.path.join(SAVE_PATH,
"torrent",
categoryName,
self.year(),
self.month(),
self.day(),
torrentName + ".torrent"))
with open(torrentPath, "wb+") as file:
file.write(torrentResponse.content)
torrentResultClass.setSavePath(torrentPath)
torrentResultClass.setCrawData(self.formatDate())
torrentResultClass.setMd5(torrentMd5)
torrentResultClass.setMagnet(self.torrentToMagnet(torrentResponse.content))
torrentResultClass.setDownloaded(0)
return torrentResultClass
except BaseException:
logger.error(traceback.format_exc())
return None
def getImageDownloadUrlGenerator(self, detailsResponse):
'''
get download page url by details page
this is a generator function
generator imageResultClass
:param detailsResponse:
:return: imageResultClassList
'''
try:
# imageResultClassList = []
if detailsResponse is not None and detailsResponse.status_code == 200:
detailsResponse.encoding = "gbk"
soup = bs(detailsResponse.text, "lxml")
title = soup.head.title.text
inputTagList = soup.find_all("input")
for inputTag in inputTagList:
try:
imageDownloadUrl = inputTag['data-src']
except KeyError:
continue
if imageDownloadUrl:
# print("is ok")
if self.existInDatabase("image", {"imageDownloadUrl": imageDownloadUrl}):
continue
if self.isImageDownloadPageUrl(imageDownloadUrl):
# imageResultClass.setCategory(self.getCategoryFromDatailsPageUrl(str(detailsResponse.url)))
imageResultClass = ImageResultClass()
imageResultClass.setCrawDate(self.formatDate())
imageResultClass.setDetailsPageUrl(str(detailsResponse.url))
imageResultClass.setDetailsPageImageCount(str(self.computeDetailsImageCount(detailsResponse)))
imageResultClass.setImageDownloadUrl(imageDownloadUrl)
imageResultClass.setResponse(self.urlToResponse([imageDownloadUrl])[0])
imageResultClass.setTitle(self.clearTitle(title))
imageResultClass.setCrawDate(self.formatDate())
yield imageResultClass
# imageResultClassList.append(imageResultClass)
else:
continue
# return imageResultClassList
except BaseException:
logger.error(traceback.format_exc())
return None
def queryIsExist(self, tableName, position):
'''
check is exist
:param table_name:
:param key: position condition --> MD5
:param value: position value --> XXX....
:return:
'''
conn = self.pool.connection()
# conn.autocommit(1)
cursor = conn.cursor()
sql = ""
try:
if len(position) == 1:
key = [key for key in position.keys()][0]
sql = "SELECT * FROM " + str(tableName) + " WHERE " + str(
key) + " = " + "'" + position[key] + "'"
sql += " FOR UPDATE"
else:
keys = [key for key in position.keys()]
sql = "SELECT * FROM " + str(tableName) + " WHERE "
for key in keys:
sql += str(
key) + " = " + "'" + position[key] + "'" + " and "
sql = sql[:len(sql) - 5]
sql += " FOR UPDATE"
# sql = "SELECT * FROM {0} WHERE {1} = '{2}'".format(table_name, key, value)
cursor.execute(sql)
conn.commit()
resultList = cursor.fetchall()
if len(resultList) != 0:
return True
else:
return False
except BaseException:
logger.error(traceback.format_exc())
logger.error("error sql ---> " + sql)
return False
finally:
cursor.close()
conn.close()
def writeToDatabase(self, torrentResultDict):
'''
information write to database
:param infoDict:
:return:
'''
if torrentResultDict is None:
logger.error("torrentResultDict is None")
return None
try:
result = self._database.queryIsExist("torrent", {"md5": torrentResultDict["md5"]})
progressBar = self.computeProgressBar(self.completed, self.totalDownload)
if not result:
logger.info("Torrent completed: {progressBar: <10}".format(progressBar=progressBar) + \
"category: {category: <20}".format(category=torrentResultDict['category']) + \
"Title:{title}".format(title=torrentResultDict['title']),
level="ALL")
self._database.insert("torrent", torrentResultDict)
else:
logger.info("Torrent completed:{progressBar: <5} torrent already exist database.".format(
progressBar=progressBar))
except BaseException:
logger.error(traceback.format_exc())
logger.error("An error occurred in the function ---> wirteToDataBase")
return None
def saveImageGenerator(self, imageResultClassGenerator):
'''
save image by image result class list
:param imageResultClassList:
:return:
'''
if imageResultClassGenerator is None:
logger.error("imageResultClass generator is None")
return None
for imageResultClass in imageResultClassGenerator:
