BoomPlayGenericPlaylistIE, BoomPlaySearchIE

yt_dlp/extractor/_extractors.py

@@ -282,9 +282,11 @@ from .bokecc import BokeCCIE
 from .bongacams import BongaCamsIE
 from .boomplay import (
     BoomPlayEpisodeIE,
+    BoomPlayGenericPlaylistIE,
     BoomPlayMusicIE,
     BoomPlayPlaylistIE,
     BoomPlayPodcastIE,
+    BoomPlaySearchIE,
     BoomPlayVideoIE,
 )
 from .boosty import BoostyIE

yt_dlp/extractor/boomplay.py

@@ -2,22 +2,29 @@ import base64
 import functools
 import json
 import re
+import urllib.parse
 
-from .common import InfoExtractor
+from .common import InfoExtractor, SearchInfoExtractor
 from ..aes import aes_cbc_decrypt_bytes, aes_cbc_encrypt_bytes, unpad_pkcs7
 from ..utils import (
     ExtractorError,
     clean_html,
+    extract_attributes,
     get_element_by_attribute,
     get_element_by_class,
     get_elements_by_attribute,
     int_or_none,
+    join_nonempty,
     merge_dicts,
+    orderedSet,
+    parse_count,
     parse_duration,
     strip_or_none,
     unified_strdate,
     url_or_none,
     urlencode_postdata,
+    urljoin,
+    variadic,
 )
 from ..utils.traversal import traverse_obj
 
@@ -27,6 +34,14 @@ class BoomPlayBaseIE(InfoExtractor):
     # Note that the real key/iv differs from `lhx.AESUtils.key`/`lhx.AESUtils.iv`
     _KEY = b'boomplayVr3xopAM'
     _IV = b'boomplay8xIsKTn9'
+    _BASE = 'https://www.boomplay.com'
+    _MEDIA_TYPES = ('songs', 'video', 'episode', 'podcasts', 'playlists', 'artists', 'albums')
+
+    @classmethod
+    def _urljoin(cls, path):
+        if not hasattr(path, 'startswith') or path.startswith('javascript:'):
+            return None
+        return url_or_none(urljoin(base=cls._BASE, path=path))
 
     def _get_playurl(self, item_id, item_type):
         resp = self._download_json(
@@ -49,7 +64,6 @@ class BoomPlayBaseIE(InfoExtractor):
         if url := url_or_none(self._get_playurl(_id, item_type)):
             return [{
                 'format_id': '0',
-                'vcodec': 'none' if item_type == 'MUSIC' else None,
                 'url': url,
                 'http_headers': {
                     'Origin': 'https://www.boomplay.com',
@@ -79,12 +93,12 @@ class BoomPlayBaseIE(InfoExtractor):
         metadata_entries.extend(re.findall(r'(?s)<li>(?P<entry>.*?)</li>', details_section) or [])
         page_metadata = {
             'id': _id,
-            'title': self._html_search_regex(r'<h1>([^<]+)</h1>', metadata_div, 'title', default=''),
+            'title': self._html_search_regex(r'<h1>([^<]+)</h1>', metadata_div, 'title', default=None),
             'thumbnail': self._html_search_meta(['og:image', 'twitter:image'],
                                                 webpage, 'thumbnail', default=''),
-            'like_count': int_or_none(get_element_by_class('btn_favorite', metadata_div)),
+            'like_count': parse_count(get_element_by_class('btn_favorite', metadata_div)),
-            'repost_count': int_or_none(get_element_by_class('btn_share', metadata_div)),
+            'repost_count': parse_count(get_element_by_class('btn_share', metadata_div)),
-            'comment_count': int_or_none(get_element_by_class('btn_comment', metadata_div)),
+            'comment_count': parse_count(get_element_by_class('btn_comment', metadata_div)),
             'duration': parse_duration(get_element_by_class('btn_duration', metadata_div)),
             'upload_date': unified_strdate(strip_or_none(get_element_by_class('btn_pubDate', metadata_div))),
             'description': description,
@@ -104,7 +118,55 @@ class BoomPlayBaseIE(InfoExtractor):
                 page_metadata['release_year'] = int_or_none(v)
         return page_metadata
 
-    extract = lambda self, url: self.write_debug(json.dumps(a := super().extract(url), indent=2)) or a  # rm
+    def _extract_suitable_links(self, webpage, media_types):
+        if not media_types:
+            media_types = self._MEDIA_TYPES
+        media_types = list(variadic(media_types))
+
+        for idx, v in enumerate(media_types):
+            media_types[idx] = re.escape(v) if v in self._MEDIA_TYPES else ''
+        media_types = join_nonempty(*media_types, delim='|')
+        return orderedSet(traverse_obj(re.finditer(
+            rf'''(?x)
+            <a
+                (?:\s(?:[^>"']|"[^"]*"|'[^']*')*)?
+                    (?<=\s)href\s*=\s*(?P<_q>['"])
+                        (?:
+                            (?!javascript:)(?P<link>/(?:{media_types})/\d+?)
+                        )
+                    (?P=_q)
+                (?:\s(?:[^>"']|"[^"]*"|'[^']*')*)?
+            ''', webpage), (..., 'link', {self._urljoin}, {self.url_result})))
+
+    def _extract_playlist_entries(self, webpage, media_types, warn=True):
+        song_list = strip_or_none(
+            get_element_by_attribute(
+                'class', r'[^\'"]*(?<=[\'"\s])morePart_musics(?=[\'"\s])[^\'"]*', webpage,
+                tag='ol', escape_value=False)
+            or get_element_by_attribute(
+                'class', r'[^\'"]*(?<=[\'"\s])morePart(?=[\'"\s])[^\'"]*', webpage,
+                tag='ol', escape_value=False)
+            or '')
+
+        entries = traverse_obj(re.finditer(
+            r'''(?x)
+            <a
+                (?:\s(?:[^>"']|"[^"]*"|'[^']*')*)?
+                    (?<=\s)class\s*=\s*(?P<_q>['"])
+                        (?:
+                            [^\'"]*(?<=[\'"\s])songName(?=[\'"\s])[^\'"]*
+                        )
+                    (?P=_q)
+                (?:\s(?:[^>"']|"[^"]*"|'[^']*')*)?
+            >
+            ''', song_list),
+            (..., 0, {extract_attributes}, 'href', {self._urljoin}, {self.url_result}))
+        if not entries:
+            if warn:
+                self.report_warning('Failed to extract playlist entries, finding suitable links instead!')
+            return self._extract_suitable_links(webpage, media_types)
+
+        return entries
 
