mirror of
https://git.sr.ht/~cadence/NewLeaf
synced 2024-11-25 08:57:30 +00:00
Add DASH manifest endpoint
This commit is contained in:
parent
cf37afec2b
commit
831d67e990
@ -13,6 +13,7 @@ These endpoints are somewhat implemented:
|
|||||||
- `/api/v1/channels/{part}/{ucid}`
|
- `/api/v1/channels/{part}/{ucid}`
|
||||||
- `/api/v1/search?q={search}`
|
- `/api/v1/search?q={search}`
|
||||||
- `/vi/{id}/{file}`
|
- `/vi/{id}/{file}`
|
||||||
|
- `/api/manifest/dash/id/{id}`
|
||||||
|
|
||||||
## The future
|
## The future
|
||||||
|
|
||||||
|
109
index.py
109
index.py
@ -7,6 +7,8 @@ import re
|
|||||||
import json
|
import json
|
||||||
import traceback
|
import traceback
|
||||||
import requests
|
import requests
|
||||||
|
import xml.etree.ElementTree as ET
|
||||||
|
from cachetools import TTLCache
|
||||||
|
|
||||||
ytdl_opts = {
|
ytdl_opts = {
|
||||||
"quiet": True,
|
"quiet": True,
|
||||||
@ -25,7 +27,14 @@ def length_text_to_seconds(text):
|
|||||||
return sum([int(x) * 60**(len(s)-i-1) for i, x in enumerate(s)])
|
return sum([int(x) * 60**(len(s)-i-1) for i, x in enumerate(s)])
|
||||||
|
|
||||||
class Second(object):
|
class Second(object):
|
||||||
|
def __init__(self):
|
||||||
|
self.video_cache = TTLCache(maxsize=50, ttl=300)
|
||||||
|
|
||||||
def _cp_dispatch(self, vpath):
|
def _cp_dispatch(self, vpath):
|
||||||
|
if vpath[:4] == ["api", "manifest", "dash", "id"]:
|
||||||
|
vpath[:4] = ["manifest"]
|
||||||
|
return self
|
||||||
|
|
||||||
if vpath[:2] == ["api", "v1"]:
|
if vpath[:2] == ["api", "v1"]:
|
||||||
endpoints = [
|
endpoints = [
|
||||||
["channels", 1, 2],
|
["channels", 1, 2],
|
||||||
@ -42,37 +51,45 @@ class Second(object):
|
|||||||
@cherrypy.expose
|
@cherrypy.expose
|
||||||
@cherrypy.tools.json_out()
|
@cherrypy.tools.json_out()
|
||||||
def videos(self, id):
|
def videos(self, id):
|
||||||
|
if id in self.video_cache:
|
||||||
|
return self.video_cache[id]
|
||||||
|
|
||||||
try:
|
try:
|
||||||
info = ytdl_save.extract_info(id, download=False)
|
info = ytdl_save.extract_info(id, download=False)
|
||||||
|
|
||||||
year = int(info["upload_date"][:4])
|
year = int(info["upload_date"][:4])
|
||||||
month = int(info["upload_date"][4:6])
|
month = int(info["upload_date"][4:6])
|
||||||
day = int(info["upload_date"][6:8])
|
day = int(info["upload_date"][6:8])
|
||||||
|
published = int(datetime.datetime(year, month, day).timestamp())
|
||||||
|
|
||||||
# Adaptive formats have either audio or video, format streams have both
|
# Adaptive formats have either audio or video, format streams have both
|
||||||
def format_is_adaptive(format):
|
def format_is_adaptive(format):
|
||||||
return format["acodec"] == "none" or format["vcodec"] == "none"
|
return format["acodec"] == "none" or format["vcodec"] == "none"
|
||||||
|
|
||||||
# just the "type" field
|
def format_mime(format):
|
||||||
def format_type(format):
|
sense = "video" if format["vcodec"] != "none" else "audio"
|
||||||
sense = "audio"
