Merge a94b69e4d3 into c5098961b0

Now also succeeds with player b12cc44b

test/test_jsinterp.py

@@ -425,6 +425,34 @@ class TestJSInterpreter(unittest.TestCase):
             self._test(jsi, [''], args=['', '-'])
             self._test(jsi, [], args=['', ''])
 
+    def test_slice(self):
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice()}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0)}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(5)}', [5, 6, 7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(99)}', [])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-2)}', [7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-99)}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0, 0)}', [])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(1, 0)}', [])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0, 1)}', [0])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(3, 6)}', [3, 4, 5])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(1, -1)}', [1, 2, 3, 4, 5, 6, 7])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-1, 1)}', [])
+        self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-3, -1)}', [6, 7])
+        self._test('function f(){return "012345678".slice()}', '012345678')
+        self._test('function f(){return "012345678".slice(0)}', '012345678')
+        self._test('function f(){return "012345678".slice(5)}', '5678')
+        self._test('function f(){return "012345678".slice(99)}', '')
+        self._test('function f(){return "012345678".slice(-2)}', '78')
+        self._test('function f(){return "012345678".slice(-99)}', '012345678')
+        self._test('function f(){return "012345678".slice(0, 0)}', '')
+        self._test('function f(){return "012345678".slice(1, 0)}', '')
+        self._test('function f(){return "012345678".slice(0, 1)}', '0')
+        self._test('function f(){return "012345678".slice(3, 6)}', '345')
+        self._test('function f(){return "012345678".slice(1, -1)}', '1234567')
+        self._test('function f(){return "012345678".slice(-1, 1)}', '')
+        self._test('function f(){return "012345678".slice(-3, -1)}', '67')
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_signature.py

@@ -174,6 +174,14 @@ _NSIG_TESTS = [
         'https://www.youtube.com/s/player/5604538d/player_ias.vflset/en_US/base.js',
         '7X-he4jjvMx7BCX', 'sViSydX8IHtdWA',
     ),
+    (
+        'https://www.youtube.com/s/player/20dfca59/player_ias.vflset/en_US/base.js',
+        '-fLCxedkAk4LUTK2', 'O8kfRq1y1eyHGw',
+    ),
+    (
+        'https://www.youtube.com/s/player/b12cc44b/player_ias.vflset/en_US/base.js',
+        'keLa5R2U00sR9SQK', 'N1OGyujjEwMnLw',
+    ),
 ]
 
 

youtube_dl/extractor/extractors.py

@@ -506,6 +506,7 @@ from .hungama import (
     HungamaSongIE,
 )
 from .hypem import HypemIE
+from .icareus import IcareusIE 
 from .ign import (
     IGNIE,
     IGNVideoIE,

