[ruutu] Improve, make more robust and fix python 2.6 support
authorSergey M․ <dstftw@gmail.com>
Sat, 6 Jun 2015 23:37:29 +0000 (05:37 +0600)
committerSergey M․ <dstftw@gmail.com>
Sat, 6 Jun 2015 23:37:29 +0000 (05:37 +0600)
youtube_dl/extractor/ruutu.py

index 59e0b12fda155a19a80b14420d5bbcf1e2315d04..4e22628d031bc462f66394384f2d398465105baf 100644 (file)
@@ -3,88 +3,117 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..compat import compat_urllib_parse_urlparse
-import re
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    xpath_text,
+)
 
 
 class RuutuIE(InfoExtractor):
-    _VALID_URL = r'http://(www\.)?ruutu\.fi/ohjelmat/(?:[^/]+/)?(?P<id>.*)$'
+    _VALID_URL = r'http://(?:www\.)?ruutu\.fi/ohjelmat/(?:[^/?#]+/)*(?P<id>[^/?#]+)'
     _TESTS = [
         {
             'url': 'http://www.ruutu.fi/ohjelmat/oletko-aina-halunnut-tietaa-mita-tapahtuu-vain-hetki-ennen-lahetysta-nyt-se-selvisi',
             'md5': 'ab2093f39be1ca8581963451b3c0234f',
             'info_dict': {
-                'id': 'oletko-aina-halunnut-tietaa-mita-tapahtuu-vain-hetki-ennen-lahetysta-nyt-se-selvisi',
+                'id': '2058907',
+                'display_id': 'oletko-aina-halunnut-tietaa-mita-tapahtuu-vain-hetki-ennen-lahetysta-nyt-se-selvisi',
                 'ext': 'mp4',
                 'title': 'Oletko aina halunnut tietää mitä tapahtuu vain hetki ennen lähetystä? - Nyt se selvisi!',
-                'description': 'Toinen toistaan huikeampia ohjelmaideoita ja täysin päätöntä sekoilua? No sitä juuri nimenomaan. Metro Helsingin Iltapäivän vieraaksi saapui Tuomas Kauhanen ja he Petra Kalliomaan kanssa keskustelivat hieman ennen lähetyksen alkua, mutta kamerat olivatkin jo päällä.',
+                'description': 'md5:cfc6ccf0e57a814360df464a91ff67d6',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'duration': 114,
+                'age_limit': 0,
             },
-            'params': {
-                'format': 'http-1000',
-            }
         },
         {
             'url': 'http://www.ruutu.fi/ohjelmat/superpesis/superpesis-katso-koko-kausi-ruudussa',
             'md5': '065a10ae4d5b8cfd9d0c3d332465e3d9',
             'info_dict': {
-                'id': 'superpesis-katso-koko-kausi-ruudussa',
+                'id': '2057306',
+                'display_id': 'superpesis-katso-koko-kausi-ruudussa',
                 'ext': 'mp4',
                 'title': 'Superpesis: katso koko kausi Ruudussa',
-                'description': 'Huippujännittävän Superpesiksen suoria ottelulähetyksiä seurataan Ruudussa kauden alusta viimeiseen finaaliin asti. Katso lisätiedot osoitteesta ruutu.fi/superpesis.',
+                'description': 'md5:44c44a99fdbe5b380ab74ebd75f0af77',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'duration': 40,
+                'age_limit': 0,
             },
-            'params': {
-                'format': 'http-1000',
-            }
         },
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        display_id = self._match_id(url)
 
