From 7ce02b5bf95cb4b1bac9b23344dbcabace5f9511 Mon Sep 17 00:00:00 2001 From: Conner <65339198+connercsbn@users.noreply.github.com> Date: Sun, 14 Nov 2021 16:00:20 -0800 Subject: [PATCH 1/5] [Podchaser] Add new extractor --- youtube_dl/extractor/extractors.py | 1 + youtube_dl/extractor/podchaser.py | 98 ++++++++++++++++++++++++++++++ 2 files changed, 99 insertions(+) create mode 100644 youtube_dl/extractor/podchaser.py diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 6e8fc3961..eaa78b16d 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -929,6 +929,7 @@ from .pluralsight import ( PluralsightIE, PluralsightCourseIE, ) +from .podchaser import PodchaserIE from .podomatic import PodomaticIE from .pokemon import PokemonIE from .polskieradio import ( diff --git a/youtube_dl/extractor/podchaser.py b/youtube_dl/extractor/podchaser.py new file mode 100644 index 000000000..80f970c1b --- /dev/null +++ b/youtube_dl/extractor/podchaser.py @@ -0,0 +1,98 @@ +# coding: utf-8 +from __future__ import unicode_literals +import re +from datetime import datetime +from ..utils import float_or_none, try_get, str_to_int +from ..compat import compat_str +from .common import InfoExtractor + + +class PodchaserIE(InfoExtractor): + _VALID_URL = r'''(?x) + https?://(?:www\.)?podchaser\.com/ + (?:(?:podcasts)|(?:creators)) + /[\w\-]+- + (?: + (?P[\d]+[\w]+)|(?P[\d]+)) + (?:/episodes/[\w\-]+- + (?P[\d]+))?''' + + _TESTS = [{ + 'url': 'https://www.podchaser.com/podcasts/cum-town-36924/episodes/ep-285-freeze-me-off-104365585', + 'info_dict': { + 'id': '104365585', + 'title': "Ep. 285 – freeze me off", + 'description': 'cam ahn', + 'thumbnail': r're:^https?://.*\.jpg$', + 'ext': 'mp3', + 'categories': ['Comedy'], + 'series': 'Cum Town', + 'duration': 3708, + 'timestamp': 1636560059, + 'upload_date': '20211110' + } + }, { + 'url': 'https://www.podchaser.com/podcasts/the-bone-zone-28853', + 'info_dict': { + 'id': '28853', + 'title': 'The Bone Zone', + 'description': 'md5:c39acd897170a8bf3ad94fc45dc25060', + }, + 'playlist_count': 6 + }, { + 'url': 'https://www.podchaser.com/creators/todd-glass-107ZzkFiEQ', + 'info_dict': { + 'id': '107ZzkFiEQ', + 'title': 'Todd Glass', + 'description': 'md5:f0302588c98d474d4e8551f4914042c9', + }, + 'playlist_mincount': 48 + }, { + 'url': 'https://www.podchaser.com/podcasts/sean-carrolls-mindscape-scienc-699349/episodes', + 'info_dict': { + 'id': '699349', + 'title': "Sean Carroll's Mindscape: Science, Society, Philosophy, Culture, Arts, and Ideas", + 'description': 'md5:a1ed5db77378a736b3a85a2331cbc05a' + }, + 'playlist_count': 25 + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + audio_id, podcast_id, creator_id = compat_str(mobj.group('id')), \ + compat_str(mobj.group('podcast_id')), compat_str(mobj.group('creator_id')) + webpage = self._download_webpage(url, audio_id) + page_title = self._html_search_meta(['title', 'og:title', 'twitter:title'], webpage, default=None) \ + or self._search_regex( + r']*>(.+?)', webpage, 'title', fatal=False, default="Podchaser Podcast") + page_description = self._html_search_meta(['description', 'og:description', 'twitter:description'], webpage, default=None) + + data = self._search_regex( + r'window.__APP_STATE__\s*=\s*(["\']?{.+?