def video_page(self, url, channel=None):
"""
Due to iqiyi import hot index instead of play count,
the crawler is updated on 2018-11-23
"""
url = self.rebuild_video_url(url)
start = time.time()
get_page = retry_get_url(url)
end = time.time() - start
print("first request costs %s seconds" % end)
if get_page is None:
print('Failed to get html page for url: %s' % url)
return None
get_page.encoding = 'utf-8'
page = get_page.text
soup = BeautifulSoup(page, 'html.parser')
page_info = soup.find("div", {"is": "i71-play"})[":page-info"]
page_dic = json.loads(page_info)
title = page_dic["tvName"]
url = page_dic["pageUrl"]
dura_str = page_dic["duration"]
duration = trans_duration(dura_str)
try:
releaser = page_dic["user"]["name"]
releaserUrl = page_dic["user"]["profileUrl"]
except:
releaser = None
releaserUrl = None
video_info = soup.find("div", {"is": "i71-play"})[":video-info"]
video_dic = json.loads(video_info)
release_time = video_dic["firstPublishTime"]
tvId = video_dic["tvId"]
start1 = time.time()
hot_idx_url = "https://pub.m.iqiyi.com/jp/h5/count/hotDisplay/?qipuId=%s" % tvId
get_hot_idx = retry_get_url(hot_idx_url)
end2 = time.time() - start1
print("second request costs %s seconds" % end2)
hot_idx_str = get_hot_idx.text
hot_idx = int(
re.findall("\d+", ' '.join(re.findall('"count":\d+',
hot_idx_str)))[0])
fetch_time = int(
datetime.datetime.timestamp(datetime.datetime.now()) * 1e3)
video_page_dict = copy.deepcopy(self.video_data)
video_page_dict["title"] = title
video_page_dict["url"] = url
video_page_dict["duration"] = duration
video_page_dict["releaser"] = releaser
video_page_dict["releaserUrl"] = releaserUrl
video_page_dict["release_time"] = release_time
video_page_dict["hot_idx"] = hot_idx
video_page_dict["fetch_time"] = fetch_time
video_page_dict["tvId"] = tvId
if channel is not None:
video_page_dict["channel"] = channel
return video_page_dict
def get_hot_videos(self,url="", max_page=10,**kwargs):
data_list = []
headers = {
"accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9",
"accept-encoding": "gzip, deflate",
"accept-language": "zh,zh-CN;q=0.9",
"cache-control": "max-age=0",
# "cookie": "pgv_pvi=3517925376; pgv_pvid=3591400976; RK=sDRQYhGkF/; ptcz=8100687e80e810853d573a8a9ced1155a9a9683321075161f61b773de19ff4c5; pac_uid=0_bf3968e8e3157; ts_uid=1260359885; tvfe_boss_uuid=082fecb8ba01b06d; QQLivePCVer=50181223; video_guid=ce0aa0f8275ad435; video_platform=2; bucket_id=9231001; mobileUV=1_1707c108811_53c13; tvfe_search_uid=3c2fd48b-03f8-4f63-af8c-bb2bd367af2b; ts_refer=www.baidu.com/link; pgv_info=ssid=s7741803072; ad_play_index=80",
# "if-modified-since": "Fri, 28 Feb 2020 08:00:00 GMT",
"referer": "https://v.qq.com/biu/ranks/?t=hotsearch&channel=hot",
"sec-fetch-mode": "navigate",
"sec-fetch-site": "none",
"sec-fetch-user": "******",
"upgrade-insecure-requests": "1",
}
res = retry_get_url(url,headers=headers,timeout=10,proxies=3)
page_text = res.content.decode("utf-8")
html = etree.HTML(page_text)
print(html)
xpath_list = html.xpath("//body[@class='page_search']/div[@class='search_container']/div[@class='wrapper']/div[@class='wrapper_main']/div")
for li in xpath_list:
title_url = li.xpath("./a/@href")
if title_url:
print(title_url)
data = crawler_qq_video_page(title_url[0])
if not data:
continue
data["is_hot"] = 1
data_list.append(data)
output_result(result_Lst=data_list,
platform=self.platform,
output_to_es_raw=True,
)
data_list.clear()
def get_releaser_follower_num(self, releaserUrl):
headers = {
"Accept-Encoding": "gzip",
# "sdk-version": "1",
"User-Agent": "ttnet okhttp/3.10.0.2",
"Host": "aweme.snssdk.com",
"Connection": "Keep-Alive",
}
releaser_id = self.get_releaser_id(releaserUrl)
time.sleep(1)
releaserUrl = 'https://{2}/aweme/v1/user/?ac=WIFI&device_id={1}&os_api=18&app_name=aweme&channel=App Store&device_platform=ipad&device_type=iPad6,11&app_version=8.7.1&js_sdk_version=1.17.2.0&version_code=8.7.1&os_version=13.2.3&screen_width=1536&user_id={0}'.format(
releaser_id, str(random.randint(40000000000, 90000000000)),
random.choice(self.api_list))
count = 0
while count < 3:
try:
count += 1
time.sleep(random.random())
get_page = retry_get_url(releaserUrl,
headers=headers,
proxies=10)
page = get_page.json()
follower_num = page["user"].get("follower_count")
print('%s follower number is %s' % (releaserUrl, follower_num))
releaser_img = page["user"].get("avatar_thumb").get(
"url_list")[0]
return follower_num, releaser_img
except:
print("can't find followers")
continue
else:
return None, None
def get_hot_words(self):
bulk_list = []
url = "https://i.snssdk.com/api/feed/hotboard_online/v1/?fe_api_version=2&is_in_channel=1&count=50&fe_source=hot_board&tab_name=tab_hot_board&is_web_refresh=1&style_type=18&client_extra_params=%7B%22hot_board_source%22%3A%22hot_board%22%2C%22fe_version%22%3A%22v11%22%7D&extra=%7B%22CardStyle%22%3A0%2C%22JumpToWebList%22%3Atrue%7D&category=hotboard_online&stream_api_version=88&tt_daymode=1&iid=105857671701&device_id=70787469432&ac=wifi&mac_address=48%3AA4%3A72%3A58%3A86%3AD5&channel=store_yingyonghui_0107&aid=13&app_name=news_article&version_code=763&version_name=7.6.3&device_platform=android&ab_version=801968%2C1419043%2C668775%2C1462526%2C1512584%2C1190522%2C1489307%2C1157750%2C1157634%2C1419598%2C1493796%2C1439625%2C1469498%2C668779%2C1417597%2C662099%2C1403340%2C668774%2C1509255%2C1396151%2C821967%2C857803%2C660830%2C1434501%2C662176%2C1491631&ab_feature=102749%2C94563&device_type=OPPO%2BR11&device_brand=OPPO&language=zh&os_api=22&os_version=5.1.1&uuid=866174725888628&openudid=48a4725886d57203&manifest_version_code=7630&resolution=900*1600&dpi=320&update_version_code=76309&plugin=0&tma_jssdk_version=1.54.0.3&rom_version=coloros__r11-user%2B5.1.1%2Bnmf26x%2B500200210%2Brelease-keys&cdid=754b9ff9-5880-48b2-ac40-3880effd3f33"
page_res = retry_get_url(url,
headers=self.headers,
proxies=3,
timeout=5)
page_json = page_res.json()
for data in page_json["data"]:
contect = data["content"]
data = json.loads(contect)
schema = data["raw_data"]["schema"]
# search_str = urllib.parse.unquote(schema)
query = urlparse(schema).query # wd=python&ie=utf-8
params = parse_qs(query) # {'wd': ['python'], 'ie': ['utf-8']}
"""所得的字典的value都是以列表的形式存在，若列表中都只有一个值"""
result = {key: params[key][0] for key in params}
search_title = result.get("keyword")
# search_json = result.get("search_json")
if search_title:
dic = {
"platform": self.platform,
"title": search_title,
"fetch_time":
int(datetime.datetime.now().timestamp() * 1e3),
"search_json": schema
}
bulk_list.append(dic)
hot_words_output_result(bulk_list)
return True
def releaser_video_sum(self, releaserUrl):
get_page = retry_get_url(releaserUrl)
get_page.encoding = 'utf-8'
page = get_page.text
soup = BeautifulSoup(page, 'html.parser')
total_video_num_str = soup.find('div', {'class':'title'}).span.text
total_video_num = total_video_num_str.replace('(', '').replace(')', '').replace(',', '')
total_video_num = trans_play_count(total_video_num)
return total_video_num
def search_page(self, title=None, search_json=None, **kwargs):
data_list = []
timestamp = int(datetime.datetime.now().timestamp() * 1e3)
title = urllib.parse.quote(title)
headers = {
"Accept-Encoding": "gzip",
# "X-SS-REQ-TICKET": "1587102750860",
"passport-sdk-version": "14",
