[elonet] Rewrite extractor (#3277)

Closes #2911
Authored by: Fam0r, pukkandan

yt_dlp/extractor/common.py

@@ -1297,8 +1297,8 @@ def _get_tfa_info(self, note='two-factor verification code'):
     @staticmethod
     def _og_regexes(prop):
         content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))'
-        property_re = (r'(?:name|property)=(?:\'og[:-]%(prop)s\'|"og[:-]%(prop)s"|\s*og[:-]%(prop)s\b)'
+        property_re = (r'(?:name|property)=(?:\'og%(sep)s%(prop)s\'|"og%(sep)s%(prop)s"|\s*og%(sep)s%(prop)s\b)'
-                       % {'prop': re.escape(prop)})
+                       % {'prop': re.escape(prop), 'sep': '(?:&#x3A;|[:-])'})
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
             template % (property_re, content_re),

yt_dlp/extractor/elonet.py

@@ -1,30 +1,22 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..utils import (
+from ..utils import determine_ext
-    base_url,
-    ExtractorError,
-    try_get,
-)
-from ..compat import compat_str
 
 
 class ElonetIE(InfoExtractor):
     _VALID_URL = r'https?://elonet\.finna\.fi/Record/kavi\.elonet_elokuva_(?P<id>[0-9]+)'
     _TESTS = [{
-        # m3u8 with subtitles
         'url': 'https://elonet.finna.fi/Record/kavi.elonet_elokuva_107867',
-        'md5': '8efc954b96c543711707f87de757caea',
         'info_dict': {
             'id': '107867',
             'ext': 'mp4',
             'title': 'Valkoinen peura',
-            'description': 'Valkoinen peura (1952) on Erik Blombergin ohjaama ja yhdessä Mirjami Kuosmasen kanssa käsikirjoittama tarunomainen kertomus valkoisen peuran hahmossa lii...',
+            'thumbnail': r're:^https?://elonet\.finna\.fi/Cover/Show\?id=kavi\.elonet_elokuva_107867.+',
-            'thumbnail': 'https://elonet.finna.fi/Cover/Show?id=kavi.elonet_elokuva_107867&index=0&size=large',
+            'description': 'md5:bded4201c9677fab10854884fe8f7312',
         },
+        'params': {'skip_download': 'dash'},
     }, {
         # DASH with subtitles
         'url': 'https://elonet.finna.fi/Record/kavi.elonet_elokuva_116539',
@@ -32,58 +24,45 @@ class ElonetIE(InfoExtractor):
             'id': '116539',
             'ext': 'mp4',
             'title': 'Minulla on tiikeri',
-            'description': 'Pienellä pojalla, joka asuu kerrostalossa, on kotieläimenä tiikeri. Se on kuitenkin salaisuus. Kerrostalon räpätäti on Kotilaisen täti, joka on aina vali...',
+            'thumbnail': r're:^https?://elonet\.finna\.fi/Cover/Show\?id=kavi\.elonet_elokuva_116539.+',
-            'thumbnail': 'https://elonet.finna.fi/Cover/Show?id=kavi.elonet_elokuva_116539&index=0&size=large&source=Solr',
+            'description': 'md5:5ab72b3fe76d3414e46cc8f277104419',
-        }
+        },
+        'params': {'skip_download': 'dash'},
+    }, {
+        # Page with multiple videos, download the main one
+        'url': 'https://elonet.finna.fi/Record/kavi.elonet_elokuva_117396',
+        'info_dict': {
+            'id': '117396',
+            'ext': 'mp4',
+            'title': 'Sampo',
+            'thumbnail': r're:^https?://elonet\.finna\.fi/Cover/Show\?id=kavi\.elonet_elokuva_117396.+',
+            'description': 'md5:ec69572a5b054d0ecafe8086b1fa96f7',
+        },
+        'params': {'skip_download': 'dash'},
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        title = self._html_search_regex(
+        src = self._parse_json(self._html_search_regex(
-            r'<meta .*property="og&#x3A;title" .*content="(.+?)"', webpage, 'title')
+            r'id=\'video-data\'[^>]+data-video-sources="([^"]+)"', webpage, 'json'), video_id)[0]['src']
-        description = self._html_search_regex(
+        ext = determine_ext(src)
-            r'<meta .*property="og&#x3A;description" .*content="(.+?)"', webpage, 'description')
-        thumbnail = self._html_search_regex(
-            r'<meta .*property="og&#x3A;image" .*content="(.+?)"', webpage, 'thumbnail')
 
-        json_s = self._html_search_regex(
+        if ext == 'm3u8':
-            r'data-video-sources="(.+?)"', webpage, 'json')
+            formats, subtitles = self._extract_m3u8_formats_and_subtitles(src, video_id, fatal=False)
-        src = try_get(
+        elif ext == 'mpd':
-            self._parse_json(json_s, video_id),
+            formats, subtitles = self._extract_mpd_formats_and_subtitles(src, video_id, fatal=False)
-            lambda x: x[0]["src"], compat_str)
-        formats = []
-        subtitles = {}
-        if re.search(r'\.m3u8\??', src):
-            res = self._download_webpage_handle(
-                # elonet servers have certificate problems
-                src.replace('https:', 'http:'), video_id,
-                note='Downloading m3u8 information',
-                errnote='Failed to download m3u8 information')
-            if res:
-                doc, urlh = res
-                url = urlh.geturl()
-                formats, subtitles = self._parse_m3u8_formats_and_subtitles(doc, url)
-                for f in formats:
-                    f['ext'] = 'mp4'
-        elif re.search(r'\.mpd\??', src):
-            res = self._download_xml_handle(
-                src, video_id,
-                note='Downloading MPD manifest',
-                errnote='Failed to download MPD manifest')
-            if res:
-                doc, urlh = res
-                url = base_url(urlh.geturl())
-                formats, subtitles = self._parse_mpd_formats_and_subtitles(doc, mpd_base_url=url)
         else:
-            raise ExtractorError("Unknown streaming format")
+            formats, subtitles = [], {}
+            self.raise_no_formats(f'Unknown streaming format {ext}')
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': title,
+            'title': self._og_search_title(webpage),
-            'description': description,
+            'description': self._og_search_description(webpage),
-            'thumbnail': thumbnail,
+            'thumbnail': self._og_search_thumbnail(webpage),
             'formats': formats,
             'subtitles': subtitles,
         }