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        media_id = self._html_search_regex(r'data-media-id="(\d+)"', webpage, 'media_id')
-        media_json = self._parse_json(self._search_regex(r'jQuery.extend\([^,]+, (.*)\);', webpage, 'media_data'), video_id)
-        xml_url = media_json['ruutuplayer']['xmlUrl'].replace('{ID}', media_id)
-        media_xml = self._download_xml(xml_url, media_id)
+        webpage = self._download_webpage(url, display_id)
+
+        video_id = self._search_regex(
+            r'data-media-id="(\d+)"', webpage, 'media id')
+
+        video_xml_url = None
+
+        media_data = self._search_regex(
+            r'jQuery\.extend\([^,]+,\s*(.+?)\);', webpage,
+            'media data', default=None)
+        if media_data:
+            media_json = self._parse_json(media_data, display_id, fatal=False)
+            if media_json:
+                xml_url = media_json.get('ruutuplayer', {}).get('xmlUrl')
+                if xml_url:
+                    video_xml_url = xml_url.replace('{ID}', video_id)
+
+        if not video_xml_url:
+            video_xml_url = 'http://gatling.ruutu.fi/media-xml-cache?id=%s' % video_id
+
+        video_xml = self._download_xml(video_xml_url, video_id)
 
         formats = []
-        parsed_urls = []
-        for fmt in media_xml.findall('.//Clip//'):
-            url = fmt.text
-            if not fmt.tag.endswith('File') or url in parsed_urls or \
-                    'NOT_USED' in url:
-                continue
+        processed_urls = []
+
+        def extract_formats(node):
+            for child in node:
+                if child.tag.endswith('Files'):
+                    extract_formats(child)
+                elif child.tag.endswith('File'):
+                    video_url = child.text
+                    if not video_url or video_url in processed_urls or 'NOT_USED' in video_url:
+                        return
+                    processed_urls.append(video_url)
+                    ext = determine_ext(video_url)
+                    if ext == 'm3u8':
+                        formats.extend(self._extract_m3u8_formats(
+                            video_url, video_id, 'mp4', m3u8_id='hls'))
+                    elif ext == 'f4m':
+                        formats.extend(self._extract_f4m_formats(
+                            video_url, video_id, f4m_id='hds'))
+                    else:
+                        proto = compat_urllib_parse_urlparse(video_url).scheme
+                        if not child.tag.startswith('HTTP') and proto != 'rtmp':
+                            continue
+                        preference = -1 if proto == 'rtmp' else 1
+                        label = child.get('label')
+                        tbr = int_or_none(child.get('bitrate'))
+                        width, height = [int_or_none(x) for x in child.get('resolution', '').split('x')]
+                        formats.append({
+                            'format_id': '%s-%s' % (proto, label if label else tbr),
+                            'url': video_url,
+                            'width': width,
+                            'height': height,
+                            'tbr': tbr,
+                            'preference': preference,
+                        })
 
-            if url.endswith('m3u8'):
-                formats.extend(self._extract_m3u8_formats(url, media_id, m3u8_id='hls'))
-                parsed_urls.append(url)
-            elif url.endswith('f4m'):
-                formats.extend(self._extract_f4m_formats(url, media_id, f4m_id='hds'))
-                parsed_urls.append(url)
-            else:
-                if not fmt.tag.startswith('HTTP'):
-                    continue
-                proto = compat_urllib_parse_urlparse(url).scheme
-                width_str, height_str = fmt.get('resolution').split('x')
-                tbr = int(fmt.get('bitrate', 0))
-                formats.append({
-                    'format_id': '%s-%d' % (proto, tbr),
-                    'url': url,
-                    'width': int(width_str),
-                    'height': int(height_str),
-                    'tbr': tbr,
-                    'ext': url.rsplit('.', 1)[-1],
-                    'live': True,
-                    'protocol': proto,
-                })
+        extract_formats(video_xml.find('./Clip'))
         self._sort_formats(formats)
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'title': self._og_search_title(webpage),
-            'formats': formats,
             'description': self._og_search_description(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
-            'duration': int(media_xml.find('.//Runtime').text),
-            'age_limit': int(media_xml.find('.//AgeLimit').text),
+            'duration': int_or_none(xpath_text(video_xml, './/Runtime', 'duration')),
+            'age_limit': int_or_none(xpath_text(video_xml, './/AgeLimit', 'age limit')),
+            'formats': formats,
         }