def get_bangumi_title(media_id: str = "",
season_id: str = "",
episode_id: str = "") -> str:
if not (media_id or season_id or episode_id):
raise ArgumentsError("media_id", "season_id", "episode_id")
if media_id:
home_url = "https://www.bilibili.com/bangumi/media/md{media_id}".format(
media_id=media_id)
res = spider.get(home_url)
regex_title = re.compile(
r'<span class="media-info-title-t">(.*?)</span>')
if regex_title.search(res.text):
title = regex_title.search(res.text).group(1)
else:
title = "呐，我也不知道是什么标题呢～"
elif season_id or episode_id:
if season_id:
play_url = "https://www.bilibili.com/bangumi/play/ss{season_id}".format(
season_id=season_id)
else:
play_url = "https://www.bilibili.com/bangumi/play/ep{episode_id}".format(
episode_id=episode_id)
res = spider.get(play_url)
regex_title = re.compile(
r'<a href=".+" target="_blank" title="(.*?)" class="media-title">(?P<title>.*?)</a>'
)
if regex_title.search(res.text):
title = regex_title.search(res.text).group("title")
else:
title = "呐，我也不知道是什么标题呢～"
return title
def get_season_id(media_id: str) -> str:
home_url = "https://www.bilibili.com/bangumi/media/md{media_id}".format(
media_id=media_id)
season_id = ""
regex_season_id = re.compile(
r'"param":{"season_id":(\d+),"season_type":\d+}')
if regex_season_id.search(spider.get(home_url).text):
season_id = regex_season_id.search(spider.get(home_url).text).group(1)
return str(season_id)
def get_subtitle(avid: str = "", bvid: str = "", cid: str = ""):
if not (avid or bvid):
raise ArgumentsError("avid", "bvid")
subtitle_api = "https://api.bilibili.com/x/player.so?id=cid:{cid}&aid={avid}&bvid={bvid}"
subtitle_url = subtitle_api.format(avid=avid, cid=cid, bvid=bvid)
res = spider.get(subtitle_url)
subtitles_info = json.loads(re.search(r"<subtitle>(.+)</subtitle>", res.text).group(1))
return [
# fmt: off
{
"lang": sub_info["lan_doc"],
"lines": spider.get("https:" + sub_info["subtitle_url"]).json()["body"]
}
for sub_info in subtitles_info["subtitles"]
]
def get_containers(context, video_dir, format, playlist=None):
season_id = context['season_id']
containers = []
info_url = info_api.format(season_id=season_id)
res = spider.get(info_url)
for i, item in enumerate(res.json()["result"]["main_section"]["episodes"]):
index = item["title"]
if re.match(r'^\d*\.?\d*$', index):
index = '第{}话'.format(index)
name = repair_filename(' '.join([index, item["long_title"]]))
file_path = os.path.join(video_dir, repair_filename(
'{}.mp4'.format(name)))
if playlist is not None:
playlist.write_path(file_path)
containers.append(BililiContainer(
id=i+1,
name=name,
path=file_path,
meta={
"aid": item["aid"],
"cid": item["cid"],
"epid": item["id"],
"bvid": ''
},
format=format,
))
if playlist is not None:
playlist.flush()
return containers
def get_bangumi_subtitle(avid: str = "", bvid: str = "", cid: str = ""):
if not (avid or bvid):
raise ArgumentsError("avid", "bvid")
subtitle_api = (
"https://api.bilibili.com/x/player/v2?cid={cid}&aid={avid}&bvid={bvid}"
)
subtitle_url = subtitle_api.format(avid=avid, bvid=bvid, cid=cid)
subtitles_info = spider.get(subtitle_url).json()["data"]["subtitle"]
return [
# fmt: off
{
"lang": sub_info["lan_doc"],
"lines":
spider.get("https:" + sub_info["subtitle_url"]).json()["body"]
} for sub_info in subtitles_info["subtitles"]
]
def get_subtitle(container):
cid, avid, bvid = container.meta["cid"], container.meta[
"avid"], container.meta["bvid"]
# 检查是否有字幕并下载
subtitle_url = subtitle_api.format(avid=avid, cid=cid, bvid=bvid)
res = spider.get(subtitle_url)
subtitles_info = json.loads(
re.search(r"<subtitle>(.+)</subtitle>", res.text).group(1))
for sub_info in subtitles_info["subtitles"]:
sub_path = os.path.splitext(
container.path)[0] + sub_info["lan_doc"] + ".srt"
subtitle = Subtitle(sub_path)
for sub_line in spider.get("https:" +
sub_info["subtitle_url"]).json()["body"]:
subtitle.write_line(sub_line["content"], sub_line["from"],