try:
# imageResultClass = imageResultClassGenerator.__next__()
if isinstance(imageResultClass, str):
continue
response = imageResultClass.getResponse()
if response is not None and response.status_code == 200:
imageByte = imageResultClass.getResponse().content
imageMd5 = self.computeMD5ByFile(imageByte)
imageResultClass.setMd5(imageMd5)
imageName = os.path.split(imageResultClass.getImageDownloadUrl())[1]
imageDirName = imageResultClass.getTitle()
# check save path is vailed
imageSavePath = self.checkDirExist(os.path.join(SAVE_PATH,
"image",
self.year(),
self.month(),
self.day(),
imageDirName,
imageName))
imageResultClass.setSavePath(imageSavePath)
with open(imageSavePath, "wb+") as fo:
fo.write(imageByte)
yield imageResultClass
# except StopIteration:
# break
except BaseException:
logger.error(traceback.format_exc())
return None
def imageResultClassPreprocessing(self, imageResultClass):
'''
the preprocessing infoClass is used to write to the database
:param infoClass:
:return:
'''
if imageResultClass is None:
logger.error("imageResultClass is None in the imageResultClassPreprocessign function")
return None
try:
imageResultDict = {}
imageResultDict.setdefault("id", None)
imageResultDict.setdefault("title", imageResultClass.getTitle())
imageResultDict.setdefault("detailsPageUrl", imageResultClass.getDetailsPageUrl())
imageResultDict.setdefault("detailsPageImageCount", imageResultClass.getDetailsPageImageCount())
imageResultDict.setdefault("imageDownloadUrl", imageResultClass.getImageDownloadUrl())
imageResultDict.setdefault("savePath", imageResultClass.getSavePath())
imageResultDict.setdefault("crawData", imageResultClass.getCrawDate())
imageResultDict.setdefault("md5", imageResultClass.getMd5())
return imageResultDict
except BaseException:
logger.error(traceback.format_exc())
return None
def writeToDatabase(self, imageResultDict):
'''
information write to database
:param infoDict:
:return:
'''
if imageResultDict is None:
logger.error("imageResultDict is None")
return None
try:
result = self._database.queryIsExist("image", {"md5": imageResultDict["md5"]})
progressBar = self.computeProgressBar(self.completed, self.totalDownload)
if not result:
logger.info("Image completed: {progressBar: <10}".format(progressBar=progressBar) + \
"Title:{title}".format(title=imageResultDict['title']),
level="ALL")
self._database.insert("image", imageResultDict)
except BaseException:
logger.error(traceback.format_exc())
logger.error("An error occurred in the function ---> wirteToDataBase")
return None
def query(self, tableName, position=None):
'''
输入表名和位置查询当前行,若位置为空则查询全部.
如果postion存在key"query_col"则请求选择列
例如:
postions = {
"query_col":"name"
}
请求name列所有内容
postions = {
"name:"wen lyuzhao",
"age":"24"
}
请求name=wen lyuzhao并且age=24的行
返回格式为一个列表字典,例如:
[
{"name":"naonao","name":"mama"}
...
]
'''
conn = self.pool.connection()
cursor = conn.cursor()
sql = ""
try:
# judge have or not positions key
if position is not None:
# has postion
# get column all values if have "query_col" key
if "query_col" in position.keys():
rows = position["query_col"]
sql = "SELECT " + str(rows) + " FROM " + str(tableName)
else:
# has postion
sql = "SELECT * FROM " + str(tableName) + " WHERE "
for key in position.keys():
sql += str(key) + " = " + "'" + str(
position[key]) + "'" + " and "
sql = sql[:len(sql) - 5]
else:
# get all
sql = "SELECT * FROM " + str(tableName)
sql += " FOR UPDATE"
cursor.execute(sql)
conn.commit()
resultTupleList = cursor.fetchall()
resultDictList = []
for result in resultTupleList:
resultDict = {}
resultDict.setdefault("id", result[0])
resultDict.setdefault("category", result[1])
resultDict.setdefault("title", result[2])
resultDict.setdefault("detailsPageUrl", result[3])
resultDict.setdefault("downloadPageUrl", result[4])
resultDict.setdefault("torrentDownloadUrl", result[5])
resultDict.setdefault("savePath", result[6])
resultDict.setdefault("crawData", result[7])
resultDict.setdefault("md5", result[8])
resultDict.setdefault("magnet", result[9])
resultDict.setdefault("downloaded", result[10])
resultDictList.append(resultDict)
return resultDictList
except BaseException as e:
logger.error(traceback.format_exc())
logger.error("error sql ---> " + sql)
return []
finally:
cursor.close()
conn.close()