"sdk-version": "2",
#"Cookie": "odin_tt=d5d96b2812637e9d20681530fbbe4d52e8f76ae1b6afa8c0a173260321611c507ac6eca10991b21fc4f023e94371d457df784f959e94db673ef29a5bd2137091; qh[360]=1; history=alrvlFic6pJZXJCTWBmSmZt6KW6mevZSz5LU3OJ7DEKX42Zw%2Bc84wMR3iYGBweFy3EzZsPcNTLyXWN1AvLYP8%2BQPMLFfEpUA8bo%2F7nNtYOK7xNwC4k3XmMHe5MtzSTiM48DluNr01dkNTDyXuHrApsi4ejkwsV%2BSmAPmSeXoMzDxXhKcAuIVrRfWAJnJJwA25fG1DoezvFBTZrzZeg6kT%2BwWSG7Gx3UJB5h4L%2FH4gXlVn%2BtAtkvFMQRcjpv%2B%2Be9TBib2S%2BwcYBuUn8xsYGK%2FJKMAkptgfXrDASaOS4yHQHJVPy6UOjDxXuI4BeJN26Fs6MDEcYn%2FEoMDAAAA%2F%2F8%3D; install_id=112651077855; ttreq=1$0b37d53ca5c301ce96959dc97a67886da420b294",
# "X-Gorgon": "0401007140017aae019cc2020b1c48dbab0ba42839014487648a",
#"X-Khronos": "1587102750",
"Host": "is.snssdk.com",
"Connection": "Keep-Alive",
"User-Agent": "okhttp/3.10.0.1",
}
url = "https://is.snssdk.com/api/search/content/?os_api=23&device_type=oneplus+a5010&from_search_subtab=synthesis&manifest_version_code=7690&source=search_subtab_switch&offset=0&is_ttwebview=0&action_type&is_incognito=0&keyword_type&rom_version=23&app_name=news_article&format=json&version_name=7.6.9&ac=wifi&host_abi=armeabi-v7a&update_version_code=76909&channel=baidu_0411&is_native_req=1&loadId=1&longitude=116.40717530841052&isIncognito=0&plugin=2050&forum=1&latitude=39.904680919672145&language=zh&pd=video&cur_tab_title=search_tab&aid=13&dpi=270&qrecImprId&fetch_by_ttnet=1&count=10&plugin_enable=3&search_position&ab_group=100167%2C94569%2C102754&keyword={0}&scm_version=1.0.2.830&search_json=%7B%22comment_ids%22%3A%5B%5D%2C%22event_discussion%22%3A74123%2C%22event_impression%22%3A17270790%2C%22forum_id%22%3A1664181806902302%2C%22forum_recall_wtt%22%3A%5B1664190666034183%2C1664192273575943%2C1664184430218253%2C1664185769175051%2C1664184985139212%2C1664196237152267%2C1664186792648732%2C1664188755414019%2C1664187055838215%2C1664184182571022%2C1664185938950148%2C1664188041995268%2C1664188322863172%2C1664190185024520%2C1664185602828300%2C1664184276484099%2C1664188211399684%2C1664187870713868%2C1664184484958211%2C1664183864289288%2C1664186825487371%2C1664195548700686%2C1664186585780228%2C1664197296210947%2C1664188146725901%2C1664191748459523%5D%2C%22group_source%22%3Anull%2C%22hot_gid%22%3A6816255461172445703%2C%22log_pb%22%3A%7B%22cluster_type%22%3A%220%22%2C%22entrance_hotspot%22%3A%22channel%22%2C%22hot_board_cluster_id%22%3A%226816091697949180424%22%2C%22hot_board_impr_id%22%3A%22202004171352010100140411610B1A7741%22%2C%22location%22%3A%22hot_board%22%2C%22rank%22%3A%225%22%2C%22source%22%3A%22trending_tab%22%2C%22style_id%22%3A%2210005%22%7D%2C%22mix_stick_ids%22%3A%5B1664190666034183%2C1664192273575943%2C1664184430218253%2C1664185769175051%2C1664184985139212%2C1664196237152267%2C1664186792648732%2C1664188755414019%2C1664187055838215%2C1664184182571022%2C1664185938950148%2C1664188041995268%2C1664188322863172%2C1664190185024520%2C1664185602828300%2C1664184276484099%2C1664188211399684%2C1664187870713868%2C1664184484958211%2C1664183864289288%2C1664186825487371%2C1664195548700686%2C1664186585780228%2C1664197296210947%2C1664188146725901%2C1664191748459523%5D%2C%22stick_group_ids%22%3A%5B%5D%7D&device_platform=android&search_id&has_count=0&version_code=769&from=video&device_id={1}&resolution=1080*1920&os_version=6.0.1&device_brand=Oneplus&search_sug=1&qc_query".format(
title, random.randint(69418800000, 69418899999))
res = retry_get_url(url, headers=headers, timeout=5, proxies=3)
page_text = res.json()
for one_video in page_text["data"]:
video_dic = {}
try:
video_dic['title'] = one_video.get('title')
video_dic['url'] = one_video.get('display').get("info").get(
"url")
releaser_id = re.findall("user_id=(\d+)",
one_video.get('user_source_url'))[0]
video_dic['releaser'] = one_video.get('media_name')
video_dic[
'releaserUrl'] = "https://www.toutiao.com/c/user/%s/" % releaser_id
release_time = int(one_video.get('create_time'))
video_dic['release_time'] = int(release_time * 1e3)
video_dic['duration'] = int(one_video.get('video_duration'))
video_dic['play_count'] = trans_play_count(
one_video.get('play_effective_count'))
video_dic['repost_count'] = 0
video_dic['comment_count'] = one_video.get('comment_count')
video_dic['favorite_count'] = one_video.get('digg_count')
video_dic['fetch_time'] = int(
datetime.datetime.now().timestamp() * 1e3)
video_dic['releaser_id_str'] = "toutiao_%s" % releaser_id
video_dic['video_img'] = one_video.get('display').get(
'self_info').get('image_url')
video_dic['platform'] = "toutiao"
video_dic["is_hot"] = 1
video_dic["data_provider"] = "CCR"
except Exception as e:
print(e)
continue
data_list.append(video_dic)
output_result(
result_Lst=data_list,
platform=self.platform,
output_to_es_raw=True,
)
data_list.clear()
def search_page(self, title=None, **kwargs):
data_list = []
headers = {
"Host": "aweme.snssdk.com",
"Connection": "keep-alive",
# "Cookie": "d_ticket=38c841789e38ea43c6338910dac65ffe192e3; odin_tt=82086544bb9028f027b5aea78724ccf512dead26658f45321be33bade615793782bf6ac7fe0c18b73b9592f4284413d5300974810d439b42ef0b3eaa761b1640; msh=cakLg8lvbK5CxiSWkIbD2UInwAI; sid_guard=09fe3dfd89dfbc79f081fb2db9dd81ee%7C1581832192%7C5184000%7CThu%2C+16-Apr-2020+05%3A49%3A52+GMT; uid_tt=da0b53b7563eca87c47da41f5f17c30f; uid_tt_ss=da0b53b7563eca87c47da41f5f17c30f; sid_tt=09fe3dfd89dfbc79f081fb2db9dd81ee; sessionid=09fe3dfd89dfbc79f081fb2db9dd81ee; sessionid_ss=09fe3dfd89dfbc79f081fb2db9dd81ee; install_id=104847319549; ttreq=1$51e484720311469c4b70f4754d730d538a074c4b",
# "X-SS-REQ-TICKET": "1583139618192",
# "X-Tt-Token": "0009fe3dfd89dfbc79f081fb2db9dd81ee013243f7134b3eb37249cc729a5276172df69a4391b56ae4bf253c3c6352322611",
"sdk-version": "1",
# "x-tt-trace-id": "00-9a797f160a107b431078db3e93480468-9a797f160a107b43-01",
"User-Agent":
"com.ss.android.ugc.aweme/990 (Linux; U; Android 5.1.1; zh_CN; OPPO R11; Build/NMF26X; Cronet/77.0.3844.0)",
"Accept-Encoding": "gzip, deflate",
# "X-Gorgon": "0401a0514001f64964a8ebef9f4305ccbef2df1aa3c92fdf955a",
# "X-Khronos": "1583139618",
}
url = "https://aweme.snssdk.com/aweme/v1/hot/search/video/list/?hotword={0}&offset=0&count=12&source=trending_page&is_ad=0&item_id_list&is_trending=0&os_api=22&device_type=OPPO%20R11&ssmix=a&manifest_version_code=990&dpi=320&uuid=866174725888628&app_name=aweme&version_name=9.9.0&ts=1583139619&app_type=normal&ac=wifi&update_version_code=9902&channel=tengxun_new&_rticket=1583139618192&device_platform=android&iid=104847319549&version_code=990&cdid=fce00742-ccef-4b14-943d-1f62b6d637b0&openudid=48a4725886d57203&device_id=70787469432&resolution=900*1600&os_version=5.1.1&language=zh&device_brand=OPPO&aid=1128&mcc_mnc=46007".format(
title)
res = retry_get_url(url, headers=headers, timeout=5, proxies=3)
page_text = res.json()
for one_video in page_text["aweme_list"]:
video_dic = {}
video_dic['title'] = one_video.get('desc')
video_dic['url'] = one_video.get('share_url')
releaser_id = one_video.get('author_user_id')
video_dic['releaser'] = one_video.get('author').get("nickname")
video_dic[
'releaserUrl'] = "https://www.iesdouyin.com/share/user/%s" % releaser_id
release_time = one_video.get('create_time')
video_dic['release_time'] = int(release_time * 1e3)
video_dic['duration'] = int(one_video.get('duration') / 1000)
video_dic['play_count'] = 0
