From 4b7b839f24c6e95a4c1047de1a0a5194ef7f8fce Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Sat, 12 Oct 2013 22:21:23 +0200 Subject: [PATCH] Add an extractor for rottentomatoes.com and improve InternetVideoArchiveIE to get the best quality --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/internetvideoarchive.py | 16 ++++++++++++++++ youtube_dl/extractor/rottentomatoes.py | 16 ++++++++++++++++ youtube_dl/extractor/videodetective.py | 2 +- 4 files changed, 34 insertions(+), 1 deletion(-) create mode 100644 youtube_dl/extractor/rottentomatoes.py diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 0f38bdd54..9dc9651ad 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -94,6 +94,7 @@ from .redtube import RedTubeIE from .ringtv import RingTVIE from .ro220 import Ro220IE +from .rottentomatoes import RottenTomatoesIE from .roxwel import RoxwelIE from .rtlnow import RTLnowIE from .sina import SinaIE diff --git a/youtube_dl/extractor/internetvideoarchive.py b/youtube_dl/extractor/internetvideoarchive.py index 52e3f9eec..5986459d6 100644 --- a/youtube_dl/extractor/internetvideoarchive.py +++ b/youtube_dl/extractor/internetvideoarchive.py @@ -4,6 +4,7 @@ from .common import InfoExtractor from ..utils import ( compat_urlparse, + compat_urllib_parse, xpath_with_ns, determine_ext, ) @@ -26,6 +27,16 @@ class InternetVideoArchiveIE(InfoExtractor): def _build_url(query): return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query + @staticmethod + def _clean_query(query): + NEEDED_ARGS = ['publishedid', 'customerid'] + query_dic = compat_urlparse.parse_qs(query) + cleaned_dic = dict((k,v[0]) for (k,v) in query_dic.items() if k in NEEDED_ARGS) + # Other player ids return m3u8 urls + cleaned_dic['playerid'] = '247' + cleaned_dic['videokbrate'] = '100000' + return compat_urllib_parse.urlencode(cleaned_dic) + def _real_extract(self, url): query = compat_urlparse.urlparse(url).query query_dic = compat_urlparse.parse_qs(query) @@ -37,6 +48,11 @@ def _real_extract(self, url): flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8')) file_url = flashconfiguration.find('file').text file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx') + # Replace some of the parameters in the query to get the best quality + # and http links (no m3u8 manifests) + file_url = re.sub(r'(?<=\?)(.+)$', + lambda m: self._clean_query(m.group()), + file_url) info_xml = self._download_webpage(file_url, video_id, u'Downloading video info') info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8')) diff --git a/youtube_dl/extractor/rottentomatoes.py b/youtube_dl/extractor/rottentomatoes.py new file mode 100644 index 000000000..c79c39413 --- /dev/null +++ b/youtube_dl/extractor/rottentomatoes.py @@ -0,0 +1,16 @@ +from .videodetective import VideoDetectiveIE + + +# It just uses the same method as videodetective.com, +# the internetvideoarchive.com is extracted from the og:video property +class RottenTomatoesIE(VideoDetectiveIE): + _VALID_URL = r'https?://www\.rottentomatoes\.com/m/[^/]+/trailers/(?P\d+)' + + _TEST = { + u'url': u'http://www.rottentomatoes.com/m/toy_story_3/trailers/11028566/', + u'file': '613340.mp4', + u'info_dict': { + u'title': u'TOY STORY 3', + u'description': u'From the creators of the beloved TOY STORY films, comes a story that will reunite the gang in a whole new way.', + }, + } diff --git a/youtube_dl/extractor/videodetective.py b/youtube_dl/extractor/videodetective.py index 265dd5b91..d89f84094 100644 --- a/youtube_dl/extractor/videodetective.py +++ b/youtube_dl/extractor/videodetective.py @@ -16,7 +16,7 @@ class VideoDetectiveIE(InfoExtractor): u'info_dict': { u'title': u'KICK-ASS 2', u'description': u'md5:65ba37ad619165afac7d432eaded6013', - u'duration': 135, + u'duration': 138, }, }