}["\']?);', webpage, 'app state') + + while isinstance(data, compat_str): + data = self._parse_json(data, audio_id) + + episodes = try_get(data, lambda x: x['podcast']['episodes']['entities'], dict) or {} + episode_list = [(episodes.get(episode_id), episode_id) for episode_id in episodes] + + entries = [{ + 'id': episode_id, + 'title': episode.get('title'), + 'description': episode.get('description'), + 'url': episode.get('audio_url'), + 'thumbnail': try_get(episode, lambda x: x['podcast']['image_url']), + 'duration': str_to_int(episode.get('length')), + 'timestamp': datetime.strptime(episode.get('air_date'), '%Y-%m-%d %H:%M:%S').timestamp(), + 'rating': float_or_none(episode.get('rating')), + 'categories': [x['text'] for x in try_get(episode, lambda x: x['podcast']['categories'], list) or []], + 'tags': [tag['text'] for tag in episode.get('tags') or []], + 'series': try_get(episode, lambda x: x['podcast']['title'], compat_str), + } for episode, episode_id in episode_list] + + if len(entries) > 1: + return self.playlist_result( + entries, playlist_id=(creator_id or podcast_id), playlist_title=page_title, + playlist_description=page_description) + + return entries[0] From 4d83c072b195f4deaf41f0ac08b909dfc87fa13a Mon Sep 17 00:00:00 2001 From: Conner <65339198+connercsbn@users.noreply.github.com> Date: Wed, 17 Nov 2021 14:41:34 -0800 Subject: [PATCH 2/5] drop unnecessary compat_str --- youtube_dl/extractor/podchaser.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/youtube_dl/extractor/podchaser.py b/youtube_dl/extractor/podchaser.py index 80f970c1b..a6b9752fe 100644 --- a/youtube_dl/extractor/podchaser.py +++ b/youtube_dl/extractor/podchaser.py @@ -59,8 +59,8 @@ class PodchaserIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - audio_id, podcast_id, creator_id = compat_str(mobj.group('id')), \ - compat_str(mobj.group('podcast_id')), compat_str(mobj.group('creator_id')) + audio_id, podcast_id, creator_id = mobj.group('id'), \ + mobj.group('podcast_id'), mobj.group('creator_id') webpage = self._download_webpage(url, audio_id) page_title = self._html_search_meta(['title', 'og:title', 'twitter:title'], webpage, default=None) \ or self._search_regex( From 89ef8b9b99a5cdf24a81dc3070106302bef64724 Mon Sep 17 00:00:00 2001 From: Conner <65339198+connercsbn@users.noreply.github.com> Date: Wed, 17 Nov 2021 14:43:21 -0800 Subject: [PATCH 3/5] regex tweak --- youtube_dl/extractor/podchaser.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/podchaser.py b/youtube_dl/extractor/podchaser.py index a6b9752fe..95509764e 100644 --- a/youtube_dl/extractor/podchaser.py +++ b/youtube_dl/extractor/podchaser.py @@ -11,7 +11,7 @@ class PodchaserIE(InfoExtractor): _VALID_URL = r'''(?x) https?://(?:www\.)?podchaser\.com/ (?:(?:podcasts)|(?:creators)) - /[\w\-]+- + /[\w-]+- (?: (?P[\d]+[\w]+)|(?P[\d]+)) (?:/episodes/[\w\-]+- From 6ddf44e2f31dc74d819cfe9fbc25b717e214e40e Mon Sep 17 00:00:00 2001 From: Conner <65339198+connercsbn@users.noreply.github.com> Date: Mon, 22 Nov 2021 17:39:26 -0800 Subject: [PATCH 4/5] use unified_timestamp --- youtube_dl/extractor/podchaser.py | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/youtube_dl/extractor/podchaser.py b/youtube_dl/extractor/podchaser.py index 95509764e..b3f80ab2a 100644 --- a/youtube_dl/extractor/podchaser.py +++ b/youtube_dl/extractor/podchaser.py @@ -2,7 +2,7 @@ from __future__ import unicode_literals import re from datetime import datetime -from ..utils import float_or_none, try_get, str_to_int +from ..utils import float_or_none, try_get, str_to_int, unified_timestamp from ..compat import compat_str from .common import InfoExtractor @@ -28,7 +28,7 @@ class PodchaserIE(InfoExtractor): 'categories': ['Comedy'], 'series': 'Cum Town', 'duration': 3708, - 'timestamp': 1636560059, + 'timestamp': 1636531259, 'upload_date': '20211110' } }, { @@ -44,7 +44,7 @@ class PodchaserIE(InfoExtractor): 'info_dict': { 'id': '107ZzkFiEQ', 'title': 'Todd Glass', - 'description': 