video_dic['repost_count'] = one_video.get('statistics').get(
'share_count')
video_dic['comment_count'] = one_video.get('statistics').get(
'comment_count')
video_dic['favorite_count'] = one_video.get('statistics').get(
'digg_count')
video_dic['video_id'] = one_video.get('aweme_id')
video_dic['fetch_time'] = int(datetime.datetime.now().timestamp() *
1e3)
video_dic['releaser_id_str'] = "抖音_%s" % releaser_id
video_dic['platform'] = "抖音"
video_dic['video_img'] = one_video.get('video').get('cover').get(
'url_list')[0]
video_dic["is_hot"] = 1
video_dic["data_provider"] = "CCR"
data_list.append(video_dic)
output_result(
result_Lst=data_list,
platform=self.platform,
output_to_es_raw=True,
)
data_list.clear()
def redirect_by_js(self, url_raw):
get_raw = retry_get_url(url_raw)
if get_raw is not None:
raw_page = get_raw.text
soup = BeautifulSoup(raw_page, 'html.parser')
soupf = soup.find_all(name='div', attrs={'class': 'cms-qipuId'})
if soupf != []:
data_qipuId = soupf[0].attrs['data-qipuid']
url_redirect = 'http://www.iqiyi.com/v_%s.html' % data_qipuId
return url_redirect
else:
return None
else:
print('Failed to get redirect by js for raw url: %s' % url_raw)
return None
def get_hot_words(self):
bulk_list = []
url = "https://apis.tudou.com/search/v1/hot?_t_={0}&e=md5&_s_=9a4abf3a92efad0605f8e31481327014&operator=CHINA+MOBILE_46007&network=WIFI".format(
int(datetime.datetime.now().timestamp()))
res = retry_get_url(url, proxies=3, headers=self.headers)
res_json = res.json()
for title in res_json["result"]["search"]["data"]:
dic = {
"platform": self.platform,
"title": title["keyword"],
"fetch_time": int(datetime.datetime.now().timestamp() * 1e3)
}
bulk_list.append(dic)
hot_words_output_result(bulk_list)
return True
def get_releaser_follower_num(self, releaserUrl):
releaser_id = self.get_releaser_id(releaserUrl)
releaserUrl = 'https://{2}/aweme/v1/user/?ac=WIFI&device_id={1}&os_api=18&app_name=aweme&channel=App Store&device_platform=ipad&device_type=iPad6,11&app_version=8.7.1&js_sdk_version=1.17.2.0&version_code=8.7.1&os_version=13.2.3&screen_width=1536&user_id={0}'.format(
releaser_id, str(random.randint(40000000000, 90000000000)), random.choice(self.api_list))
count = 0
while count < 3:
try:
count += 1
time.sleep(random.randint(1, 2))
get_page = retry_get_url(releaserUrl, headers=self.headers, proxies=1)
page = get_page.json()
follower_num = page["user"].get("follower_count")
print('%s follower number is %s' % (releaserUrl, follower_num))
releaser_img = page["user"].get("avatar_thumb").get("url_list")[0]
return follower_num, releaser_img
except:
print("can't find followers")
continue
else:
return None, None
def get_hot_words(self):
bulk_list = []
url = "http://c.m.163.com/nc/search/hotWord.html"
page_res = retry_get_url(url,
headers=self.headers,
proxies=3,
timeout=5)
page_json = page_res.json()
for data in page_json["hotWordList"]:
title = data["searchWord"]
if title:
dic = {
"platform": self.platform,
"title": title,
"fetch_time":
int(datetime.datetime.now().timestamp() * 1e3),
}
bulk_list.append(dic)
hot_words_output_result(bulk_list)
return True
def get_hot_videos(self, title=None, max_page=10, **kwargs):
page = 1
while page <= max_page:
get_dic = {
"keyword": title,
# "pid": "6c23a6957198fad2",
# "guid": "2139ff131a8a7d9ef7d3014cc8b97010",
"mac": "",
"imei": "null",
"ver": "6.39.1",
"_t_": int(datetime.datetime.now().timestamp()),
"e": "md5",
# "_s_": "b905d3a9738d7d2f815687428563d8f7",
"operator": "CHINA+MOBILE_46007",
"network": "WIFI",
"ftype": "0",
"cateId": "0",
"seconds": "0",
"seconds_end": "0",
"ob": "",
"pg": str(page),
"pz": "30",
# "aaid": "1.58259884569785E+20",
"brand": "OPPO",
"btype": "OPPO+R11",
"sdkver": "2",
"apad": "0",
# "utdid": "XkjV9GsfBysDACyQ2%2BiF8MOw",
"srid": "1",
"userType": "guest",
}
requests_res = retry_get_url(
"https://apis.tudou.com/search/v2/integration?%s" %
urllib.parse.urlencode(get_dic),
headers=self.headers,
proxies=3)
requests_json = requests_res.json()
page += 1
print(requests_json)
for data in requests_json["results"]["ugc"]["data"]:
print(data)
def get_hot_words(self):
bulk_list = []
url = "https://api3-normal-c-lf.amemv.com/aweme/v1/hot/search/list/?detail_list=1&mac_address=48%3AA4%3A72%3A58%3A86%3AD5&os_api=22&device_type=OPPO%20R11&ssmix=a&manifest_version_code=990&dpi=320&uuid=866174725888628&app_name=aweme&version_name=9.9.0&app_type=normal&ac=wifi&update_version_code=9902&channel=tengxun_new&device_platform=android&iid=104847319549&version_code=990&cdid=fce00742-ccef-4b14-943d-1f62b6d637b0&openudid=48a4725886d57203&device_id=70787469432&resolution=900*1600&os_version=5.1.1&language=zh&device_brand=OPPO&aid=1128&mcc_mnc=46007"
page_res = retry_get_url(url,
headers=self.headers,
proxies=3,
timeout=5)
page_json = page_res.json()
for data in page_json["data"]["word_list"]:
title = data["word"]
if title:
dic = {
"platform": self.platform,
"title": title,
"fetch_time":
int(datetime.datetime.now().timestamp() * 1e3),
"hot_value": data.get("hot_value"),
"top": data.get("position"),
}
bulk_list.append(dic)
hot_words_output_result(bulk_list)
return True
def search_page(self, title):
data_list = []
encodestr = base64.b64encode(title.encode('utf-8'))
encodestr = str(encodestr, 'utf-8')
url = "http://c.m.163.com/search/comp2/Kg%3D%3D/20/{0}.html?".format(
encodestr)
para = "deviceId=2zx5YfHmoBb72ayxYpQVUg%3D%3D&version=newsclient.32.1.android&channel=VDEzNDg2NDc5MDkxMDc%3D&canal=bmV3c19sZl9jcGFfMg%3D%3D&dtype=0&tabname=shipin&position=5YiX6KGo6aG26YOo&ts={0}&sign=Di3opZw%2FFIPDdgreSK4VCKlnMSpm6FPoel5LeY88RgZ48ErR02zJ6%2FKXOnxX046I&spever=FALSE&open=scheme_%E9%BB%98%E8%AE%A4&openpath=/video/VT5O1KVCO".format(
str(int(datetime.datetime.now().timestamp())))
res = retry_get_url(url + para,
headers=self.headers,
timeout=5,
proxies=3)
page_text = res.json()
for data in page_text["doc"]["result"]:
print(data)
data_list.append(data)
output_result(
result_Lst=data_list,
platform=self.platform,
output_to_es_raw=True,
)
data_list.clear()
def get_hot_words(self):
bulk_list = []
timestamp = int(datetime.datetime.now().timestamp() * 1e3)
url = "https://w.inews.qq.com/searchPage?pagefrom=moreHotDetail&adcode=310112&is_special_device=0&mid=0&dpi=320.0&qqnetwork=wifi&rom_type=R11-user%205.1.1%20NMF26X%20500200210%20release-keys&isColdLaunch=1&real_device_width=2.81&net_proxy=DIRECT@&net_bssid=48:A4:72:58:86:D5&isMainUserLogin=0¤tChannelId=_qqnews_custom_search&isElderMode=0&apptype=android&islite=0&hw=OPPO_OPPOR11&baseid=&global_session_id={0}&screen_width=900&omgbizid=&isClosePersonalized=0&sceneid=&videoAutoPlay=1&imsi=460077203886213&fix_store=&cpuabi=armeabi-v7a&isoem=0¤tTabId=news_news&lite_version=&startTimestamp={1}&net_slot=0&qn-time={2}&pagestartfrom=icon&mac=48:A4:72:58:86:D5&activefrom=icon&net_ssid=R1148a4725886d57203&store=17&screen_height=1600&top_activity=DailyHotDetailActivity&real_device_height=5.0&origin_imei=866174725888628&network_type=wifi&origCurrentTab=top&global_info=1|1|1|1|1|14|4|1|0|6|1|1|1||0|J309P000000000:J902P000000000:J601P900000000:A601P800217702:A601P700321102:B601P600286205:A601P500154501:A601P400161601:J601P300000000:B601P200096102:A601P100272502:A601P000261102:J601P904000000:J601P903000000:A601P902266601:A601P901291001:J601P811000000:A601P701226201