def consumer(self):
'''
used to consume details response
get a details page response by queue, then get image download url by details page response,
then save image to local dir, then write to database
:return:
'''
while not self._taskQueue.empty() or self._queueProducerStatus:
# get detailsResponse
detailsResponse = self._taskQueue.get()
if detailsResponse is None:
continue
# get image download url result is a imageResultClass generator
imageDownloadUrlGenerator = self.getImageDownloadUrlGenerator(detailsResponse)
if imageDownloadUrlGenerator is None:
continue
# save image to dir result is a imageResultClass
imageResultClassGenerator = self.saveImageGenerator(imageDownloadUrlGenerator)
for imageResultClass in imageResultClassGenerator:
if imageResultClass is None:
continue
# imageResultClass to imageResultDict
imageResultDict = self.imageResultClassPreprocessing(imageResultClass)
if imageResultDict is None:
continue
# imageResultDict write to database
self.writeToDatabase(imageResultDict)
self.completed += 1
progressBar = self.computeProgressBar(self.completed, self.totalDownload)
logger.info("Image completed:{0}".format(progressBar))
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 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 addToAria2(self, torentResultDict):
magnet = torentResultDict["magnet"]
savePath = self.getSavePath(torentResultDict)
torrenName = torentResultDict["title"]
logger.info("add to aria2, torrent file: {0}".format(torrenName))
self._pyaria2.addUrls(magnet, savePath)
self._completed += 1
def downloadImage(cls):
logger.info("start download image")
taskQueue = Queue(maxsize=128)
imageSpider = Image(taskQueue)
threading.Thread(target=imageSpider.producer, name="producer").start()
for i in range(DOWNLOAD_CONCURRENT_COUNT * 2):
threading.Thread(target=imageSpider.consumer, name="ThreadID:{0}".format(str(i))).start()
def downloadTorrent(cls):
logger.info("start download torrent")
taskQueue = Queue(maxsize=128)
torrentSpider = Torrent(taskQueue)
threading.Thread(target=torrentSpider.producer, name="producer").start()
for i in range(DOWNLOAD_CONCURRENT_COUNT):
threading.Thread(target=torrentSpider.consumer, name="ThreadID:{0}".format(str(i))).start()
def torrentResultDictGenerator(self):
torrentResultDictList = self._database.query("torrent",
{"downloaded": 0})
self._totalDownload = len(torrentResultDictList)
logger.info("not downloaded torrent count {0}".format(
len(torrentResultDictList)))
for torentResultDict in torrentResultDictList:
yield torentResultDict
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 downloadVideo(cls):
logger.info("start download video")
downloadVideo = Video()
threading.Thread(target=downloadVideo.downloadScheduler, name="downloadVideo").start()
