[dramefever] Improve and simplify

youtube_dl/extractor/dramafever.py

@@ -1,104 +1,111 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
+import itertools
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_HTTPError,
+    compat_urlparse,
+)
+from ..utils import (
+    ExtractorError,
+    clean_html,
+    determine_ext,
+    int_or_none,
+    parse_iso8601,
+)
 
 
 class DramaFeverIE(InfoExtractor):
     IE_NAME = 'dramafever'
-    _VALID_URL = r'^https?://(?:www\.)?dramafever\.com/drama/(?P<id>[0-9]+/[0-9]+)/'
-    _TESTS = [{
+    _VALID_URL = r'https?://(?:www\.)?dramafever\.com/drama/(?P<id>[0-9]+/[0-9]+)'
+    _TEST = {
         'url': 'http://www.dramafever.com/drama/4512/1/Cooking_with_Shin/',
         'info_dict': {
             'id': '4512.1',
             'ext': 'flv',
             'title': 'Cooking with Shin 4512.1',
+            'description': 'md5:a8eec7942e1664a6896fcd5e1287bfd0',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'timestamp': 1404336058,
             'upload_date': '20140702',
-            'description': 'Served at all special occasions and featured in the hit drama Heirs, Shin cooks Red Bean Rice.',
+            'duration': 343,
         }
-    }]
+    }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url).replace("/", ".")
-
-        consumer_secret = self._get_consumer_secret(video_id)
+        video_id = self._match_id(url).replace('/', '.')
 
-        ep_json = self._download_json(
-            "http://www.dramafever.com/amp/episode/feed.json?guid=%s" % video_id,
-            video_id, note='Downloading episode metadata',
-            errnote="Video may not be available for your location")["channel"]["item"]
+        try:
+            feed = self._download_json(
+                'http://www.dramafever.com/amp/episode/feed.json?guid=%s' % video_id,
+                video_id, 'Downloading episode JSON')['channel']['item']
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError):
+                raise ExtractorError(
+                    'Currently unavailable in your country.', expected=True)
+            raise
 
-        title = ep_json["media-group"]["media-title"]
-        description = ep_json["media-group"]["media-description"]
-        thumbnail = ep_json["media-group"]["media-thumbnail"]["@attributes"]["url"]
-        duration = int(ep_json["media-group"]["media-content"][0]["@attributes"]["duration"])
-        mobj = re.match(r"([0-9]{4})-([0-9]{2})-([0-9]{2})", ep_json["pubDate"])
-        upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3) if mobj is not None else None
+        media_group = feed.get('media-group', {})
 
         formats = []
-        for vid_format in ep_json["media-group"]["media-content"]:
-            src = vid_format["@attributes"]["url"]
-            if '.f4m' in src:
-                formats.extend(self._extract_f4m_formats(src, video_id))
-
+        for media_content in media_group['media-content']:
+            src = media_content.get('@attributes', {}).get('url')
+            if not src:
+                continue
+            ext = determine_ext(src)
+            if ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(
+                    src, video_id, f4m_id='hds'))
+            elif ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    src, video_id, 'mp4', m3u8_id='hls'))
+            else:
+                formats.append({
+                    'url': src,
+                })
         self._sort_formats(formats)
-        video_subtitles = self.extract_subtitles(video_id, consumer_secret)
+
+        title = media_group.get('media-title')
+        description = media_group.get('media-description')
+        duration = int_or_none(media_group['media-content'][0].get('@attributes', {}).get('duration'))
+        thumbnail = self._proto_relative_url(
+            media_group.get('media-thumbnail', {}).get('@attributes', {}).get('url'))
+        timestamp = parse_iso8601(feed.get('pubDate'), ' ')
+
+        subtitles = {}
+        for media_subtitle in media_group.get('media-subTitle', []):
+            lang = media_subtitle.get('@attributes', {}).get('lang')
+            href = media_subtitle.get('@attributes', {}).get('href')
+            if not lang or not href:
+                continue
+            subtitles[lang] = [{
+                'ext': 'ttml',
+                'url': href,
+            }]
 
         return {
             'id': video_id,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
-            'upload_date': upload_date,
+            'timestamp': timestamp,
             'duration': duration,
             'formats': formats,
-            'subtitles': video_subtitles,
+            'subtitles': subtitles,
         }
 