:A601P622269601:A601P621294101:A601P620269601:J601P111000000:J601P110000000:A601P109107102:A601P105118803:A601P019237403:A601P016212405:J601P006000000:J603P000000000:J401P100000000:A401P000050901:J602P900000000:J602P800000000:J602P700000000:J602P600000000:A602P500267502:B602P400286004:J602P300000000:J602P200000000:J602P100000000:B602P000315504:A602P901257901:J602P616000000:A602P615304801:A602P613271701:A602P611253801:A602P516234601:A602P414259901:A602P307160708:J602P302000000:A602P208205801:J602P117000000:A602P007272801:A602P003136401:J304P000000000:J310P700000000:A310P200210802:J310P100000000:B310P020314103:A310P010301701:B310P000267107:B701P000323002:A703P000322204:A704P000309801:J702P000000000:J405P000000000:J064P400000000:J064P300000000:B064P100243802:B064P020290902:J064P010000000:J064P000000000:A085P000087701:B074P200238202:J074P040000000:B074P030315703:A074P020315602:A074P010315401:B074P000142402:J903P000000000:A267P300215801:A267P200263601:A267P100299801:B267P000300102:A073P040317201:B073P030314503:A073P020313801:J073P010000000:B073P000313603:J060P700000000:J060P300000000:J060P200000000:B060P100299703:A060P090287301:J060P020000000:J060P010000000:B060P000311102:J060P099000000:J060P016000000:A406P000313203:J403P700000000:J403P600000000:A403P200206702:B403P100246105:J403P010000000:A403P000310401:A403P602218702:B404P200262402:A404P000263407:J055P200000000:J055P090000000:J055P080000000:J055P070000000:J055P060000000:J055P050000000:J055P010000000:A055P000265801:J402P100000000:J402P090000000:J402P080000000:J402P060000000:J402P020000000:A402P000301403:J054P400000000:J054P300000000:J054P200000000:A054P100269701:B054P090289604:A054P080289702:J054P050000000:J054P040000000:A054P030288501:J054P010000000:A054P000319901:J056P000000000:A901P200252304:B901P100226405:B901P000232405:J407P000000000|1402|0|1|25|25|0|0|0||3|3|1|1|1|1|1|1|-1|0|0|5|2|0|0|0|3|0|0|1|3|0|2|0|0|2|0|0|1|0|1|1|0|0|1|0|4|0|1|1|11|20|1|0|1|1|0|0|1|4|0|1|1|41|2|51|60|0|1|0|0|1|5|1|0|0|71|0|0|1|71&imsi_history=460077203886213&net_apn=0&uid=48a4725886d57203&omgid=&trueVersion=6.0.40&qimei=866174725888628&devid=866174725888628&appver=22_android_6.0.40&Cookie=lskey%3D;skey%3D;uin%3D;%20luin%3D;logintype%3D0;%20main_login%3D;%20&qn-sig=74f9daefb0544a8cff5f2d1e0b465fd0&qn-rid=1002_1218eb69-f164-474c-9330-04d620a35c93&qn-newsig=68896f7d5f840c8540a9dff8877c88c277879f095408ca81edffe1a8b05ba94f".format(
timestamp, int(timestamp / 1000), timestamp)
page_res = retry_get_url(url,
headers=self.headers,
proxies=3,
timeout=10)
page_json = page_res.json()
for data in page_json["showInfo"]:
search_title = data["desc"]
if search_title:
dic = {
"platform": self.platform,
"title": search_title,
"fetch_time":
int(datetime.datetime.now().timestamp() * 1e3),
}
bulk_list.append(dic)
hot_words_output_result(bulk_list)
return True
def get_hot_words(self):
bulk_list = []
timestamp = int(datetime.datetime.now().timestamp())
url = "https://v.qq.com/biu/ranks/?t=hotsearch&channel=hot"
headers = {
"accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9",
"accept-encoding": "gzip, deflate",
"accept-language": "zh,zh-CN;q=0.9",
"cache-control": "max-age=0",
# "cookie": "pgv_pvi=3517925376; pgv_pvid=3591400976; RK=sDRQYhGkF/; ptcz=8100687e80e810853d573a8a9ced1155a9a9683321075161f61b773de19ff4c5; pac_uid=0_bf3968e8e3157; ts_uid=1260359885; tvfe_boss_uuid=082fecb8ba01b06d; QQLivePCVer=50181223; video_guid=ce0aa0f8275ad435; video_platform=2; bucket_id=9231001; mobileUV=1_1707c108811_53c13; tvfe_search_uid=3c2fd48b-03f8-4f63-af8c-bb2bd367af2b; ts_refer=www.baidu.com/link; ad_play_index=71; pgv_info=ssid=s7741803072; ts_last=v.qq.com/biu/ranks/",
"if-modified-since": "Fri, 28 Feb 2020 07:10:00 GMT",
"sec-fetch-mode": "navigate",
"sec-fetch-site": "none",
"sec-fetch-user": "******",
"upgrade-insecure-requests": "1",
"user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/79.0.3945.130 Safari/537.36",
}
page_res = retry_get_url(url, headers=headers, proxies=3, timeout=5)
page_text = page_res.content.decode("utf-8")
html = etree.HTML(page_text)
print(html)
xpath_list = html.xpath("//ul[@class='table_list']/li")
for li in xpath_list:
title = li.xpath("./div[1]/a/@title")
title_url = li.xpath("./div[1]/a/@href")
if title:
dic = {
"platform": self.platform,
"title": title[0],
"fetch_time": int(datetime.datetime.now().timestamp() * 1e3),
"url":title_url[0]
}
bulk_list.append(dic)
hot_words_output_result(bulk_list)
return True
def list_page_single(self, listurl, channel=None):
"""
To be solved: video collection page, mainly include TV series
and cartoon series, such as
http://www.iqiyi.com/a_19rrh7z5vx.html#vfrm=2-4-0-1
http://www.iqiyi.com/a_19rrk3ndgl.html#vfrm=2-4-0-1
and zongyi show, such as
http://www.iqiyi.com/a_19rrhbfb4d.html#vfrm=2-4-0-1
"""
list_page_Lst = []
get_page = retry_get_url(listurl)
if get_page is None:
print('Failed to get singe list page for url: %s' % listurl)
return None
print(listurl)
get_page.encoding = 'utf-8'
page = get_page.text
soup = BeautifulSoup(page, 'html.parser')
iqiyi = soup.find_all('div', {'class': 'site-piclist_pic'})
for data_line in iqiyi:
try:
url = data_line.find('a')['href']
except TypeError:
url = None
if url is not None:
video_page_dict = self.video_page(url, channel)
if video_page_dict is not None:
list_page_Lst.append(video_page_dict)
else:
print('Got None on video page url: %s' % url)
# # for test
# video_page_dict = {'url': url}
# list_page_Lst.append(video_page_dict)
else:
pass
return list_page_Lst
def releaser_page_web(self,
releaserUrl,
output_to_file=False,
filepath=None,
releaser_page_num_max=30,
output_to_es_raw=False,
es_index=None,
doc_type=None,
output_to_es_register=False,
push_to_redis=False,
proxies_num=None):
releaser_id = self.get_releaser_id(releaserUrl)
# releaser = self.get_releaser_name(releaserUrl)
releaserUrl = 'https://id.tudou.com/i/%s/videos' % releaser_id
json_headers = {
"accept": "application/json, text/javascript, */*; q=0.01",
"accept-encoding": "gzip, deflate, br",
"accept-language": "zh,zh-CN;q=0.9",
# "cookie": "cna=W99aFOvX+QACAXL4fBJI3rAw; __ysuid=1541219939103JPW; ykss=e93bad5ef9c26af71c8e7ee5; P_ck_ctl=47F163FE35A5B1B2E479B158A12376A7; __ayvstp=16; __aysvstp=16; _zpdtk=ecd18a6d5d86a28b786b653356133cfb606dd1dc; isg=BOzsOnpUnhIGhYq8YxHgZ36EvcoepZBPH_JJJ0Yt-Rc6UY5bbrVJ3rr3dxdpWcin",
"referer": releaserUrl,
"sec-fetch-dest": "empty",
"sec-fetch-mode": "cors",
"sec-fetch-site": "same-origin",
"user-agent":
"Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 Mobile/15E148 Safari/604.1",
"x-csrf-token": "ecd18a6d5d86a28b786b653356133cfb606dd1dc",
"x-requested-with": "XMLHttpRequest",
}
json_cookies = {
"cna":
"W99aFOvX+QACAXL4fBJI3rAw",
"__ysuid":
"1541219939103JPW",
"ykss":
"e93bad5ef9c26af71c8e7ee5",
"P_ck_ctl":
"47F163FE35A5B1B2E479B158A12376A7",
"__ayvstp":
"16",
"__aysvstp":
"16",
"_zpdtk":
"ecd18a6d5d86a28b786b653356133cfb606dd1dc",
"isg":
"BOzsOnpUnhIGhYq8YxHgZ36EvcoepZBPH_JJJ0Yt-Rc6UY5bbrVJ3rr3dxdpWcin",
}
firsh_page_headers = {
"accept":
"text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9",
"accept-encoding":
"gzip, deflate, br",
"accept-language":
"zh,zh-CN;q=0.9",
