def download_video(self, aweme_id):
query_params = self.common_params
query_params['aweme_id'] = aweme_id
params = {**query_params, **self.common_params}
url = URL.video_detail_url()
real_url = gen_url(self.token, url, params)
post_data = {
"aweme_id": aweme_id
}
cookies = self.__COOKIES
cookies['install_id'] = str(self.__device["install_id"])
resp = fetch(real_url,
data=post_data,
cookies=cookies,
headers=self.__HEADERS,
timeout=3).json()
try:
play_addr_raw = resp['aweme_detail']['video']['play_addr']['url_list']
play_addr = play_addr_raw[0]
print(play_addr)
content = fetch(play_addr).content
except:
self.logger.warning("提取视频信息失败...")
content = None
return content
def gen_url(token, raw_url, query):
if isinstance(query, dict):
query = params2str(query)
url = raw_url + "?" + query
resp = requests.post(URL.api_sign(token), json={"url": url}).json()
real_url = resp['url']
return real_url
def download_video(self, aweme_id):
query_params = self.common_params
query_params['aweme_id'] = aweme_id
sign = getSign(self.__get_token(), query_params)
params = {**query_params, **sign}
post_data = {"aweme_id": aweme_id}
resp = requests.get(URL.video_detail_url(),
params=params,
data=post_data,
verify=False,
headers=self.__HEADERS)
resp_result = resp.json()
# print(resp_result)
try:
play_addr_raw = resp_result['aweme_detail']['video']['play_addr'][
'url_list']
play_addr = play_addr_raw[0]
content = requests.get(play_addr).content
except:
self.logger.warning("提取视频信息失败...")
content = None
return content
def get_device(token):
device_info = requests.get(URL.api_device(token), timeout=10)
device_info_json = device_info.json()
return device_info_json
def gen_url(token, raw_url, query):
if isinstance(query, dict):
query = params2str(query)
url = raw_url + "?" + query
resp = requests.post(URL.api_sign(token), json={"url": url})
if resp.status_code != 200:
print("您当日的 api 次数已经使用完毕, 请明日再来爬取吧...")
return
real_url = resp.json()['url']
return real_url
def gen_real_url(token, raw_url, query):
if isinstance(query, dict):
query = params2str(query)
url = raw_url + "?" + query
resp = fetch(URL.api_sign(token), json={"url": url}, method="post")
if not resp:
print("你的当日 token 次数已经用完，请明天再来尝试吧...")
raise BaseException("you have run out of token.Please try tomorrow")
resp_json = resp.json()
real_url = resp_json['url']
return real_url
def grab_video(self, user_id, action, content, max_cursor=0):
url = URL.favorite_url() if action == "USER_LIKE" else URL.post_url()
favorite_params = copy.deepcopy(self.__USER_VIDEO_PARAMS)
favorite_params['user_id'] = user_id
favorite_params['max_cursor'] = max_cursor
query_params = {**favorite_params, **self.common_params}
real_url = gen_url(self.token, url, query_params)
# 目前支持两种类型爬取，用户喜欢过的，和当前用户所有已发布的视频
cookies = self.__COOKIES
cookies['install_id'] = str(self.__device["install_id"])
resp = requests.get(real_url,
verify=False,
cookies=cookies,
headers={"User-Agent": "okhttp/3.10.0.1"},
timeout=3)
favorite_info = json.loads(resp.content.decode("utf-8"))
hasmore = favorite_info.get('has_more')
max_cursor = favorite_info.get('max_cursor')
video_infos = favorite_info.get('aweme_list')
for per_video in video_infos:
author_nick_name = per_video['author'].get("nickname")
author_uid = per_video['author'].get('uid')
video_desc = per_video.get('desc')
music_id = per_video['music']['play_url'].get('uri') if content == "-m" else None
download_item = {
"author_nick_name": author_nick_name,
"video_desc": video_desc,
"author_uid": author_uid,
"music_id": music_id
}
aweme_id = per_video.get("aweme_id")
self.download_user_video(aweme_id, **download_item)
time.sleep(5)
return hasmore, max_cursor
def download_music(self, music_id):
if not music_id:
return
url = URL.music_url(music_id)
resp = requests.get(url, headers=self.__HEADERS, verify=False)
music_content = resp.content
return music_content
def grab_video(self, user_id, action, content, max_cursor=0):
favorite_params = copy.deepcopy(self.__USER_VIDEO_PARAMS)
favorite_params['user_id'] = user_id
favorite_params['max_cursor'] = max_cursor
query_params = {**favorite_params, **self.common_params}