youtube_dl/extractor/icareus.py

@@ -0,0 +1,230 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    determine_ext,
+    get_element_by_attribute,
+    get_element_by_class,
+    int_or_none,
+    parse_bitrate,
+    parse_resolution,
+    unified_timestamp,
+    urlencode_postdata,
+    url_or_none,
+)
+
+
+class IcareusIE(InfoExtractor):
+    _VALID_URL = r'''(?x)
+                    https?://(?:www\.)?
+                        (?:
+                            asahitv\.fi|
+                            helsinkikanava\.fi|
+                            hyvinvointitv\.fi|
+                            inez\.fi|
+                            midastv\.ke|
+                            permanto\.fi|
+                            suite.icareus.com|
+                            videos\.minifiddlers\.org
+                        )
+                        /.+/player/.*(?:assetId|eventId)=(?P<id>\d+).*'''
+    _TESTS = [{
+        'url': 'https://www.helsinkikanava.fi/fi_FI/web/helsinkikanava/player/vod?assetId=68021894',
+        'md5': 'ca0b62ffc814a5411dfa6349cf5adb8a',
+        'info_dict': {
+            'id': '68021894',
+            'ext': 'mp4',
+            'title': 'Perheiden parhaaksi',
+            'description': 'md5:fe4e4ec742a34f53022f3a0409b0f6e7',
+            'thumbnail': 'https://www.helsinkikanava.fi/image/image_gallery?img_id=68022501',
+            'upload_date': '20200924',
+            'timestamp': 1600938300,
+        },
+    }, {  # Recorded livestream
+        'url': 'https://www.helsinkikanava.fi/fi/web/helsinkikanava/player/event/view?eventId=76241489',
+        'md5': '014327e69dfa7b949fcc861f6d162d6d',
+        'info_dict': {
+            'id': '76258304',
+            'ext': 'mp4',
+            'title': 'Helsingin kaupungin ja HUSin tiedotustilaisuus koronaepidemiatilanteesta 24.11.2020',
+            'description': 'md5:3129d041c6fbbcdc7fe68d9a938fef1c',
+            'thumbnail': 'https://icareus-suite.secure2.footprint.net/image/image_gallery?img_id=76288630',
+            'upload_date': '20201124',
+            'timestamp': 1606206600,
+        },
+    }, {
+        'url': 'https://asahitv.fi/fi/web/asahi/player/vod?assetId=89415818',
+        'only_matching': True
+    }, {
+        'url': 'https://hyvinvointitv.fi/fi/web/hyvinvointitv/player/vod?assetId=89149730',
+        'only_matching': True
+    }, {
+        'url': 'https://inez.fi/fi/web/inez-media/player/vod?assetId=71328822',
+        'only_matching': True
+    }, {
+        'url': 'https://www.midastv.ke/en/web/midas-tv/player/embed/vod?assetId=65714535',
+        'only_matching': True
+    }, {
+        'url': 'https://www.permanto.fi/fi/web/alfatv/player/vod?assetId=95010095',
+        'only_matching': True
+    }, {
+        'url': 'https://suite.icareus.com/fi/web/westend-indians/player/vod?assetId=47567389',
+        'only_matching': True
+    }, {
+        'url': 'https://videos.minifiddlers.org/web/international-minifiddlers/player/vod?assetId=1982759',
+        'only_matching': True
+    }]
+    _API2_PATH = '/icareus-suite-api-portlet/publishing'
+
+    def _real_extract(self, url):
+        maybe_id = self._match_id(url)
+        page = self._download_webpage(url, maybe_id)
+        video_id = self._search_regex(
+            r"_icareus\['itemId'\]='(\d+)'", page, "video_id")
+        api_base = self._search_regex(
+            r'var publishingServiceURL = "(http.*?)";', page, "api_base")
+        organization_id = self._search_regex(
+            r"_icareus\['organizationId'\]='(\d+)'", page, "organization_id")
+        token = self._search_regex(
+            r"_icareus\['token'\]='([a-f0-9]+)'", page, "token")
+
+        token2 = self._search_regex(
+            r'''data\s*:\s*{action:"getAsset".*?token:'([a-f0-9]+)'}''', page,
+            "token2", default=None, fatal=False)
+        metajson = get_element_by_attribute('type', 'application/ld+json', page)
+
+        metad = None
+        if metajson:
+            # The description can contain newlines, HTML tags, quote chars etc.
+            # so we'll extract it manually
+            mo = re.match(
+                r'(.*",)\s*"description": "(.*?)",(\s*"thumbnailUrl":.*)',
+                metajson, flags=re.DOTALL)
+            if mo:
+                desc_text = mo.group(2)
+                metajson = mo.group(1) + mo.group(3)
+                metad = self._parse_json(metajson, video_id, fatal=False)