# "cookie": "cna=W99aFOvX+QACAXL4fBJI3rAw; __ysuid=1541219939103JPW; ykss=e93bad5ef9c26af71c8e7ee5; P_ck_ctl=47F163FE35A5B1B2E479B158A12376A7; __ayvstp=16; __aysvstp=16; _zpdtk=9053e5d58ee0c51b1f3da8008dd4bda164ecd846; isg=BHl5FRo0A8WDkd_DnlItMBsXiOVThm042sF8-Juu9KAfIpu049ZUCb80oCjUmgVw",
"referer":
releaserUrl,
"sec-fetch-dest":
"document",
"sec-fetch-mode":
"navigate",
"sec-fetch-site":
"same-origin",
"sec-fetch-user":
"******",
"upgrade-insecure-requests":
"1",
"user-agent":
"Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0.3 Mobile/15E148 Safari/604.1",
}
first_page_res = retry_get_url(releaserUrl,
headers=firsh_page_headers,
proxies=proxies_num)
json_cookies.update(dict(first_page_res.cookies))
user_id = re.findall('uid="(\d+)"', first_page_res.text)[0]
zptk_url = "https://id.tudou.com/i/h5/id_%s/playlisttab?uid=%s" % (
user_id, user_id)
playlisttab_res = retry_get_url(zptk_url,
headers=json_headers,
proxies=proxies_num,
cookies=json_cookies)
# print(dict(playlisttab_res.cookies))
json_cookies.update(dict(playlisttab_res.cookies))
json_headers["x-csrf-token"] = dict(playlisttab_res.cookies)["_zpdtk"]
count = 1
retry_time = 0
result_list = []
self.video_data['releaserUrl'] = releaserUrl
print("working on releaser_id: %s" % (releaser_id))
while count <= releaser_page_num_max and retry_time < 5:
proxies = get_proxy(proxies_num)
api_url = 'https://id.tudou.com/i/h5/id_%s/videos?ajax=1&pn=%s&pl=20' % (
user_id, count)
print(api_url)
if proxies:
get_page = requests.get(api_url,
headers=json_headers,
proxies=proxies,
timeout=3,
cookies=json_cookies)
else:
get_page = requests.get(api_url,
headers=json_headers,
timeout=3,
cookies=json_cookies)
_zpdtk = dict(get_page.cookies)
json_cookies.update(_zpdtk)
# print(dict(get_page.cookies))
json_headers["x-csrf-token"] = _zpdtk["_zpdtk"]
page_dic = get_page.json()
releaser_page_num_max = page_dic["page"]["pz"]
releaser = page_dic['channelOwnerInfo']["data"]["nickname"]
# has_more = page_dic.get('has_more')
try:
data_list = page_dic['data']["data"]
time.sleep(0.25)
except:
retry_time += 1
time.sleep(0.25)
print("no more data at page: %s try_time: %s" %
(count, retry_time))
continue
if data_list == []:
retry_time += 1
time.sleep(0.25)
print("no more data at page: %s try_time: %s" %
(count, retry_time))
continue
else:
retry_time = 0
print("get data at page: %s" % (count))
count += 1
for info_dic in data_list:
video_info = copy.deepcopy(self.video_data)
video_info['video_id'] = info_dic["videoid"]
video_info['title'] = info_dic["title"]
video_info['releaser'] = releaser
video_info[
'url'] = 'https://video.tudou.com/v/%s.html' % info_dic[
"videoid"]
video_info['duration'] = int(info_dic.get('seconds') / 1e3)
video_info['releaser_id_str'] = "new_tudou_%s" % (
releaser_id)
video_info['comment_count'] = int(
info_dic.get('total_comment'))
video_info['favorite_count'] = int(
info_dic.get('total_up'))
# favorite_count in database means 点赞数, while in web page the factor
# named praiseNumber
# in web page facorite_count means 收藏数
video_info['video_img'] = info_dic.get('thumburl')
video_info['play_count'] = info_dic.get('total_vv')
video_info['release_time'] = int(
info_dic.get('publishtime') * 1e3)
# print(video_info['release_time'])
# if '天前' in release_time_str:
# video_info['release_time'] = self.video_page(video_info['url'])['release_time']
# else:
# video_info['release_time'] = trans_strtime_to_timestamp(input_time=release_time_str,
# missing_year=True)
video_info['fetch_time'] = int(time.time() * 1e3)
yield video_info
def search_short_video_page(self, title=None, search_json=None, **kwargs):
data_list = []
timestamp = int(datetime.datetime.now().timestamp() * 1e3)
title = urllib.parse.quote(title)
headers = {
"Accept-Encoding": "gzip",
# "X-SS-REQ-TICKET": "1587103224961",
"passport-sdk-version": "14",
"sdk-version": "2",
#"Cookie": "odin_tt=d5d96b2812637e9d20681530fbbe4d52e8f76ae1b6afa8c0a173260321611c507ac6eca10991b21fc4f023e94371d457df784f959e94db673ef29a5bd2137091; qh[360]=1; history=alrvlFic6pJZXJCTWBmSmZt6KW6mevZSz5LU3OJ7DEKX42Zw%2Bc84wMR3iYGBweFy3EzZsPcNTLyXWN1AvLYP8%2BQPMLFfEpUA8bo%2F7nNtYOK7xNwC4k3XmMHe5MtzSTiM48DluNr01dkNTDyXuHrApsi4ejkwsV%2BSmAPmSeXoMzDxXhKcAuIVrRfWAJnJJwA25fG1DoezvFBTZrzZeg6kT%2BwWSG7Gx3UJB5h4L%2FH4gXlVn%2BtAtkvFMQRcjpv%2B%2Be9TBib2S%2BwcYBuUn8xsYGK%2FJKMAkptgfXrDASaOS4yHQHJVPy6UOjDxXuI4BeJN26Fs6MDEcYn%2FEoMDAAAA%2F%2F8%3D; install_id=112651077855; ttreq=1$0b37d53ca5c301ce96959dc97a67886da420b294",
# "X-Gorgon": "0401e08b4001a628dcf96b16d01278ad842e915d905b213dc48f",
# "X-Khronos": "1587103224",
"Host": "is.snssdk.com",
"Connection": "Keep-Alive",
"User-Agent": "okhttp/3.10.0.1",
}
url = "https://is.snssdk.com/api/search/content/?os_api=23&device_type=oneplus%2Ba5010&from_search_subtab=video&manifest_version_code=7690&source=search_subtab_switch&offset=0&is_ttwebview=0&uuid=440000000189785&action_type&is_incognito=0&keyword_type&rom_version=23&app_name=news_article&format=json&version_name=7.6.9&ac=wifi&host_abi=armeabi-v7a&update_version_code=76909&channel=baidu_0411&is_native_req=1&loadId=1&longitude=113.40717530841052&isIncognito=0&plugin=2050&openudid=e44cc0264b92bcbf&forum=1&latitude=39.904680919672145&search_start_time=1587102733626&language=zh&pd=xiaoshipin&cur_tab_title=search_tab&aid=13&pos=5r_-9Onkv6e_eBEKeScxeCUfv7G_8fLz-vTp6Pn4v6esrKuzqa2qrKqorq2lqaytqK-xv_H86fTp6Pn4v6eupLOkramrpa2krKSrqq-sqaixv_zw_O3e9Onkv6e_eBEKeScxeCUfv7G__PD87dHy8_r06ej5-L-nrKyrs6mtqqyqqK6tpamsraivsb_88Pzt0fzp9Ono-fi_p66ks6StqaulraSspKuqr6ypqOA%253D&dpi=270&qrecImprId&fetch_by_ttnet=1&count=10&plugin_enable=3&search_position&ab_group=100167%252C94569%252C102754&keyword={0}&scm_version=1.0.2.830&search_json=%257B%2522comment_ids%2522%253A%255B%255D%252C%2522event_discussion%2522%253A74123%252C%2522event_impression%2522%253A17270790%252C%2522forum_id%2522%253A1664181806902302%252C%2522forum_recall_wtt%2522%253A%255B1664190666034183%252C1664192273575943%252C1664184430218253%252C1664185769175051%252C1664184985139212%252C1664196237152267%252C1664186792648732%252C1664188755414019%252C1664187055838215%252C1664184182571022%252C1664185938950148%252C1664188041995268%252C1664188322863172%252C1664190185024520%252C1664185602828300%252C1664184276484099%252C1664188211399684%252C1664187870713868%252C1664184484958211%252C1664183864289288%252C1664186825487371%252C1664195548700686%252C1664186585780228%252C1664197296210947%252C1664188146725901%252C1664191748459523%255D%252C%2522group_source%2522%253Anull%252C%2522hot_gid%2522%253A6816255461172445703%252C%2522log_pb%2522%253A%257B%2522cluster_type%2522%253A%25220%2522%252C%2522entrance_hotspot%2522%253A%2522channel%2522%252C%2522hot_board_cluster_id%2522%253A%25226816091697949180424%2522%252C%2522hot_board_impr_id%2522%253A%2522202004171352010100140411610B1A7741%2522%252C%2522location%2522%253A%2522hot_board%2522%252C%2522rank%2522%253A%25225%2522%252C%2522source%2522%253A%2522trending_tab%2522%252C%2522style_id%2522%253A%252210005%2522%257D%252C%2522mix_stick_ids%2522%253A%255B1664190666034183%252C1664192273575943%252C1664184430218253%252C1664185769175051%252C1664184985139212%252C1664196237152267%252C1664186792648732%252C1664188755414019%252C1664187055838215%252C1664184182571022%252C1664185938950148%252C1664188041995268%252C1664188322863172%252C1664190185024520%252C1664185602828300%252C1664184276484099%252C1664188211399684%252C1664187870713868%252C1664184484958211%252C1664183864289288%252C1664186825487371%252C1664195548700686%252C1664186585780228%252C1664197296210947%252C1664188146725901%252C1664191748459523%255D%252C%2522stick_group_ids%2522%253A%255B%255D%257D&device_platform=android&search_id&has_count=0&version_code=769&mac_address=08%253A00%253A27%253A1F%253A7E%253AA0&from=xiaoshipin&device_id={1}&resolution=810*1440&os_version=6.0.1&device_brand=Oneplus&search_sug=1&qc_query".format(