|
return "{}/{}".format(sense, format["ext"])
|
||||||
|
|
||||||
|
def format_codecs(format):
|
||||||
codecs = []
|
codecs = []
|
||||||
if format["vcodec"] != "none":
|
if format["vcodec"] != "none":
|
||||||
sense = "video"
|
|
||||||
codecs.append(format["vcodec"])
|
codecs.append(format["vcodec"])
|
||||||
if format["acodec"] != "none":
|
if format["acodec"] != "none":
|
||||||
codecs.append(format["acodec"])
|
codecs.append(format["acodec"])
|
||||||
return '{}/{}; codecs="{}"'.format(sense, format["ext"], ", ".join(codecs))
|
return codecs
|
||||||
|
|
||||||
|
def format_type(format):
|
||||||
|
return '{}; codecs="{}"'.format(format_mime(format), ", ".join(format_codecs(format)))
|
||||||
|
|
||||||
result = {
|
result = {
|
||||||
"type": "video",
|
"type": "video",
|
||||||
"title": info["title"],
|
"title": info["title"],
|
||||||
"videoId": info["id"],
|
"videoId": info["id"],
|
||||||
"videoThumbnails": None,
|
"videoThumbnails": [],
|
||||||
"storyboards": None,
|
"storyboards": None,
|
||||||
"description": info["description"],
|
"description": info["description"],
|
||||||
"descriptionHtml": None,
|
"descriptionHtml": None,
|
||||||
"published": int(datetime.datetime(year, month, day).timestamp()),
|
"published": published,
|
||||||
"publishedText": None,
|
"publishedText": None,
|
||||||
"keywords": None,
|
"keywords": None,
|
||||||
"viewCount": info["view_count"],
|
"viewCount": info["view_count"],
|
||||||
@ -107,6 +124,8 @@ class Second(object):
|
|||||||
"url": format["url"],
|
"url": format["url"],
|
||||||
"itag": format["format_id"],
|
"itag": format["format_id"],
|
||||||
"type": format_type(format),
|
"type": format_type(format),
|
||||||
|
"second__mime": format_mime(format),
|
||||||
|
"second__codecs": format_codecs(format),
|
||||||
"clen": str(format["filesize"]),
|
"clen": str(format["filesize"]),
|
||||||
"lmt": None,
|
"lmt": None,
|
||||||
"projectionType": None,
|
"projectionType": None,
|
||||||
@ -122,6 +141,7 @@ class Second(object):
|
|||||||
"url": format["url"],
|
"url": format["url"],
|
||||||
"itag": format["format_id"],
|
"itag": format["format_id"],
|
||||||
"type": format_type(format),
|
"type": format_type(format),
|
||||||
|
"second__mime": format_mime(format),
|
||||||
"quality": None,
|
"quality": None,
|
||||||
"fps": format["fps"],
|
"fps": format["fps"],
|
||||||
"container": format["ext"],
|
"container": format["ext"],
|
||||||
@ -144,11 +164,12 @@ class Second(object):
|
|||||||
if len(possible_files) == 1:
|
if len(possible_files) == 1:
|
||||||
filename = possible_files[0]
|
filename = possible_files[0]
|
||||||
with open(filename) as file:
|
with open(filename) as file:
|
||||||
r = re.compile(r"""^\s*window\["ytInitialData"\] = (\{.*\});\n?$""")
|
r_yt_intial_data = re.compile(r"""^\s*window\["ytInitialData"\] = (\{.*\});\n?$""")
|
||||||
|
r_yt_player_config = re.compile(r"""^\s*[^"]+"cfg"[^"]+ytplayer\.config = (\{.*\});ytplayer\.web_player_context_config = {".""")