sub_line["to"])
def get_containers(context, video_dir, format, playlist=None):
avid, bvid = context['avid'], context['bvid']
containers = []
info_url = info_api.format(avid=avid, bvid=bvid)
res = spider.get(info_url)
for i, item in enumerate(res.json()["data"]):
file_path = os.path.join(video_dir, repair_filename(
'{}.mp4'.format(item["part"])))
if playlist is not None:
playlist.write_path(file_path)
containers.append(BililiContainer(
id=i+1,
name=item["part"],
path=file_path,
meta={
"avid": avid,
"bvid": bvid,
"cid": item["cid"]
},
format=format,
))
if playlist is not None:
playlist.flush()
return containers
def get_danmaku(container):
# 下载弹幕
danmaku_url = danmaku_api.format(cid=container.meta['cid'])
res = spider.get(danmaku_url)
res.encoding = "utf-8"
danmaku_path = os.path.splitext(container.path)[0] + ".xml"
with open(danmaku_path, "w", encoding="utf-8") as f:
f.write(res.text)
def get_context(home_url):
context = {
'season_id': '',
}
season_id = re.search(
r'"param":{"season_id":(\d+),"season_type":\d+}', spider.get(home_url).text).group(1)
context['season_id'] = season_id
return context
def get_acg_video_list(avid: str = "", bvid: str = ""):
if not (avid or bvid):
raise ArgumentsError("avid", "bvid")
list_api = "https://api.bilibili.com/x/player/pagelist?aid={avid}&bvid={bvid}&jsonp=jsonp"
res = spider.get(list_api.format(avid=avid, bvid=bvid))
return [
# fmt: off
{
'id': i + 1,
'name': item['part'],
'cid': str(item['cid'])
} for i, item in enumerate(res.json()['data'])
]
def get_acg_video_title(avid: str = "", bvid: str = "") -> str:
if not (avid or bvid):
raise ArgumentsError("avid", "bvid")
home_url = ("https://www.bilibili.com/video/{bvid}".format(bvid=bvid)
if bvid else "https://www.bilibili.com/video/av{avid}".format(
avid=avid))
res = spider.get(home_url)
regex_title = re.compile(
r"<title .*>(.*)_哔哩哔哩 \(゜-゜\)つロ 干杯~-bilibili</title>")
if regex_title.search(res.text):
title = regex_title.search(res.text).group(1)
else:
title = "呐，我也不知道是什么标题呢～"
return title
def get_video_info(avid: str = "", bvid: str = ""):
if not (avid or bvid):
raise ArgumentsError("avid", "bvid")
info_api = "http://api.bilibili.com/x/web-interface/view?aid={avid}&bvid={bvid}"
res = spider.get(info_api.format(avid=avid, bvid=bvid))
res_json_data = res.json()["data"]
episode_id = ""
if res_json_data.get("redirect_url") and regex_bangumi_ep.match(res_json_data["redirect_url"]):
episode_id = regex_bangumi_ep.match(res_json_data["redirect_url"]).group("episode_id")
return {
"avid": str(res_json_data["aid"]),
"bvid": res_json_data["bvid"],
"picture": res_json_data["pic"],
"episode_id": episode_id,
}
def get_bangumi_list(episode_id: str = "", season_id: str = ""):
if not (season_id or episode_id):
raise ArgumentsError("season_id", "episode_id")
list_api = "http://api.bilibili.com/pgc/view/web/season?season_id={season_id}&ep_id={episode_id}"
res = spider.get(
list_api.format(episode_id=episode_id, season_id=season_id))
return [{
"id":
i + 1,
"name":
" ".join([
"第{}话".format(item["title"])
if re.match(r"^\d*\.?\d*$", item["title"]) else item["title"],
item["long_title"],
]),
"cid":
str(item["cid"]),
"episode_id":
str(item["id"]),
"avid":
str(item["aid"]),
"bvid":
item["bvid"],
} for i, item in enumerate(res.json()["result"]["episodes"])]
def get_bangumi_playurl(
avid: str = "",
bvid: str = "",
episode_id: str = "",
cid: str = "",
quality: int = 120,
audio_quality: int = 30280,
type: str = "dash",
):
video_quality_sequence = gen_quality_sequence(quality, type=Media.VIDEO)
audio_quality_sequence = gen_quality_sequence(audio_quality,
type=Media.AUDIO)
play_api = "https://api.bilibili.com/pgc/player/web/playurl?avid={avid}&bvid={bvid}&ep_id={episode_id}&cid={cid}&qn={quality}"
if type == "flv":
touch_message = spider.get(
play_api.format(avid=avid,
bvid=bvid,
episode_id=episode_id,
cid=cid,