title, random.randint(69418800000, 69418899999))
res = retry_get_url(url, headers=headers, timeout=5, proxies=3)
page_text = res.json()
for one_video in page_text["data"]:
video_dic = {}
try:
one_video = one_video["raw_data"]
video_dic['title'] = one_video.get('title')
video_dic['url'] = one_video.get('share').get("share_url")
releaser_id = one_video.get('user').get("info").get("user_id")
video_dic['releaser'] = one_video.get('user').get("info").get(
"name")
video_dic[
'releaserUrl'] = "https://www.toutiao.com/c/user/%s/" % releaser_id
release_time = int(one_video.get('create_time'))
video_dic['release_time'] = int(release_time * 1e3)
video_dic['duration'] = int(
one_video.get('video').get("duration"))
video_dic['play_count'] = one_video.get('action').get(
"play_count")
video_dic['repost_count'] = one_video.get('action').get(
"share_count")
video_dic['comment_count'] = one_video.get('action').get(
"comment_count")
video_dic['favorite_count'] = one_video.get('action').get(
"digg_count")
video_dic['fetch_time'] = int(
datetime.datetime.now().timestamp() * 1e3)
video_dic['releaser_id_str'] = "toutiao_%s" % releaser_id
video_dic['video_img'] = one_video.get('video').get(
'origin_cover').get('url_list')[0]
video_dic['platform'] = "toutiao"
if "iesdouyin" in video_dic['url']:
video_dic[
'releaserUrl'] = "https://www.douyin.com/share/user/%s/" % releaser_id
video_dic['platform'] = "抖音"
video_dic['releaser_id_str'] = "抖音_%s" % releaser_id
video_dic['play_count'] = 0
video_dic["is_hot"] = 1
video_dic["data_provider"] = "CCR"
except:
continue
data_list.append(video_dic)
output_result(
result_Lst=data_list,
platform=self.platform,
output_to_es_raw=True,
)
data_list.clear()
def list_page(self,
rid,
page_num=1,
channel=None,
output_to_file=False,
filepath=None,
output_to_es_raw=False,
output_to_es_register=False,
push_to_redis=False,
page_num_max=34,
output_es_index=None,
output_doc_type=None,
proxy_dic=None):
result_lst = []
fail_time = 0
while page_num <= page_num_max and fail_time < 5:
lst_url = (
'https://api.bilibili.com/x/web-interface/newlist?rid=' + rid +
'&type=0&pn=' + str(page_num) + '&ps=20')
if proxy_dic is not None:
raw_proxy_dic = extract_data_to_use()
record_id = raw_proxy_dic['id']
proxy_dic = build_proxy_dic(raw_proxy_dic=raw_proxy_dic)
print('get proxy_dic %s' % proxy_dic)
try:
get_page = retry_get_url(lst_url,
proxies=proxy_dic,
timeout=15)
fail_time = 0
page_num += 1
except:
update_status(record_id=record_id, availability=0)
fail_time += 1
print('%s has failed %s times' % (lst_url, fail_time))
continue
print('get page at %s' % (page_num - 1))
page_dic = get_page.json()
total_video = int(page_dic['data']['page']['count'])
if page_num == 1:
if int(total_video / 20) == total_video / 20:
total_page_num = int(total_video / 20)
else:
total_page_num = int(total_video / 20) + 1
if total_page_num <= page_num_max:
page_num_max = total_page_num
video_dic = page_dic['data']['archives']
for one_video in video_dic:
video_dic = copy.deepcopy(self.video_data)
video_dic['title'] = one_video['title']
aid = one_video['aid']
video_dic['aid'] = one_video['aid']
try:
attribute = one_video['attribute']
except:
attribute = 0
video_dic['attribute'] = attribute
video_dic['url'] = 'https://www.bilibili.com/video/av' + str(
aid)
video_dic['releaser'] = one_video['owner']['name']
video_dic['releaser_id'] = one_video['owner']['mid']
video_dic['video_intro'] = one_video['desc']
video_dic['duration'] = one_video['duration']
video_dic['play_count'] = one_video['stat']['view']
video_dic['danmuku'] = one_video['stat']['danmaku']
video_dic['release_time'] = (one_video['pubdate']) * 1e3
fetch_time = int(
datetime.datetime.timestamp(datetime.datetime.now()) * 1e3)
video_dic['fetch_time'] = fetch_time
result_lst.append(video_dic)
if len(result_lst) >= 100:
if output_es_index is None and output_doc_type is None:
output_result(
result_Lst=result_lst,
platform=self.platform,
output_to_file=output_to_file,
filepath=filepath,
output_to_es_raw=output_to_es_raw,
output_to_es_register=output_to_es_register,
push_to_redis=push_to_redis)
result_lst.clear()
elif output_es_index is not None and output_doc_type is not None:
output_result(
result_Lst=result_lst,
platform=self.platform,
output_to_file=output_to_file,
filepath=filepath,
output_to_es_raw=output_to_es_raw,
output_to_es_register=output_to_es_register,
push_to_redis=push_to_redis,
es_index=output_es_index,
doc_type=output_doc_type)
result_lst.clear()
time.sleep(0)
if result_lst != []:
if output_es_index is None and output_doc_type is None:
output_result(result_Lst=result_lst,
platform=self.platform,
output_to_file=output_to_file,
filepath=filepath,
output_to_es_raw=output_to_es_raw,
output_to_es_register=output_to_es_register,
push_to_redis=push_to_redis)
elif output_es_index is not None and output_doc_type is not None:
output_result(result_Lst=result_lst,
platform=self.platform,
output_to_file=output_to_file,
filepath=filepath,
output_to_es_raw=output_to_es_raw,
output_to_es_register=output_to_es_register,
push_to_redis=push_to_redis,
es_index=output_es_index,
doc_type=output_doc_type)
return result_lst
def get_releaser_page(self, releaserUrl):
headers = {
"Accept-Encoding": "gzip",
# "sdk-version": "1",
"User-Agent":
"Linux; U; Android 8.1.0; zh-CN; EML-AL00 Build/HUAWEIEML-AL00",
"Host": "aweme.snssdk.com",
"Connection": "Keep-Alive",
}
releaser_id = self.get_releaser_id(releaserUrl)
time.sleep(1)
releaserUrl = 'https://{2}/aweme/v1/user/?ac=WIFI&device_id={1}&os_api=18&app_name=aweme&channel=App Store&device_platform=ipad&device_type=iPad6,11&app_version=8.9.1&js_sdk_version=1.17.2.0&version_code=8.7.1&os_version=13.2.3&screen_width=1536&user_id={0}'.format(
releaser_id, str(random.randint(40000000000, 90000000000)),
random.choice(self.api_list))
count = 0
while count < 3:
try:
count += 1
time.sleep(random.random())
get_page = retry_get_url(releaserUrl,
headers=headers,
proxies=0)
page = get_page.json()
follower_num = page["user"].get("follower_count")
age = page["user"].get('birthday')
location = page["user"].get('location')
province = page["user"].get('province')
huozan = page["user"].get("total_favorited")
sex = page["user"].get("gender")
verify = page["user"].get("enterprise_verify_reason")
signature = page["user"].get('signature').replace("\r",
"").replace(
"\n", "")
if page["user"].get('is_gov_media_vip'):
signature_type = "gov"
elif page["user"].get('is_star'):
signature_type = "star"
elif page["user"].get('is_effect_artist'):
signature_type = "artist"
elif page["user"].get('is_verified'):
signature_type = "已认证"
else:
signature_type = ""
dic = {
"age": age,
"location": location,
"huozan": huozan,
"sex": sex,
"verify": verify,
"signature": signature,
"province": province,
"follower_num": follower_num,
"signature_type": signature_type,