+            else:
+                self.report_warning("Could not fix metadata JSON", video_id)
+
+        livestream_title = get_element_by_class(
+            'unpublished-info-item future-event-title', page)
+
+        duration = None
+        thumbnail = None
+        if metad:
+            title = metad.get('name')
+            description = desc_text
+            timestamp = unified_timestamp(metad.get('uploadDate'))
+            thumbnail = url_or_none(metad.get('thumbnailUrl'))
+        elif token2:
+            base_url = self._search_regex(r'(https?://[^/]+)/', url, 'base_url')
+            data = {
+                "version": "03",
+                "action": "getAsset",
+                "organizationId": organization_id,
+                "assetId": video_id,
+                "languageId": "en_US",
+                "userId": "0",
+                "token": token2,
+            }
+            metad = self._download_json(base_url + self._API2_PATH, video_id,
+                                        data=urlencode_postdata(data))
+            title = metad.get('name')
+            description = metad.get('description')
+            timestamp = int_or_none(metad.get('date'), scale=1000)
+            duration = int_or_none(metad.get('duration'))
+            thumbnail = url_or_none(metad.get('thumbnailMedium'))
+        elif livestream_title:  # Recorded livestream
+            title = livestream_title
+            description = get_element_by_class(
+                'unpublished-info-item future-event-description', page)
+            timestamp = int_or_none(self._search_regex(
+                r"var startEvent\s*=\s*(\d+);", page, "uploadDate",
+                fatal=False), scale=1000)
+        else:
+            self.report_warning("Could not extract metadata", video_id)
+            description = None
+            timestamp = None
+
+        title = title if title else video_id
+        description = clean_html(description)
+
+        data = {
+            "version": "03",
+            "action": "getAssetPlaybackUrls",
+            "organizationId": organization_id,
+            "assetId": video_id,
+            "token": token,
+        }
+        jsond = self._download_json(api_base, video_id,
+                                    data=urlencode_postdata(data))
+
+        if thumbnail is None:
+            thumbnail = url_or_none(jsond.get('thumbnail'))
+
+        formats = []
+        for item in jsond.get('urls', []):
+            video_url = url_or_none(item.get('url'))
+            ext = determine_ext(video_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, video_id, 'mp4',
+                    entry_protocol='m3u8_native', m3u8_id='hls',
+                    fatal=False))
+            else:
+                fd = {'url': video_url}
+                fmt = item.get('name')
+                if fmt:
+                    fd['format'] = fmt
+                    fd.update(parse_resolution(fmt))
+                    mo = re.search(r'\((\d+)\s*kbps\)\s*\+\s*(\d+)\s*kbps', fmt)
+                    if mo:
+                        fd['vbr'] = int_or_none(mo.group(1))
+                        fd['abr'] = int_or_none(mo.group(2))
+                    else:
+                        fd['tbr'] = parse_bitrate(fmt)
+                fmt_id = item.get('id')
+                if fmt_id:
+                    fd['format_id'] = str(fmt_id)
+                formats.append(fd)
+
+        for item in jsond.get('audio_urls', []):
+            fmt = item.get('name')
+            mo = re.match(r'.*\((\d+)k\).*', fmt if fmt else '')
+            abr = int_or_none(mo.group(1)) if mo else None
+            fd = {
+                'format': fmt,
+                'format_id': 'audio',
+                'url': url_or_none(item.get('url')),
+                'vcodec': 'none',
+            }
+            if abr:
+                fd['abr'] = abr
+            formats.append(fd)
+
+        subtitles = {}
+        for sub in jsond.get('subtitles', []):
+            scode, sdesc, surl = sub
+            lang = sdesc.split(' ')[0]
+            lang = lang[:-1] if lang.endswith(':') else lang
+            subtitles[lang] = [{"url": url_or_none(surl)}]
+
+        info = {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'formats': formats,
+        }
+        if duration:
+            info['duration'] = duration
+        if subtitles:
+            info['subtitles'] = subtitles
+
+        return info