sign = getSign(self.__get_token(), query_params)
params = {**query_params, **sign}
# 目前支持两种类型爬取，用户喜欢过的，和当前用户所有已发布的视频
url = URL.favorite_url() if action == "USER_LIKE" else URL.post_url()
resp = requests.get(url,
params=params,
verify=False,
headers=self.__HEADERS)
favorite_info = resp.json()
hasmore = favorite_info.get('has_more')
max_cursor = favorite_info.get('max_cursor')
video_infos = favorite_info.get('aweme_list')
for per_video in video_infos:
author_nick_name = per_video['author'].get("nickname")
author_uid = per_video['author'].get('uid')
video_desc = per_video.get('desc')
music_id = per_video['music']['play_url'].get(
'uri') if content == "-m" else None
download_item = {
"author_nick_name": author_nick_name,
"video_desc": video_desc,
"author_uid": author_uid,
"music_id": music_id
}
aweme_id = per_video.get("aweme_id")
self.download_user_video(aweme_id, **download_item)
time.sleep(5)
return hasmore, max_cursor
def __grab_comment(self, aweme_id, cursor, upvote_bound=10):
url = URL.comment_url()
comment_params = copy.deepcopy(self.__COMMENT_LIST_PARAMS)
comment_params['aweme_id'] = aweme_id
comment_params['cursor'] = cursor
params = {**comment_params, **self.common_params}
real_url = gen_url(self.token, url, params)
cookies = self.__COOKIES
cookies['install_id'] = str(self.__device["install_id"])
resp = requests.get(real_url,
verify=False,
cookies=cookies,
headers=self.__HEADERS)
comment_content = json.loads(resp.content.decode("utf-8"))
comments = comment_content.get("comments")
for per_comment in comments:
is_reply = per_comment.get("reply_comment")
if is_reply:
upvote_count = is_reply[0].get("digg_count")
comment_info = {
"text": is_reply[0].get("text"),
"upvote_count": upvote_count,
"nick_name": is_reply[0]['user'].get("nickname"),
"user_id": is_reply[0]['user'].get("uid"),
}
else:
upvote_count = per_comment.get("digg_count")
comment_info = {
"text": per_comment.get("text"),
"upvote_count": upvote_count,
"nick_name": per_comment['user'].get("nickname"),
"user_id": per_comment['user'].get("uid"),
}
if int(upvote_count) < upvote_bound:
return -2
self.download_comment(aweme_id, **comment_info)
# print(text,upvote_count,nick_name,user_id)
hasmore = int(comment_content.get("hasmore"))
return hasmore
def __grab_comment(self, aweme_id, cursor, upvote_bound=10):
comment_params = copy.deepcopy(self.__COMMENT_LIST_PARAMS)
comment_params['aweme_id'] = aweme_id
comment_params['cursor'] = cursor
query_params = {**comment_params, **self.common_params}
sign = getSign(self.__get_token(), query_params)
params = {**query_params, **sign}
resp = requests.get(URL.comment_url(),
params=params,
verify=False,
headers=self.__HEADERS)
comment_content = resp.json()
comments = comment_content.get("comments")
for per_comment in comments:
is_reply = per_comment.get("reply_comment")
if is_reply:
upvote_count = is_reply[0].get("digg_count")
comment_info = {
"text": is_reply[0].get("text"),
"upvote_count": upvote_count,
"nick_name": is_reply[0]['user'].get("nickname"),
"user_id": is_reply[0]['user'].get("uid"),
}
else:
upvote_count = per_comment.get("digg_count")
comment_info = {
"text": per_comment.get("text"),
"upvote_count": upvote_count,
"nick_name": per_comment['user'].get("nickname"),
"user_id": per_comment['user'].get("uid"),
}
print(upvote_count)
if int(upvote_count) < upvote_bound:
return -2
self.download_comment(aweme_id, **comment_info)
# print(text,upvote_count,nick_name,user_id)
hasmore = int(comment_content.get("hasmore"))
return hasmore
def like_video(self, aweme_id):
query_params = {**{"pass-region": "1"}, **self.common_params}
sign = getSign(self.__get_token(), query_params)
params = {**query_params, **sign}
form_params = {"aweme_id": aweme_id, "type": 1}
headers = copy.deepcopy(self.__HEADERS)
headers["sdk-version"] = '1'
headers["Accept-Encoding"] = 'br, gzip, deflate'
print(self.__request.cookies)
result = self.__request.post(URL.like_video_url(),
params=params,
data=form_params,
verify=False,
headers=headers)
print(result.json())
def gen_device(token):
resp = fetch(URL.api_device(token), timeout=10).json()
return resp