quality=80)).json()
if touch_message["code"] != 0:
raise CannotDownloadError(touch_message["code"],
touch_message["message"])
if touch_message["result"]["is_preview"] == 1:
raise IsPreviewError()
accept_quality = touch_message["result"]["accept_quality"]
for quality in video_quality_sequence:
if quality in accept_quality:
break
play_url = play_api.format(avid=avid,
bvid=bvid,
episode_id=episode_id,
cid=cid,
quality=quality)
res = spider.get(play_url)
return [{
"id": i + 1,
"url": segment["url"],
"mirrors": segment["backup_url"],
"quality": quality,
"height": video_quality_map[quality]["height"],
"width": video_quality_map[quality]["width"],
"size": segment["size"],
"type": "flv_segment",
} for i, segment in enumerate(res.json()["result"]["durl"])]
elif type == "dash":
result = []
play_api_dash = play_api + "&fnver=0&fnval=16&fourk=1"
play_info = spider.get(
play_api_dash.format(avid=avid,
bvid=bvid,
episode_id=episode_id,
cid=cid,
quality=video_quality_sequence[0])).json()
if play_info["code"] != 0:
raise CannotDownloadError(play_info["code"], play_info["message"])
if play_info["result"].get("dash") is None:
raise UnsupportTypeError("dash")
if play_info["result"]["is_preview"] == 1:
raise IsPreviewError()
accept_video_quality = set(
[video["id"] for video in play_info["result"]["dash"]["video"]])
for video_quality in video_quality_sequence:
if video_quality in accept_video_quality:
break
else:
video_quality = 120
accept_audio_quality = set(
[audio["id"] for audio in play_info["result"]["dash"]["audio"]])
for audio_quality in audio_quality_sequence:
if audio_quality in accept_audio_quality:
break
else:
audio_quality = 30280
if play_info["result"]["dash"]["video"]:
videos = play_info["result"]["dash"]["video"]
for video in videos:
if video["id"] == video_quality:
result.append({
"id":
1,
"url":
video["base_url"],
"mirrors":
video["backup_url"],
"quality":
quality,
"height":
video["height"],
"width":
video["width"],
"size":
touch_url(video["base_url"], spider)[0],
"type":
"dash_video",
})
break
if play_info["result"]["dash"]["audio"]:
audios = play_info["result"]["dash"]["audio"]
for audio in audios:
if audio["id"] == audio_quality:
result.append({
"id":
2,
"url":
audio["base_url"],
"mirrors":
audio["backup_url"],
"quality":
audio_quality,
"height":
None,
"width":
None,
"size":
touch_url(audio["base_url"], spider)[0],
"type":
"dash_audio",
})
break
return result
elif type == "mp4":
raise UnsupportTypeError("mp4")
else:
raise UnknownTypeError()
def get_danmaku(cid: str) -> str:
danmaku_api = "http://comment.bilibili.com/{cid}.xml"
res = spider.get(danmaku_api.format(cid=cid))
res.encoding = "utf-8"
return res.text
def get_title(home_url):
""" 获取视频标题 """
res = spider.get(home_url)
title = re.search(
r'<span class="media-info-title-t">(.*?)</span>', res.text).group(1)
return title
def parse_segments(container, quality_sequence, block_size):
cid, avid, bvid = container.meta["cid"], container.meta["avid"], container.meta["bvid"]
if container.format == "flv":
# 检查是否可以下载，同时搜索支持的清晰度，并匹配最佳清晰度
touch_message = spider.get(parse_api.format(
avid=avid, cid=cid, bvid=bvid, qn=80)).json()
if touch_message["code"] != 0:
print("warn: 无法下载 {} ，原因： {}".format(
container.name, touch_message["message"]))
return
accept_quality = touch_message['data']['accept_quality']
for qn in quality_sequence:
if qn in accept_quality:
break
parse_url = parse_api.format(avid=avid, cid=cid, bvid=bvid, qn=qn)
res = spider.get(parse_url)
for i, segment in enumerate(res.json()['data']['durl']):
container.append_media(
id=i+1,
url=segment["url"],
qn=qn,
height=quality_map[qn]['height'],
width=quality_map[qn]['width'],
size=segment["size"],
type="segment",
block_size=block_size,
)
elif container.format == "m4s":
# 检查是否可以下载，同时搜索支持的清晰度，并匹配最佳清晰度
parse_api_m4s = parse_api + "&fnver=0&fnval=16&fourk=1"
play_info = spider.get(parse_api_m4s.format(