'md5:f0302588c98d474d4e8551f4914042c9', + 'description': 'md5:0771e81d879f304f11254e5a56a97a58', }, 'playlist_mincount': 48 }, { @@ -52,7 +52,7 @@ class PodchaserIE(InfoExtractor): 'info_dict': { 'id': '699349', 'title': "Sean Carroll's Mindscape: Science, Society, Philosophy, Culture, Arts, and Ideas", - 'description': 'md5:a1ed5db77378a736b3a85a2331cbc05a' + 'description': 'md5:8692ce0c50cb900c5e4eb27b437dd67b' }, 'playlist_count': 25 }] @@ -83,7 +83,7 @@ class PodchaserIE(InfoExtractor): 'url': episode.get('audio_url'), 'thumbnail': try_get(episode, lambda x: x['podcast']['image_url']), 'duration': str_to_int(episode.get('length')), - 'timestamp': datetime.strptime(episode.get('air_date'), '%Y-%m-%d %H:%M:%S').timestamp(), + 'timestamp': unified_timestamp(episode.get('air_date')), 'rating': float_or_none(episode.get('rating')), 'categories': [x['text'] for x in try_get(episode, lambda x: x['podcast']['categories'], list) or []], 'tags': [tag['text'] for tag in episode.get('tags') or []], From 94c633d58a598334beb49bf2871de125839d4861 Mon Sep 17 00:00:00 2001 From: Conner <65339198+connercsbn@users.noreply.github.com> Date: Sat, 27 Nov 2021 06:41:44 -0800 Subject: [PATCH 5/5] Change to using Podchaser's API for data retrieval I've temporarily removed the `podchaser.com/creators/creator-` path from _VALID_URL, because I haven't figured out how to get a Creator's 4 or 5 digit numeric ID from the alphanumeric ID in the URL--at least not without downloading the page's HTML, which looks tricky atm with cloudflare --- youtube_dl/extractor/podchaser.py | 92 ++++++++++++++++++------------- 1 file changed, 54 insertions(+), 38 deletions(-) diff --git a/youtube_dl/extractor/podchaser.py b/youtube_dl/extractor/podchaser.py index b3f80ab2a..83fa2246d 100644 --- a/youtube_dl/extractor/podchaser.py +++ b/youtube_dl/extractor/podchaser.py @@ -1,8 +1,8 @@ # coding: utf-8 from __future__ import unicode_literals +import json import re -from datetime import datetime -from ..utils import float_or_none, try_get, str_to_int, unified_timestamp +from ..utils import float_or_none, try_get, str_to_int, unified_timestamp, merge_dicts from ..compat import compat_str from .common import InfoExtractor @@ -10,10 +10,8 @@ from .common import InfoExtractor class PodchaserIE(InfoExtractor): _VALID_URL = r'''(?x) https?://(?:www\.)?podchaser\.com/ - (?:(?:podcasts)|(?:creators)) - /[\w-]+- (?: - (?P[\d]+[\w]+)|(?P[\d]+)) + (?:podcasts/[\w-]+-(?P[\d]+))) (?:/episodes/[\w\-]+- (?P[\d]+))?''' @@ -26,6 +24,7 @@ class PodchaserIE(InfoExtractor): 'thumbnail': r're:^https?://.*\.jpg$', 'ext': 'mp3', 'categories': ['Comedy'], + 'tags': ['comedy', 'dark humor'], 'series': 'Cum Town', 'duration': 3708, 'timestamp': 1636531259, @@ -36,63 +35,80 @@ class PodchaserIE(InfoExtractor): 'info_dict': { 'id': '28853', 'title': 'The Bone Zone', - 'description': 'md5:c39acd897170a8bf3ad94fc45dc25060', + 'description': 'Podcast by The Bone Zone', }, - 'playlist_count': 6 - }, { - 'url': 'https://www.podchaser.com/creators/todd-glass-107ZzkFiEQ', - 'info_dict': { - 'id': '107ZzkFiEQ', - 'title': 'Todd Glass', - 'description': 'md5:0771e81d879f304f11254e5a56a97a58', - }, - 'playlist_mincount': 48 + 'playlist_count': 275 }, { 'url': 'https://www.podchaser.com/podcasts/sean-carrolls-mindscape-scienc-699349/episodes', 'info_dict': { 'id': '699349', 'title': "Sean Carroll's Mindscape: Science, Society, Philosophy, Culture, Arts, and Ideas", - 'description': 'md5:8692ce0c50cb900c5e4eb27b437dd67b' + 'description': 'md5:2cbd8f4749891a84dc8235342e0b5ff1' }, - 'playlist_count': 25 + 'playlist_mincount': 199 }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - audio_id, podcast_id, creator_id = mobj.group('id'), \ - mobj.group('podcast_id'), mobj.group('creator_id') - webpage = self._download_webpage(url, audio_id) - page_title = self._html_search_meta(['title', 'og:title', 'twitter:title'], webpage, default=None) \ - or self._search_regex( - r']*>(.