diff options
Diffstat (limited to 'hypervideo_dl/extractor/abematv.py')
| -rw-r--r-- | hypervideo_dl/extractor/abematv.py | 114 |
1 files changed, 40 insertions, 74 deletions
diff --git a/hypervideo_dl/extractor/abematv.py b/hypervideo_dl/extractor/abematv.py index 80046af..8f962ba 100644 --- a/hypervideo_dl/extractor/abematv.py +++ b/hypervideo_dl/extractor/abematv.py @@ -22,80 +22,23 @@ from ..utils import ( int_or_none, intlist_to_bytes, OnDemandPagedList, - request_to_url, time_seconds, traverse_obj, update_url_query, ) -# NOTE: network handler related code is temporary thing until network stack overhaul PRs are merged (#2861/#2862) - -def add_opener(ydl, handler): - ''' Add a handler for opening URLs, like _download_webpage ''' - # https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L426 - # https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L605 - assert isinstance(ydl._opener, urllib.request.OpenerDirector) - ydl._opener.add_handler(handler) - - -def remove_opener(ydl, handler): - ''' - Remove handler(s) for opening URLs - @param handler Either handler object itself or handler type. - Specifying handler type will remove all handler which isinstance returns True. - ''' +def add_opener(ydl, handler): # FIXME: Create proper API in .networking + """Add a handler for opening URLs, like _download_webpage""" # https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L426 # https://github.com/python/cpython/blob/main/Lib/urllib/request.py#L605 - opener = ydl._opener - assert isinstance(ydl._opener, urllib.request.OpenerDirector) - if isinstance(handler, (type, tuple)): - find_cp = lambda x: isinstance(x, handler) - else: - find_cp = lambda x: x is handler - - removed = [] - for meth in dir(handler): - if meth in ["redirect_request", "do_open", "proxy_open"]: - # oops, coincidental match - continue - - i = meth.find("_") - protocol = meth[:i] - condition = meth[i + 1:] - - if condition.startswith("error"): - j = condition.find("_") + i + 1 - kind = meth[j + 1:] - try: - kind = int(kind) - except ValueError: - pass - lookup = opener.handle_error.get(protocol, {}) - opener.handle_error[protocol] = lookup - elif condition == "open": - kind = protocol - lookup = opener.handle_open - elif condition == "response": - kind = protocol - lookup = opener.process_response - elif condition == "request": - kind = protocol - lookup = opener.process_request - else: - continue - - handlers = lookup.setdefault(kind, []) - if handlers: - handlers[:] = [x for x in handlers if not find_cp(x)] - - removed.append(x for x in handlers if find_cp(x)) - - if removed: - for x in opener.handlers: - if find_cp(x): - x.add_parent(None) - opener.handlers[:] = [x for x in opener.handlers if not find_cp(x)] + rh = ydl._request_director.handlers['Urllib'] + if 'abematv-license' in rh._SUPPORTED_URL_SCHEMES: + return + opener = rh._get_instance(cookiejar=ydl.cookiejar, proxies=ydl.proxies) + assert isinstance(opener, urllib.request.OpenerDirector) + opener.add_handler(handler) + rh._SUPPORTED_URL_SCHEMES = (*rh._SUPPORTED_URL_SCHEMES, 'abematv-license') class AbemaLicenseHandler(urllib.request.BaseHandler): @@ -137,11 +80,11 @@ class AbemaLicenseHandler(urllib.request.BaseHandler): return intlist_to_bytes(aes_ecb_decrypt(encvideokey, enckey)) def abematv_license_open(self, url): - url = request_to_url(url) + url = url.get_full_url() if isinstance(url, urllib.request.Request) else url ticket = urllib.parse.urlparse(url).netloc response_data = self._get_videokey_from_ticket(ticket) return urllib.response.addinfourl(io.BytesIO(response_data), headers={ - 'Content-Length': len(response_data), + 'Content-Length': str(len(response_data)), }, url=url, code=200) @@ -156,7 +99,7 @@ class AbemaTVBaseIE(InfoExtractor): def _generate_aks(cls, deviceid): deviceid = deviceid.encode('utf-8') # add 1 hour and then drop minute and secs - ts_1hour = int((time_seconds(hours=9) // 3600 + 1) * 3600) + ts_1hour = int((time_seconds() // 3600 + 1) * 3600) time_struct = time.gmtime(ts_1hour) ts_1hour_str = str(ts_1hour).encode('utf-8') @@ -190,6 +133,16 @@ class AbemaTVBaseIE(InfoExtractor): if self._USERTOKEN: return self._USERTOKEN + username, _ = self._get_login_info() + AbemaTVBaseIE._USERTOKEN = username and self.cache.load(self._NETRC_MACHINE, username) + if AbemaTVBaseIE._USERTOKEN: + # try authentication with locally stored token + try: + self._get_media_token(True) + return + except ExtractorError as e: + self.report_warning(f'Failed to login with cached user token; obtaining a fresh one ({e})') + AbemaTVBaseIE._DEVICE_ID = str(uuid.uuid4()) aks = self._generate_aks(self._DEVICE_ID) user_data = self._download_json( @@ -203,10 +156,7 @@ class AbemaTVBaseIE(InfoExtractor): }) AbemaTVBaseIE._USERTOKEN = user_data['token'] - # don't allow adding it 2 times or more, though it's guarded - remove_opener(self._downloader, AbemaLicenseHandler) add_opener(self._downloader, AbemaLicenseHandler(self)) - return self._USERTOKEN def _get_media_token(self, invalidate=False, to_show=True): @@ -300,6 +250,11 @@ class AbemaTVIE(AbemaTVBaseIE): _TIMETABLE = None def _perform_login(self, username, password): + self._get_device_token() + if self.cache.load(self._NETRC_MACHINE, username) and self._get_media_token(): + self.write_debug('Skipping logging in') + return + if '@' in username: # don't strictly check if it's email address or not ep, method = 'user/email', 'email' else: @@ -319,6 +274,7 @@ class AbemaTVIE(AbemaTVBaseIE): AbemaTVBaseIE._USERTOKEN = login_response['token'] self._get_media_token(True) + self.cache.store(self._NETRC_MACHINE, username, AbemaTVBaseIE._USERTOKEN) def _real_extract(self, url): # starting download using infojson from this extractor is undefined behavior, @@ -416,10 +372,20 @@ class AbemaTVIE(AbemaTVBaseIE): f'https://api.abema.io/v1/video/programs/{video_id}', video_id, note='Checking playability', headers=headers) - ondemand_types = traverse_obj(api_response, ('terms', ..., 'onDemandType'), default=[]) + ondemand_types = traverse_obj(api_response, ('terms', ..., 'onDemandType')) if 3 not in ondemand_types: # cannot acquire decryption key for these streams self.report_warning('This is a premium-only stream') + info.update(traverse_obj(api_response, { + 'series': ('series', 'title'), + 'season': ('season', 'title'), + 'season_number': ('season', 'sequence'), + 'episode_number': ('episode', 'number'), + })) + if not title: + title = traverse_obj(api_response, ('episode', 'title')) + if not description: + description = traverse_obj(api_response, ('episode', 'content')) m3u8_url = f'https://vod-abematv.akamaized.net/program/{video_id}/playlist.m3u8' elif video_type == 'slots': @@ -489,7 +455,7 @@ class AbemaTVTitleIE(AbemaTVBaseIE): }) yield from ( self.url_result(f'https://abema.tv/video/episode/{x}') - for x in traverse_obj(programs, ('programs', ..., 'id'), default=[])) + for x in traverse_obj(programs, ('programs', ..., 'id'))) def _entries(self, playlist_id, series_version): return OnDemandPagedList( |