avid=avid, cid=cid, bvid=bvid, qn=quality_sequence[0])).json()
if play_info["code"] != 0:
print("warn: 无法下载 {} ，原因： {}".format(
container.name, play_info["message"]))
return
if play_info['data'].get('dash') is None:
raise Exception('该视频尚不支持 M4S format 哦~')
# accept_quality = play_info['data']['accept_quality']
accept_quality = set([video['id']
for video in play_info['data']['dash']['video']])
for qn in quality_sequence:
if qn in accept_quality:
break
parse_url = parse_api_m4s.format(avid=avid, cid=cid, bvid=bvid, qn=qn)
play_info = spider.get(parse_url).json()
for video in play_info['data']['dash']['video']:
if video['id'] == qn:
container.append_media(
id=1,
url=video['base_url'],
qn=qn,
height=video['height'],
width=video['width'],
size=touch_url(video['base_url'], spider)[0],
type="video",
block_size=block_size,
)
break
for audio in play_info['data']['dash']['audio']:
container.append_media(
id=2,
url=audio['base_url'],
height=None,
width=None,
size=touch_url(audio['base_url'], spider)[0],
qn=qn,
type="audio",
block_size=block_size,
)
break
elif container.format == 'mp4':
# 检查是否可以下载，同时搜索支持的清晰度，并匹配最佳清晰度
parse_api_mp4 = parse_api + "&platform=html5&high_quality=1"
play_info = spider.get(parse_api_mp4.format(
avid=avid, cid=cid, bvid=bvid, qn=120)).json()
if play_info["code"] != 0:
print("warn: 无法下载 {} ，原因： {}".format(
container.name, play_info["message"]))
return
container.append_media(
id=1,
url=play_info['data']['durl'][0]['url'],
qn=play_info['data']['quality'],
height=quality_map[play_info['data']['quality']]['height'],
width=quality_map[play_info['data']['quality']]['width'],
size=play_info['data']['durl'][0]['size'],
type="container",
block_size=block_size,
)
else:
print("Unknown format {}".format(container.format))
def get_title(home_url):
res = spider.get(home_url)
title = re.search(
r'<title .*>(.*)_哔哩哔哩 \(゜-゜\)つロ 干杯~-bilibili</title>', res.text).group(1)
return title
def get_acg_video_playurl(
avid: str = "",
bvid: str = "",
cid: str = "",
quality: int = 120,
audio_quality: int = 30280,
type: str = "dash",
):
if not (avid or bvid):
raise ArgumentsError("avid", "bvid")
video_quality_sequence = gen_quality_sequence(quality, type=Media.VIDEO)
audio_quality_sequence = gen_quality_sequence(audio_quality,
type=Media.AUDIO)
play_api = "https://api.bilibili.com/x/player/playurl?avid={avid}&bvid={bvid}&cid={cid}&qn={quality}&type=&otype=json"
if type == "flv":
touch_message = spider.get(
play_api.format(avid=avid, bvid=bvid, cid=cid, quality=80)).json()
if touch_message["code"] != 0:
raise CannotDownloadError(touch_message["code"],
touch_message["message"])
accept_quality = touch_message["data"]["accept_quality"]
for quality in video_quality_sequence:
if quality in accept_quality:
break
play_url = play_api.format(avid=avid,
bvid=bvid,
cid=cid,
quality=quality)
res = spider.get(play_url)
return [{
"id": i + 1,
"url": segment["url"],
"mirrors": segment["backup_url"],
"quality": quality,
"height": video_quality_map[quality]["height"],
"width": video_quality_map[quality]["width"],
"size": segment["size"],
"type": "flv_segment",
} for i, segment in enumerate(res.json()["data"]["durl"])]
elif type == "dash":
result = []
play_api_dash = play_api + "&fnver=0&fnval=16&fourk=1"
touch_message = spider.get(
play_api_dash.format(avid=avid,
bvid=bvid,
cid=cid,
quality=video_quality_sequence[0])).json()
if touch_message["code"] != 0:
raise CannotDownloadError(touch_message["code"],
touch_message["message"])
if touch_message["data"].get("dash") is None:
raise UnsupportTypeError("dash")
video_accept_quality = set(
[video["id"] for video in touch_message["data"]["dash"]["video"]])
for video_quality in video_quality_sequence:
if video_quality in video_accept_quality:
break
else:
video_quality = 120
audio_accept_quality = set(
[audio["id"] for audio in touch_message["data"]["dash"]["audio"]])