-    def _get_consumer_secret(self, video_id):
-        df_js = self._download_webpage(
-            "http://www.dramafever.com/static/126960d/v2/js/plugins/jquery.threadedcomments.js", video_id)
-        return self._search_regex(r"'cs': '([0-9a-zA-Z]+)'", df_js, "cs")
-
-    def _get_episodes(self, series_id, consumer_secret, episode_filter=None):
-        _PAGE_SIZE = 60
-
-        curr_page = 1
-        max_pages = curr_page + 1
-        results = []
-        while max_pages >= curr_page:
-            page_url = "http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_size=%d&page_number=%d" % \
-                       (consumer_secret, series_id, _PAGE_SIZE, curr_page)
-            series = self._download_json(
-                page_url, series_id, note="Downloading series json page #%d" % curr_page)
-            max_pages = series['num_pages']
-            results.extend([ep for ep in series['value'] if episode_filter is None or episode_filter(ep)])
-            curr_page += 1
-        return results
-
-    def _get_subtitles(self, video_id, consumer_secret):
-
-        res = None
-        info = self._get_episodes(
-            video_id.split(".")[0], consumer_secret,
-            episode_filter=lambda x: x['guid'] == video_id)
-
-        if len(info) == 1 and info[0]['subfile'] != '':
-            res = {'en': [{'url': info[0]['subfile'], 'ext': 'srt'}]}
-        return res
-
-
-class DramaFeverSeriesIE(DramaFeverIE):
+
+class DramaFeverSeriesIE(InfoExtractor):
     IE_NAME = 'dramafever:series'
-    _VALID_URL = r'^https?://(?:www\.)?dramafever\.com/drama/(?P<id>[0-9]+)/\d*[a-zA-Z_][a-zA-Z0-9_]*/'
+    _VALID_URL = r'https?://(?:www\.)?dramafever\.com/drama/(?P<id>[0-9]+)(?:/(?:(?!\d).+)?)?$'
     _TESTS = [{
         'url': 'http://www.dramafever.com/drama/4512/Cooking_with_Shin/',
         'info_dict': {
             'id': '4512',
             'title': 'Cooking with Shin',
-            'description': 'Professional chef and cooking instructor Shin Kim takes some of the delicious dishes featured in your favorite dramas and shows you how to make them right at home.',
+            'description': 'md5:84a3f26e3cdc3fb7f500211b3593b5c1',
         },
         'playlist_count': 4,
     }, {
@@ -106,25 +113,48 @@ class DramaFeverSeriesIE(DramaFeverIE):
         'info_dict': {
             'id': '124',
             'title': 'IRIS',
-            'description': 'Lee Byung Hun and Kim Tae Hee star in this powerhouse drama and ratings megahit of action, intrigue and romance.',
+            'description': 'md5:b3a30e587cf20c59bd1c01ec0ee1b862',
         },
         'playlist_count': 20,
     }]
 
+    _CONSUMER_SECRET = 'DA59dtVXYLxajktV'
+    _PAGE_SIZE = 5  # max is 60 (see http://api.drama9.com/#get--api-4-episode-series-)
+
+    def _get_consumer_secret(self, video_id):
+        mainjs = self._download_webpage(
+            'http://www.dramafever.com/static/51afe95/df2014/scripts/main.js',
+            video_id, 'Downloading main.js', fatal=False)
+        if not mainjs:
+            return self._CONSUMER_SECRET
+        return self._search_regex(
+            r"var\s+cs\s*=\s*'([^']+)'", mainjs,
+            'consumer secret', default=self._CONSUMER_SECRET)
+
     def _real_extract(self, url):
         series_id = self._match_id(url)
+
         consumer_secret = self._get_consumer_secret(series_id)
 
-        series_json = self._download_json(
-            "http://www.dramafever.com/api/4/series/query/?cs=%s&series_id=%s" % (consumer_secret, series_id),
-            series_id, note='Downloading series metadata')["series"][series_id]
+        series = self._download_json(
+            'http://www.dramafever.com/api/4/series/query/?cs=%s&series_id=%s'
+            % (consumer_secret, series_id),
+            series_id, 'Downloading series JSON')['series'][series_id]
 
-        title = series_json["name"]
-        description = series_json["description_short"]
+        title = clean_html(series['name'])
+        description = clean_html(series.get('description') or series.get('description_short'))
 
-        episodes = self._get_episodes(series_id, consumer_secret)
         entries = []
-        for ep in episodes:
-            entries.append(self.url_result(
-                'http://www.dramafever.com%s' % ep['episode_url'], 'DramaFever', ep['guid']))
+        for page_num in itertools.count(1):
+            episodes = self._download_json(
+                'http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_size=%d&page_number=%d'
+                % (consumer_secret, series_id, self._PAGE_SIZE, page_num),
+                series_id, 'Downloading episodes JSON page #%d' % page_num)
+            for episode in episodes.get('value', []):
+                entries.append(self.url_result(
+                    compat_urlparse.urljoin(url, episode['episode_url']),
+                    'DramaFever', episode.get('guid')))
+            if page_num == episodes['num_pages']:
+                break
+
         return self.playlist_result(entries, series_id, title, description)