youtube_dl/extractor/youtube.py

@@ -1659,17 +1659,46 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
     def _extract_n_function_name(self, jscode):
         func_name, idx = self._search_regex(
             # new: (b=String.fromCharCode(110),c=a.get(b))&&c=nfunc[idx](c)
-            # or:  (b="nn"[+a.D],c=a.get(b))&&(c=nfunc[idx](c)s
-            # old: .get("n"))&&(b=nfunc[idx](b)
-            # older: .get("n"))&&(b=nfunc(b)
+            # or:  (b="nn"[+a.D],c=a.get(b))&&(c=nfunc[idx](c)
+            # or:  (PL(a),b=a.j.n||null)&&(b=nfunc[idx](b)
+            # or:  (b="nn"[+a.D],vL(a),c=a.j[b]||null)&&(c=narray[idx](c),a.set(b,c),narray.length||nfunc("")
+            # old: (b=a.get("n"))&&(b=nfunc[idx](b)(?P<c>[a-z])\s*=\s*[a-z]\s*
+            # older: (b=a.get("n"))&&(b=nfunc(b)
             r'''(?x)
-                (?:\(\s*(?P<b>[a-z])\s*=\s*(?:
-                    String\s*\.\s*fromCharCode\s*\(\s*110\s*\)|
-                    "n+"\[\s*\+?s*[\w$.]+\s*]
-                )\s*,(?P<c>[a-z])\s*=\s*[a-z]\s*)?
-                \.\s*get\s*\(\s*(?(b)(?P=b)|"n{1,2}")(?:\s*\)){2}\s*&&\s*\(\s*(?(c)(?P=c)|b)\s*=\s*
-                (?P<nfunc>[a-zA-Z_$][\w$]*)(?:\s*\[(?P<idx>\d+)\])?\s*\(\s*[\w$]+\s*\)
-            ''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
+                \((?:[\w$()\s]+,)*?\s*      # (
+                (?P<b>[a-z])\s*=\s*         # b=
+                (?:
+                    (?:                     # expect ,c=a.get(b) (etc)
+                        String\s*\.\s*fromCharCode\s*\(\s*110\s*\)|
+                        "n+"\[\s*\+?s*[\w$.]+\s*]
+                    )\s*(?:,[\w$()\s]+(?=,))*|
+                       (?P<old>[\w$]+)      # a (old[er])
+                   )\s*
+                   (?(old)
+                                            # b.get("n")
+                       (?:\.\s*[\w$]+\s*|\[\s*[\w$]+\s*]\s*)*?
+                       (?:\.\s*n|\[\s*"n"\s*]|\.\s*get\s*\(\s*"n"\s*\))
+                       |                    # ,c=a.get(b)
+                       ,\s*(?P<c>[a-z])\s*=\s*[a-z]\s*
+                       (?:\.\s*[\w$]+\s*|\[\s*[\w$]+\s*]\s*)*?
+                       (?:\[\s*(?P=b)\s*]|\.\s*get\s*\(\s*(?P=b)\s*\))
+                   )
+                                            # interstitial junk
+                   \s*(?:\|\|\s*null\s*)?(?:\)\s*)?&&\s*(?:\(\s*)?
+               (?(c)(?P=c)|(?P=b))\s*=\s*   # [c|b]=
+                                            # nfunc|nfunc[idx]
+                   (?P<nfunc>[a-zA-Z_$][\w$]*)(?:\s*\[(?P<idx>\d+)\])?\s*\(\s*[\w$]+\s*\)
+            ''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx'),
+            default=(None, None))
+        # thx bashonly: yt-dlp/yt-dlp/pull/10611
+        if not func_name:
+            self.report_warning('Falling back to generic n function search')
+            return self._search_regex(
+                r'''(?xs)
+                    (?:(?<=[^\w$])|^)       # instead of \b, which ignores $
+                    (?P<name>(?!\d)[a-zA-Z\d_$]+)\s*=\s*function\((?!\d)[a-zA-Z\d_$]+\)
+                    \s*\{(?:(?!};).)+?["']enhanced_except_
+                ''', jscode, 'Initial JS player n function name', group='name')
         if not idx:
             return func_name
 

youtube_dl/jsinterp.py

@@ -925,9 +925,16 @@ class JSInterpreter(object):
                     obj.reverse()
                     return obj
                 elif member == 'slice':
-                    assertion(isinstance(obj, list), 'must be applied on a list')
-                    assertion(len(argvals) == 1, 'takes exactly one argument')
-                    return obj[argvals[0]:]
+                    assertion(isinstance(obj, (list, compat_str)), 'must be applied on a list or string')
+                    # From [1]:
+                    # .slice() - like [:]
+                    # .slice(n) - like [n:] (not [slice(n)]
+                    # .slice(m, n) - like [m:n] or [slice(m, n)]
+                    # [1] https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Array/slice
+                    assertion(len(argvals) <= 2, 'takes between 0 and 2 arguments')
+                    if len(argvals) < 2:
+                        argvals += (None,)
+                    return obj[slice(*argvals)]
                 elif member == 'splice':
                     assertion(isinstance(obj, list), 'must be applied on a list')
                     assertion(argvals, 'takes one or more arguments')