 
 class BoomPlayMusicIE(BoomPlayBaseIE):
@@ -143,7 +205,7 @@ class BoomPlayMusicIE(BoomPlayBaseIE):
                 'artists': ('byArtist', ..., 'name'),
                 'duration': ('duration', {parse_duration}),
             }), {
-                'formats': self._extract_formats(song_id, 'MUSIC'),
+                'formats': self._extract_formats(song_id, 'MUSIC', vcodec='none'),
             })
 
 
@@ -217,6 +279,7 @@ class BoomPlayPodcastIE(BoomPlayBaseIE):
             'thumbnail': 'https://source.boomplaymusic.com/group10/M00/12/22/6f9cf97ad6f846a0a7882c98dfcf4f8c_320_320.jpg',
             'repost_count': int,
             'comment_count': int,
+            'like_count': int,
         },
     }
 
@@ -253,6 +316,7 @@ class BoomPlayPlaylistIE(BoomPlayBaseIE):
             'thumbnail': 'https://source.boomplaymusic.com/group10/M00/08/19/d05d431ee616412caeacd7f78f4f68f5_320_320.jpeg',
             'repost_count': int,
             'comment_count': int,
+            'like_count': int,
             'description': 'md5:7ebdffc5137c77acb62acb3c89248445',
         },
         'playlist_count': 10,
@@ -281,3 +345,75 @@ class BoomPlayPlaylistIE(BoomPlayBaseIE):
                 'artists': ('byArtist', ..., 'name'),
                 'channel_url': ('byArtist', 0, '@id'),
             })))
+
+
+class BoomPlayGenericPlaylistIE(BoomPlayBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?boomplay\.com/.+'
+    _TESTS = [{
+        'url': 'https://www.boomplay.com/search/default/Rise%20of%20the%20Fallen%20Heroes',
+        'md5': 'c5fb4f23e6aae98064230ef3c39c2178',
+        'info_dict': {
+            'id': '165481965',
+            'ext': 'mp3',
+            'title': 'Rise of the Fallen Heroes',
+            'duration': 125.0,
+            'genres': ['Metal'],
+            'artists': ['fatbunny'],
+            'thumbnail': 'https://source.boomplaymusic.com/group10/M00/04/29/375ecda38f6f48179a93c72ab909118f_464_464.jpg',
+            'channel_url': 'https://www.boomplay.com/artists/52723101',
+            'comment_count': int,
+            'repost_count': int,
+            'album': 'Legendary Battle',
+            'release_year': 2024,
+            'like_count': int,
+        },
+    }, {
+        'url': 'https://www.boomplay.com/search/video/%20Autumn%20blues',
+        'md5': 'd9b67ad333d2292a82922062d065352d',
+        'info_dict': {
+            'id': '1154892',
+            'title': 'Autumn blues',
+            'ext': 'mp4',
+            'timestamp': 1728599214,
+            'view_count': int,
+            'thumbnail': 'https://source.boomplaymusic.com/group10/M00/10/10/2171dee9e1f8452e84021560729edb88.jpg',
+            'description': 'Autumn blues by Lugo',
+            'upload_date': '20241010',
+            'duration': 177.0,
+        },
+        'expected_warnings': ['Failed to extract playlist entries, finding suitable links instead!'],
+        'params': {'playlist_items': '1'},
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        if not any(ie.suitable(url) for ie in (
+            BoomPlayEpisodeIE,
+            BoomPlayMusicIE,
+            BoomPlayPlaylistIE,
+            BoomPlayPodcastIE,
+            BoomPlayVideoIE,
+        )):
+            return super().suitable(url)
+        return False
+
+    def _real_extract(self, url):
+        _id = self._generic_id(url)
+        webpage = self._download_webpage(url, _id)
+        # TODO: pass media types based on search types
+        return self.playlist_result(
+            self._extract_playlist_entries(webpage, self._MEDIA_TYPES),
+            **self._extract_page_metadata(webpage, _id))
+
+
+class BoomPlaySearchIE(SearchInfoExtractor):
+    _SEARCH_KEY = 'boomplaysearch'
+    _RETURN_TYPE = 'url'
+    _TEST = {
+        'url': 'boomplaysearch:rise of the fallen heroes',
+        'only_matching': True,
+    }
+
+    def _search_results(self, query):
+        yield self.url_result(
+            f'https://www.boomplay.com/search/default/{urllib.parse.quote(query)}')