[cda] Improve and simplify (Closes #8805)

youtube_dl/extractor/cda.py

@@ -12,30 +12,30 @@ from ..utils import (
 
 
 class CDAIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:(?:www|ebd)\.)?cda\.pl/(?:video|[0-9]+x[0-9]+)/(?P<id>[0-9a-z]+)'
+    _VALID_URL = r'https?://(?:(?:www\.)?cda\.pl/video|ebd\.cda\.pl/[0-9]+x[0-9]+)/(?P<id>[0-9a-z]+)'
-    _TESTS = [
+    _TESTS = [{
-        {
+        'url': 'http://www.cda.pl/video/5749950c',
-            'url': 'http://www.cda.pl/video/5749950c',
+        'md5': '6f844bf51b15f31fae165365707ae970',
-            'md5': '6f844bf51b15f31fae165365707ae970',
+        'info_dict': {
-            'info_dict': {
+            'id': '5749950c',
-                'id': '5749950c',
+            'ext': 'mp4',
-                'ext': 'mp4',
+            'height': 720,
-                'height': 720,
+            'title': 'Oto dlaczego przed zakrętem należy zwolnić.',
-                'title': 'Oto dlaczego przed zakrętem należy zwolnić.',
+            'duration': 39
-                'duration': 39
+        }
-            }
+    }, {
-        },
+        'url': 'http://www.cda.pl/video/57413289',
-        {
+        'md5': 'a88828770a8310fc00be6c95faf7f4d5',
-            'url': 'http://www.cda.pl/video/57413289',
+        'info_dict': {
-            'md5': 'a88828770a8310fc00be6c95faf7f4d5',
+            'id': '57413289',
-            'info_dict': {
+            'ext': 'mp4',
-                'id': '57413289',
+            'title': 'Lądowanie na lotnisku na Maderze',
-                'ext': 'mp4',
+            'duration': 137
-                'title': 'Lądowanie na lotnisku na Maderze',
-                'duration': 137
-            }
         }
-    ]
+    }, {
+        'url': 'http://ebd.cda.pl/0x0/5749950c',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -44,53 +44,53 @@ class CDAIE(InfoExtractor):
         if 'Ten film jest dostępny dla użytkowników premium' in webpage:
             raise ExtractorError('This video is only available for premium users.', expected=True)
 
-        title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title', fatal=False)
+        title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title')
-
-        def _get_format(page, version=''):
-            unpacked = decode_packed_codes(page)
-            duration = self._search_regex(r"duration:\\'(.+?)\\'", unpacked, 'duration', fatal=False)
-            format_id = None
-            height = None
-
-            m = re.search(r'<a data-quality="(?P<format_id>[^"]+)" href="[^"]+" class="quality-btn quality-btn-active">(?P<height>[0-9]+)p<\/a>', page)
-            if m:
-                format_id = m.group('format_id')
-                height = int(m.group('height'))
-
-            url = self._search_regex(r"url:\\'(.+?)\\'", unpacked, version + ' url', fatal=False)
-            if url is None:
-                return None
-
-            return {
-                'format_id': format_id,
-                'height': height,
-                'url': url
-            }, parse_duration(duration)
 
         formats = []
 
-        format_desc, duration = _get_format(webpage) or (None, None)
+        info_dict = {
-        if format_desc is not None:
+            'id': video_id,
-            formats.append(format_desc)
+            'title': title,
+            'formats': formats,
+            'duration': None,
+        }
 
-        pattern = re.compile(r'<a data-quality="[^"]+" href="([^"]+)" class="quality-btn">([0-9]+p)<\/a>')
+        def extract_format(page, version):
-        for version in re.findall(pattern, webpage):
+            unpacked = decode_packed_codes(page)
-            webpage = self._download_webpage(version[0], video_id, 'Downloading %s version information' % version[1], fatal=False)
+            format_url = self._search_regex(
+                r"url:\\'(.+?)\\'", unpacked, '%s url' % version, fatal=False)
+            if not format_url:
+                return
+            f = {
+                'url': format_url,
+            }
+            m = re.search(
+                r'<a[^>]+data-quality="(?P<format_id>[^"]+)"[^>]+href="[^"]+"[^>]+class="[^"]*quality-btn-active[^"]*">(?P<height>[0-9]+)p',
+                page)
+            if m:
+                f.update({
+                    'format_id': m.group('format_id'),
+                    'height': int(m.group('height')),
+                })
+            info_dict['formats'].append(f)
+            if not info_dict['duration']:
+                info_dict['duration'] = parse_duration(self._search_regex(
+                    r"duration:\\'(.+?)\\'", unpacked, 'duration', fatal=False))
+
+        extract_format(webpage, 'default')
+
+        for href, resolution in re.findall(
+                r'<a[^>]+data-quality="[^"]+"[^>]+href="([^"]+)"[^>]+class="quality-btn"[^>]*>([0-9]+p)',
+                webpage):
+            webpage = self._download_webpage(
+                href, video_id, 'Downloading %s version information' % resolution, fatal=False)
             if not webpage:
-                # Manually report warning because empty page is returned when invalid version is requested.
+                # Manually report warning because empty page is returned when
-                self.report_warning('Unable to download %s version information' % version[1])
+                # invalid version is requested.
+                self.report_warning('Unable to download %s version information' % resolution)
                 continue
-
+            extract_format(webpage, resolution)
-            format_desc, duration_ = _get_format(webpage, version[1]) or (None, None)
-            duration = duration or duration_
-            if format_desc is not None:
-                formats.append(format_desc)
 
         self._sort_formats(formats)
 
-        return {
+        return info_dict
-            'id': video_id,
-            'title': title,
-            'formats': formats,
-            'duration': duration
-        }