}
print(dic)
return dic
except:
print("can't find followers")
continue
else:
return None
def releaser_page(self,
releaserUrl,
output_to_file=False,
filepath=None,
output_to_es_raw=False,
output_to_es_register=False,
push_to_redis=False,
releaser_page_num_max=30,
es_index=None,
doc_type=None):
videos_per_page = 42
releaser_page_Lst = []
releaser_id = self.rebuild_releaserUrl(releaserUrl)
if releaser_id == '' or releaser_id is None:
print('Failed to get releaser id: %s' % releaserUrl)
return None
real_releaserUrl = 'https://www.iqiyi.com/u/' + releaser_id + '/v'
get_page = retry_get_url(real_releaserUrl)
if get_page is None:
print('Failed to get releaser page: %s' % releaserUrl)
return None
get_page.encoding = 'utf-8'
page = get_page.text
soup = BeautifulSoup(page, 'html.parser')
try:
videonum_str = soup.find('span', {'class': 'icon-num'}).text
videonum_f = re.findall('[0-9]+', videonum_str)
except:
print('Failed to get total video number: %s' % releaserUrl)
videonum_f = []
if videonum_f != []:
videonum = int(videonum_f[0])
totalpage = videonum // videos_per_page + 1
else:
videonum = None
totalpage = 1000 # assign an arbitary number
def process_one_line(data_line):
url = data_line.find('p', {
'class': 'site-piclist_info_title_twoline'
}).a['href']
if url[:6] != 'https:' or url[:5] != 'http:':
url = 'https:' + url
get_video_dict = self.video_page(url)
if get_video_dict is None:
return None
return get_video_dict
releaser_url_body_f = re.findall('https://www.iqiyi.com/u/[0-9]+/v',
releaserUrl)
if releaser_url_body_f != []:
releaser_url_body = releaser_url_body_f[0]
else:
releaser_url_body_f = re.findall('http://www.iqiyi.com/u/[0-9]+/v',
releaserUrl)
if releaser_url_body_f != []:
releaser_url_body = releaser_url_body_f[0]
else:
return None
if releaser_page_num_max > totalpage:
releaser_page_num_max = totalpage
else:
pass
video_page_url = [
releaser_url_body + '?page={}&video_type=1'.format(str(i))
for i in range(1, releaser_page_num_max + 1)
]
for urls in video_page_url:
get_page = retry_get_url(urls)
if get_page is None:
continue
print("get %s successfully" % urls)
page = get_page.text
soup = BeautifulSoup(page, 'html.parser')
iqiyi = soup.find_all('li', {'j-delegate': 'colitem'})
for data_line in iqiyi:
one_video_dic = process_one_line(data_line)
releaser_page_Lst.append(one_video_dic)
if len(releaser_page_Lst) >= 100:
output_result(releaser_page_Lst,
self.platform,
output_to_file=output_to_file,
filepath=filepath,
output_to_es_raw=output_to_es_raw,
output_to_es_register=output_to_es_register,
push_to_redis=push_to_redis,
es_index=es_index,
doc_type=doc_type)
releaser_page_Lst.clear()
if releaser_page_Lst != []:
output_result(releaser_page_Lst,
self.platform,
output_to_file=output_to_file,
filepath=filepath,
output_to_es_raw=output_to_es_raw,
output_to_es_register=output_to_es_register,
push_to_redis=push_to_redis,
es_index=es_index,
doc_type=doc_type)
def releaser_page(self, releaserUrl,
output_to_file=False,
filepath=None,
releaser_page_num_max=5000,
output_to_es_raw=False,
es_index=None,
doc_type=None,
output_to_es_register=False,
push_to_redis=False, proxies_num=None, **kwargs):
"""
get video info from api instead of web page html
the most scroll page is 1000
"""
result_list = []
has_more = True
count = 1
count_false = 0
releaser_id = self.find_releaser_id(releaserUrl)
offset = "0"
# vid = "AB5483CA-FCDC-42F1-AFB1-077A1%sDA" % random.randint(100000, 999999)
# ccid = "F153594D-1310-4984-A4C3-A679D4D%s" % random.randint(10000, 99999)
# openudid = "5d44f2ea1b74e3731b27e5ed8039ac29f%s" % random.randint(1000000, 9999999)
# idfa = "E3FC9054-384B-485F-9B4C-936F33D7D%s" % random.randint(100, 999)
# iid = str(random.randint(40000000000, 70000000000))
device_id = str(random.randint(56884000000, 96890000000))
# proxies = get_proxy(proxies_num)
while has_more and count <= releaser_page_num_max:
# print(str(releaser_id)+str(max_behot_time))
# js_head = json.loads(get_js(str(releaser_id)+str(max_behot_time)))
time.sleep(random.randint(1,2))
print("get %s video on page %s" % (releaser_id, count))
# url_dic = {
# "source": "0",
# "max_cursor": offset,
# "user_id": releaser_id,
# "count": "21",
# "os_api": "23",
# "device_type": "Huawei P20",
# "ssmix": "a",
# "manifest_version_code": "985",
# "dpi": "429",
# "uuid": "440000000189785",
# "app_name": "douyin",
# "version_name": "9.8.5",
# # "ts": "1585532172",
# "app_type": "normal",
# "ac": "wifi",
# "update_version_code": "9852",
# "channel": "baidu",
# # "_rticket": "1585532172572",
# "device_platform": "android",
# # "iid": "109688778422",
# "version_code": "985",
# # "cdid": "87cc1c77-cc3c-41a1-8df6-1e060b9c510b",
# # "openudid": "e44cc0264b92bcbf",
# "device_id": device_id,
# "resolution": "1080*2244",
# "os_version": "9.0.1",
# "language": "zh",
# "device_brand": "Huawei",
# "aid": "2329",
# "mcc_mnc": "46005",
# }
# url_dic = {
# "ac": "WIFI",
# # "iid": iid,
# "device_id": device_id,
# "os_api": "18",
# "app_name": "aweme",
# "channel": "App Store",
# # "idfa": "7AED33DD-0F97-418D-AFAA-72ED0578A44E",
# # "idfa": idfa,
# "device_platform": "iphone",
# "build_number": "92113",
# # "vid": "21B39A50-8C28-4E7E-AEB8-A67B12B1A82B",
# # "vid": vid,
# # "openudid": "b1021c76124449e0e9f0e43bdf51f3314aac263b",
# # "openudid": openudid,
# "device_type": "iPhone9,4",
# "app_version": "9.2.1",
# "js_sdk_version": "1.43.0.1",
# "version_code": "9.2.1",
# "os_version": "13.3",
# "screen_width": "1242",
# "aid": "2329",
# "mcc_mnc": "",
# "user_id": releaser_id,
# "max_cursor": offset,
# "count": "21",
# "source": "0",
# }
url_dic = {
"source": "0",
"max_cursor": offset,
"user_id": releaser_id,
"count": "21",
"os_api": "23",
"device_type": "Huawei P20",
"ssmix": "a",
"manifest_version_code": "10000%s" % random.randint(1,5),
"dpi": "429",
# "uuid": "440000000189785",
"app_name": "douyin_lite",
"version_name": "10.0.0",
# "ts": "1585532172",
"app_type": "normal",
"ac": "wifi",
"update_version_code": "10009900",
"channel": "baidu",
# "_rticket": "1585532172572",
"device_platform": "android",
# "iid": "1697284012668695",
"version_code": "100%s00" %random.randint(1,5),
# "cdid": "87cc1c77-cc3c-41a1-8df6-1e060b9c510b",
# "openudid": "e44cc0264b92bcbf",
"device_id": device_id,
# "device_id": 69418894872,
"resolution": "1080*2244",
"os_version": "9.0.1",
"language": "zh",
"device_brand": "Huawei",
"aid": "2329",
"mcc_mnc": "46001",
}
# self.headers["Host"] = host
url = "https://{1}/aweme/v1/aweme/post/?{0}".format(urllib.parse.urlencode(url_dic),random.choice(self.api_list))
try:
#proxies = get_proxy(proxies_num)
if proxies_num:
# get_page = requests.get(url, headers=self.headers, proxies=proxies, timeout=10)
get_page = retry_get_url(url, headers=self.headers, proxies=proxies_num, timeout=10)
else:
get_page = requests.get(url, headers=self.headers, timeout=10)
except Exception as e:
proxies = get_proxy(proxies_num)
print(e)
continue
page_dic = {}
# print(get_page.text)
try:
page_dic = get_page.json()
# print(get_page)
# print(page_dic)
data_list = page_dic.get('aweme_list')
if not data_list:
get_page = requests.get(url, headers=self.headers, timeout=10)
page_dic = get_page.json()
data_list = page_dic.get('aweme_list')
if not data_list:
raise ValueError