|
||||||
for line in file:
|
for line in file:
|
||||||
match_result = re.search(r, line)
|
m_yt_initial_data = re.search(r_yt_intial_data, line)
|
||||||
if match_result:
|
if m_yt_initial_data:
|
||||||
yt_initial_data = json.loads(match_result.group(1))
|
yt_initial_data = json.loads(m_yt_initial_data.group(1))
|
||||||
views = yt_initial_data["contents"]["twoColumnWatchNextResults"]["results"]["results"]["contents"][0]\
|
views = yt_initial_data["contents"]["twoColumnWatchNextResults"]["results"]["results"]["contents"][0]\
|
||||||
["videoPrimaryInfoRenderer"]["viewCount"]["videoViewCountRenderer"]
|
["videoPrimaryInfoRenderer"]["viewCount"]["videoViewCountRenderer"]
|
||||||
result["second__viewCountText"] = views["viewCount"]["simpleText"]
|
result["second__viewCountText"] = views["viewCount"]["simpleText"]
|
||||||
@ -208,6 +229,24 @@ class Second(object):
|
|||||||
"viewCount": get_view_count(r)
|
"viewCount": get_view_count(r)
|
||||||
} for r in [get_useful_recommendation_data(r) for r in recommendations if get_useful_recommendation_data(r)])
|
} for r in [get_useful_recommendation_data(r) for r in recommendations if get_useful_recommendation_data(r)])
|
||||||
|
|
||||||
|
m_yt_player_config = re.search(r_yt_player_config, line)
|
||||||
|
if m_yt_player_config:
|
||||||
|
yt_player_config = json.loads(m_yt_player_config.group(1))
|
||||||
|
player_response = json.loads(yt_player_config["args"]["player_response"])
|
||||||
|
itagDict = {}
|
||||||
|
for f in player_response["streamingData"]["adaptiveFormats"]:
|
||||||
|
itagDict[str(f["itag"])] = {
|
||||||
|
"initRange": f["initRange"],
|
||||||
|
"indexRange": f["indexRange"],
|
||||||
|
"audioChannels": f["audioChannels"] if "audioChannels" in f else None
|
||||||
|
}
|
||||||
|
for f in result["adaptiveFormats"]:
|
||||||
|
if f["itag"] in itagDict:
|
||||||
|
i = itagDict[f["itag"]]
|
||||||
|
f["init"] = "{}-{}".format(i["initRange"]["start"], i["initRange"]["end"])
|
||||||
|
f["index"] = "{}-{}".format(i["indexRange"]["start"], i["indexRange"]["end"])
|
||||||
|
f["second__audioChannels"] = i["audioChannels"]
|
||||||
|
|
||||||
except Exception:
|
except Exception:
|
||||||
traceback.print_exc()
|
traceback.print_exc()
|
||||||
|
|
||||||
@ -215,7 +254,7 @@ class Second(object):
|
|||||||
for file in possible_files:
|
for file in possible_files:
|
||||||
os.unlink(file)
|
os.unlink(file)
|
||||||
|
|
||||||
# return recommendations
|
self.video_cache[id] = result
|
||||||
return result
|
return result
|
||||||
|
|
||||||
except youtube_dl.DownloadError:
|
except youtube_dl.DownloadError:
|
||||||
@ -224,6 +263,50 @@ class Second(object):
|
|||||||
"identifier": "VIDEO_DOES_NOT_EXIST"
|
"identifier": "VIDEO_DOES_NOT_EXIST"
|
||||||
}
|
}
|
||||||
|
|
||||||
|
@cherrypy.expose
|
||||||
|
@cherrypy.tools.encode()
|
||||||
|
def manifest(self, id):
|
||||||
|
id = id.split(".")[0] # remove extension if present
|
||||||
|
video = self.videos(id)
|
||||||
|
|
||||||
|
if "error" in video:
|
||||||
|
return video
|
||||||
|
|
||||||
|
adaptation_sets_dict = {}