+?)', webpage, 'title', fatal=False, default="Podchaser Podcast") - page_description = self._html_search_meta(['description', 'og:description', 'twitter:description'], webpage, default=None) + audio_id, podcast_id = mobj.group('id'), mobj.group('podcast_id') - data = self._search_regex( - r'window.__APP_STATE__\s*=\s*(["\']?{.+?}["\']?);', webpage, 'app state') + # If one episode + if audio_id: + episodes = [self._download_json("https://api.podchaser.com/episodes/%s" % audio_id, audio_id)] - while isinstance(data, compat_str): - data = self._parse_json(data, audio_id) + # Else get every episode available + else: + total_episode_count = self._download_json( + "https://api.podchaser.com/list/episode", podcast_id, + headers={'Content-Type': 'application/json;charset=utf-8'}, + data=json.dumps({ + "filters": {"podcast_id": podcast_id} + }).encode()).get('total') + episodes = [] + print(total_episode_count) + for i in range(total_episode_count // 100 + 1): + curr_episodes_data = self._download_json( + "https://api.podchaser.com/list/episode", podcast_id, + headers={'Content-Type': 'application/json;charset=utf-8'}, + data=json.dumps({ + "start": i * 100, + "count": (i + 1) * 100, + "sort_order": "SORT_ORDER_RECENT", + "filters": { + "podcast_id": podcast_id + }, "options": {} + }).encode()) + curr_episodes = curr_episodes_data.get('entities') or [] + if len(curr_episodes) + len(episodes) <= total_episode_count: + episodes.extend(curr_episodes) - episodes = try_get(data, lambda x: x['podcast']['episodes']['entities'], dict) or {} - episode_list = [(episodes.get(episode_id), episode_id) for episode_id in episodes] + podcast_data = merge_dicts( + self._download_json("https://api.podchaser.com/podcasts/%s" % podcast_id, audio_id or podcast_id) or {}, + episodes[0].get('podcast') or {} if episodes else {}) entries = [{ - 'id': episode_id, + 'id': compat_str(episode.get('id')), 'title': episode.get('title'), 'description': episode.get('description'), 'url': episode.get('audio_url'), - 'thumbnail': try_get(episode, lambda x: x['podcast']['image_url']), + 'thumbnail': episode.get('image_url'), 'duration': str_to_int(episode.get('length')), 'timestamp': unified_timestamp(episode.get('air_date')), 'rating': float_or_none(episode.get('rating')), - 'categories': [x['text'] for x in try_get(episode, lambda x: x['podcast']['categories'], list) or []], - 'tags': [tag['text'] for tag in episode.get('tags') or []], - 'series': try_get(episode, lambda x: x['podcast']['title'], compat_str), - } for episode, episode_id in episode_list] + 'categories': [ + x.get('text') for x in + podcast_data.get('categories') + or try_get(podcast_data, lambda x: x['summary']['categories'], list) or []], + 'tags': [tag.get('text') for tag in podcast_data.get('tags') or []], + 'series': podcast_data.get('title'), + } for episode in episodes] if len(entries) > 1: + # Return playlist return self.playlist_result( - entries, playlist_id=(creator_id or podcast_id), playlist_title=page_title, - playlist_description=page_description) + entries, playlist_id=compat_str(podcast_data.get('id')), + playlist_title=podcast_data.get('title'), + playlist_description=podcast_data.get('description')) + # Return episode return entries[0]