for audio_quality in audio_quality_sequence:
if audio_quality in audio_accept_quality:
break
else:
audio_quality = 30280
res = spider.get(
play_api_dash.format(avid=avid,
bvid=bvid,
cid=cid,
quality=quality))
if res.json()["data"]["dash"]["video"]:
videos = res.json()["data"]["dash"]["video"]
for video in videos:
if video["id"] == video_quality:
result.append({
"id":
1,
"url":
video["base_url"],
"mirrors":
video["backup_url"],
"quality":
video_quality,
"height":
video["height"],
"width":
video["width"],
"size":
touch_url(video["base_url"], spider)[0],
"type":
"dash_video",
})
break
if res.json()["data"]["dash"]["audio"]:
audios = res.json()["data"]["dash"]["audio"]
for audio in audios:
if audio["id"] == audio_quality:
result.append({
"id":
2,
"url":
audio["base_url"],
"mirrors":
audio["backup_url"],
"quality":
audio_quality,
"height":
None,
"width":
None,
"size":
touch_url(audio["base_url"], spider)[0],
"type":
"dash_audio",
})
break
return result
elif type == "mp4":
play_api_mp4 = play_api + "&platform=html5&high_quality=1"
play_info = spider.get(
play_api_mp4.format(avid=avid, bvid=bvid, cid=cid,
quality=120)).json()
if play_info["code"] != 0:
raise CannotDownloadError(play_info["code"], play_info["message"])
return [{
"id":
1,
"url":
play_info["data"]["durl"][0]["url"],
"mirrors": [],
"quality":
play_info["data"]["quality"],
"height":
video_quality_map[play_info["data"]["quality"]]["height"],
"width":
video_quality_map[play_info["data"]["quality"]]["width"],
"size":
play_info["data"]["durl"][0]["size"],
"type":
"mp4_container",
}]
else:
raise UnknownTypeError(type)
def parse_segments(container, quality_sequence, block_size):
aid, cid, ep_id, bvid = container.meta["aid"], container.meta[
"cid"], container.meta["epid"], container.meta["bvid"]
if container.format == "flv":
# 检查是否可以下载，同时搜索支持的清晰度，并匹配最佳清晰度
touch_message = spider.get(parse_api.format(
avid=aid, cid=cid, ep_id=ep_id, qn=80)).json()
if touch_message["code"] != 0:
print("warn: 无法下载 {} ，原因： {}".format(
container.name, touch_message["message"]))
return
if touch_message["result"]["is_preview"] == 1:
print("warn: {} 为预览版视频".format(container.name))
accept_quality = touch_message['result']['accept_quality']
for qn in quality_sequence:
if qn in accept_quality:
break
parse_url = parse_api.format(avid=aid, cid=cid, ep_id=ep_id, qn=qn)
res = spider.get(parse_url)
for i, segment in enumerate(res.json()['result']['durl']):
container.append_media(
id=i+1,
url=segment["url"],
qn=qn,
height=quality_map[qn]['height'],
width=quality_map[qn]['width'],
size=segment["size"],
type="segment",
block_size=block_size,
)
elif container.format == "m4s":
# 检查是否可以下载，同时搜索支持的清晰度，并匹配最佳清晰度
parse_api_m4s = parse_api + "&fnver=0&fnval=16&fourk=1"
play_info = spider.get(parse_api_m4s.format(
avid=aid, cid=cid, ep_id=ep_id, qn=quality_sequence[0], bvid=bvid)).json()
if play_info["code"] != 0:
print("warn: 无法下载 {} ，原因： {}".format(
container.name, play_info["message"]))
return
if play_info["result"]["is_preview"] == 1:
print("warn: {} 为预览版视频".format(container.name))
# accept_quality = play_info['result']['accept_quality']
accept_quality = set([video['id']
for video in play_info['result']['dash']['video']])
for qn in quality_sequence:
if qn in accept_quality:
break
for video in play_info['result']['dash']['video']:
if video['id'] == qn:
container.append_media(
id=1,
url=video['base_url'],
qn=qn,
height=video['height'],
width=video['width'],
size=touch_url(video['base_url'], spider)[0],
type="video",
block_size=block_size,
)
break
for audio in play_info['result']['dash']['audio']:
container.append_media(
id=2,
url=audio['base_url'],
height=None,
width=None,
size=touch_url(audio['base_url'], spider)[0],
qn=qn,
type="audio",
block_size=block_size,
)
break
elif container.format == 'mp4':
print("番剧不支持 mp4 format")
else:
print("Unknown format {}".format(container.format))