|
||||||
|
for f in video["adaptiveFormats"]:
|
||||||
|
mime = f["second__mime"]
|
||||||
|
if not mime in adaptation_sets_dict:
|
||||||
|
adaptation_sets_dict[mime] = []
|
||||||
|
ads = adaptation_sets_dict[mime]
|
||||||
|
|
||||||
|
representation_attributes = {"id": f["itag"], "codecs": ", ".join(f["second__codecs"]), "bandwidth": f["bitrate"]}
|
||||||
|
if f["second__width"]:
|
||||||
|
representation_attributes["width"] = str(f["second__width"])
|
||||||
|
representation_attributes["height"] = str(f["second__height"])
|
||||||
|
representation_attributes["startWithSAP"] = "1"
|
||||||
|
representation_attributes["maxPlayoutRate"] = "1"
|
||||||
|
representation_attributes["frameRate"] = str(f["fps"])
|
||||||
|
representation = ET.Element("Representation", representation_attributes)
|
||||||
|
if f["second__audioChannels"]:
|
||||||
|
ET.SubElement(representation, "AudioChannelConfiguration", {"schemeIdUri": "urn:mpeg:dash:23003:3:audio_channel_configuration:2011", "value": str(f["second__audioChannels"])})
|
||||||
|
ET.SubElement(representation, "BaseURL").text = f["url"]
|
||||||
|
et_segment_base = ET.SubElement(representation, "SegmentBase", {"indexRange": f["index"]})
|
||||||
|
ET.SubElement(et_segment_base, "Initialization", {"range": f["init"]})
|
||||||
|
ads.append(representation)
|
||||||
|
|
||||||
|
s_meta = B'<?xml version="1.0" encoding="UTF-8"?>'
|
||||||
|
et_mpd = ET.Element("MPD", {"xmlns": "urn:mpeg:dash:schema:mpd:2011", "profiles": "urn:mpeg:dash:profile:full:2011", "minBufferTime": "PT1.5S", "type": "static", "mediaPresentationDuration": "PT282S"})
|
||||||
|
et_period = ET.SubElement(et_mpd, "Period")
|
||||||
|
for (index, key) in list(enumerate(adaptation_sets_dict)):
|
||||||
|
ads = adaptation_sets_dict[key]
|
||||||
|
et_adaptation_set = ET.SubElement(et_period, "AdaptationSet", {"id": str(index), "mimeType": key, "startWithSAP": "1", "subsegmentAlignment": "true"})
|
||||||
|
for representation in ads:
|
||||||
|
et_adaptation_set.append(representation)
|
||||||
|
manifest = s_meta + ET.tostring(et_mpd)
|
||||||
|
|
||||||
|
cherrypy.response.headers["content-type"] = "application/dash+xml"
|
||||||
|
return manifest
|
||||||
|
|
||||||
@cherrypy.expose
|
@cherrypy.expose
|
||||||
@cherrypy.tools.json_out()
|
@cherrypy.tools.json_out()
|
||||||
def channels(self, *suffix):
|
def channels(self, *suffix):
|
||||||
@ -290,7 +373,7 @@ class Second(object):
|
|||||||
|
|
||||||
@cherrypy.expose
|
@cherrypy.expose
|
||||||
@cherrypy.tools.json_out()
|
@cherrypy.tools.json_out()
|
||||||
def search(self, *, q, sort_by):
|
def search(self, *, q, sort_by, page, date, duration, type):
|
||||||
info = ytdl.extract_info("ytsearchall:{}".format(q), download=False)
|
info = ytdl.extract_info("ytsearchall:{}".format(q), download=False)
|
||||||
return list({
|
return list({
|
||||||
"type": "video",
|
"type": "video",
|
||||||
|
@ -1,2 +1,3 @@
|
|||||||
cherrypy
|
cherrypy
|
||||||
youtube-dl
|
youtube-dl
|
||||||
|
cachetools
|
||||||
|
Loading…
Reference in New Issue
Block a user