has_more = page_dic.get('has_more')
offset = str(page_dic.get("max_cursor"))
except:
if not data_list:
proxies = get_proxy(proxies_num)
count_false += 1
if count_false >= 2:
break
else:
continue
# offset = page_dic.get('offset')
if has_more is None:
has_more = False
if data_list == []:
print("no data in releaser %s page %s" % (releaser_id, count))
# print(page_dic)
# print(url)
proxies = get_proxy(1)
count_false += 1
if count_false >= 2:
has_more = False
continue
else:
count_false = 0
count += 1
for one_video in data_list:
# info_str = one_video.get('content')
video_dic = copy.deepcopy(self.video_data)
video_dic['title'] = one_video.get('desc')
video_dic['url'] = one_video.get('share_url')
video_dic['releaser'] = one_video.get('author').get("nickname")
video_dic['releaserUrl'] = releaserUrl
release_time = one_video.get('create_time')
video_dic['release_time'] = int(release_time * 1e3)
try:
video_dic['duration'] = int(one_video.get('duration') / 1000)
except:
video_dic['duration'] = 0
video_dic['play_count'] = 0
video_dic['repost_count'] = one_video.get('statistics').get('share_count')
video_dic['comment_count'] = one_video.get('statistics').get('comment_count')
video_dic['favorite_count'] = one_video.get('statistics').get('digg_count')
video_dic['video_id'] = one_video.get('aweme_id')
video_dic['fetch_time'] = int(datetime.datetime.now().timestamp() * 1e3)
video_dic['releaser_id_str'] = "抖音_%s" % releaser_id
try:
video_dic['video_img'] = one_video.get('video').get('cover').get('url_list')[0]
except:
pass
yield video_dic
def key_customer(self,
releaserUrl,
releaser_page_num_max=1000,
output_to_es_raw=False,
es_index='crawler-data-raw',
doc_type='doc'):
"""
input releaserUrl must be strict as https://id.tudou.com/i/UMjc5MzI5NDA==/videos?
end with /videos otherwise when scrolling it will make mistakes
"""
releaser_id = self.get_releaser_id(releaserUrl)
print("working on releaser: %s" % releaser_id)
releaserUrl = 'https://id.tudou.com/i/%s/videos' % releaser_id
result_lst = []
get_page = retry_get_url(releaserUrl)
get_page.encoding = 'utf-8'
page = get_page.text
soup = BeautifulSoup(page, 'html.parser')
try:
releaser = soup.find('div', {'class': 'user-name'}).a.text
except:
releaser = None
try:
total_video_num_str = soup.find('div', {
'class': 'title'
}).span.text
total_video_num = total_video_num_str.replace('(', '').replace(
')', '').replace(',', '')
total_video_num = trans_play_count(total_video_num)
except:
print(releaserUrl)
if total_video_num % 50 == 0:
total_page_num = int(total_video_num / 50)
else:
total_page_num = int(total_video_num / 50) + 1
if releaser_page_num_max > total_page_num:
releaser_page_num_max = total_page_num
print("releaser page num max is %s" % releaser_page_num_max)
video_lst = soup.find_all('div', {'class': 'v'})
for line in video_lst:
video_info = self.process_one_video(line)
video_info['releaserUrl'] = releaserUrl
video_info['releaser'] = releaser
result_lst.append(video_info)
if releaser_page_num_max >= 2:
page_num = 2
try:
partial_releaserUrl = soup.find('li', {
'class': 'next'
}).a['href']
new_releaserUrl = 'https://id.tudou.com%s' % partial_releaserUrl
except:
print(new_releaserUrl)
while page_num <= releaser_page_num_max:
get_page = retry_get_url(new_releaserUrl)
get_page.encoding = 'utf-8'
page = get_page.text
soup = BeautifulSoup(page, 'html.parser')
if page_num != releaser_page_num_max:
try:
new_releaserUrl = 'https://id.tudou.com' + soup.find(
'li', {
'class': 'next'
}).a['href']
except:
new_releaserUrl = (
'https://id.tudou.com/i/%s/videos?order=1&page=%s'
% (releaser_id, page_num))
video_lst = soup.find_all('div', {'class': 'v'})
for line in video_lst:
video_info = self.process_one_video(line)
video_info['releaserUrl'] = releaserUrl
video_info['releaser'] = releaser
result_lst.append(video_info)
print('get page %s list length is %s' %
(page_num, len(result_lst)))
page_num += 1
output_result(result_Lst=result_lst,
platform=self.platform,
output_to_es_raw=output_to_es_raw,
es_index=es_index,
doc_type=doc_type)
result_lst.clear()
if result_lst != []:
output_result(result_Lst=result_lst,
platform=self.platform,
output_to_es_raw=output_to_es_raw,
es_index=es_index,
doc_type=doc_type)
result_lst.clear()
def parse_video_page_single_process(self,
output_to_file=False,
filepath=None,
push_to_redis=False,
output_to_es_raw=True,
es_index="crawler-data-raw",
doc_type="doc",
output_to_es_register=False):
key = 'iqiyi_video_page_html'
result_list = []
pid = os.getpid()
while connect_with_redis.length_of_lst(key) > 0:
video_page_html = connect_with_redis.retrieve_video_page_html_from_redis(
platform=self.platform)
soup = BeautifulSoup(video_page_html, 'html.parser')
try:
page_info = soup.find("div", {"is": "i71-play"})[":page-info"]
page_info = page_info.replace("'", '"')
page_dic = json.loads(page_info)
except:
page_dic = None
if page_dic is not None:
title = page_dic["tvName"]
url = page_dic["pageUrl"]
dura_str = page_dic["duration"]
duration = trans_duration(dura_str)
try:
releaser = page_dic["user"]["name"]
releaserUrl = page_dic["user"]["profileUrl"]
except:
releaser = None
releaserUrl = None
else:
title = None
url = None
duration = None
releaser = None
releaserUrl = None
try:
video_info = soup.find("div",
{"is": "i71-play"})[":video-info"]
video_dic = json.loads(video_info)
except:
video_dic = None
if video_dic is not None:
if title is None:
title = video_dic['name']
if url is None:
url = video_dic['url']
if releaser is None:
try:
releaser = video_dic["user"]["name"]
releaserUrl = video_dic["user"]["profileUrl"]
except:
releaser = None
releaserUrl = None
release_time = video_dic["firstPublishTime"]
tvId = video_dic["tvId"]
hot_idx_url = "https://pub.m.iqiyi.com/jp/h5/count/hotDisplay/?qipuId=%s" % tvId
get_hot_idx = retry_get_url(hot_idx_url)
hot_idx_str = get_hot_idx.text
hot_idx = int(
re.findall(
"\d+", ' '.join(re.findall('"count":\d+',
hot_idx_str)))[0])
fetch_time = int(
datetime.datetime.timestamp(datetime.datetime.now()) * 1e3)
if releaser is None:
try:
releaser = soup.find('span', {
'class': 'intro-iterm__txt'
}).text
except:
releaser = None
video_page_dict = copy.deepcopy(self.video_data)
video_page_dict["title"] = title
video_page_dict["url"] = url
video_page_dict["duration"] = duration
video_page_dict["releaser"] = releaser
video_page_dict["releaserUrl"] = releaserUrl
video_page_dict["release_time"] = release_time
video_page_dict["hot_idx"] = hot_idx
video_page_dict["fetch_time"] = fetch_time
video_page_dict["tvId"] = tvId
result_list.append(video_page_dict)
print(
"platform: %s, action: parse video page, process_id: %s, has done: %s"
% (self.platform, pid, len(result_list)))
if len(result_list) >= 1000:
output_result(result_Lst=result_list,
platform=self.platform,
output_to_file=output_to_file,
filepath=filepath,
push_to_redis=push_to_redis,
output_to_es_raw=output_to_es_raw,
es_index=es_index,
doc_type=doc_type,
output_to_es_register=output_to_es_register)
result_list.clear()
if result_list != []:
output_result(result_Lst=result_list,
platform=self.platform,
output_to_file=output_to_file,
filepath=filepath,
push_to_redis=push_to_redis,
output_to_es_raw=output_to_es_raw,
es_index=es_index,
doc_type=doc_type,
output_to_es_register=output_to_es_register)
result_list.clear()
