From 65f4c1de3d442a49367597a80687fddcf3d142a2 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Sun, 9 Oct 2016 18:58:15 +0800 Subject: [PATCH] [allocine] Fix extraction (closes #10860) I change the URL of the third test case, because now the original URL does not contain a video anymore, and there's no easy to get the real URL from the /film/ one. --- ChangeLog | 1 + youtube_dl/extractor/allocine.py | 57 ++++++++++++-------------------- 2 files changed, 22 insertions(+), 36 deletions(-) diff --git a/ChangeLog b/ChangeLog index f74c6b5a4..6c6053a2a 100644 --- a/ChangeLog +++ b/ChangeLog @@ -4,6 +4,7 @@ Core * [Makefile] Support for GNU make < 4 is fixed; BSD make dropped (#9387) Extractors +* [allocine] Fix extraction (#10860) + [nextmedia] Recognize action news on AppleDaily diff --git a/youtube_dl/extractor/allocine.py b/youtube_dl/extractor/allocine.py index 7d280d871..b292ffdd9 100644 --- a/youtube_dl/extractor/allocine.py +++ b/youtube_dl/extractor/allocine.py @@ -1,29 +1,25 @@ # coding: utf-8 from __future__ import unicode_literals -import re -import json - from .common import InfoExtractor -from ..compat import compat_str from ..utils import ( qualities, - unescapeHTML, - xpath_element, + url_basename, ) class AllocineIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?allocine\.fr/(?Particle|video|film)/(fichearticle_gen_carticle=|player_gen_cmedia=|fichefilm_gen_cfilm=|video-)(?P[0-9]+)(?:\.html)?' + _VALID_URL = r'https?://(?:www\.)?allocine\.fr/(?:article|video|film)/(?:fichearticle_gen_carticle=|player_gen_cmedia=|fichefilm_gen_cfilm=|video-)(?P[0-9]+)(?:\.html)?' _TESTS = [{ 'url': 'http://www.allocine.fr/article/fichearticle_gen_carticle=18635087.html', 'md5': '0c9fcf59a841f65635fa300ac43d8269', 'info_dict': { 'id': '19546517', + 'display_id': '18635087', 'ext': 'mp4', 'title': 'Astérix - Le Domaine des Dieux Teaser VF', - 'description': 'md5:abcd09ce503c6560512c14ebfdb720d2', + 'description': 'md5:4a754271d9c6f16c72629a8a993ee884', 'thumbnail': 're:http://.*\.jpg', }, }, { @@ -31,19 +27,21 @@ class AllocineIE(InfoExtractor): 'md5': 'd0cdce5d2b9522ce279fdfec07ff16e0', 'info_dict': { 'id': '19540403', + 'display_id': '19540403', 'ext': 'mp4', 'title': 'Planes 2 Bande-annonce VF', 'description': 'Regardez la bande annonce du film Planes 2 (Planes 2 Bande-annonce VF). Planes 2, un film de Roberts Gannaway', 'thumbnail': 're:http://.*\.jpg', }, }, { - 'url': 'http://www.allocine.fr/film/fichefilm_gen_cfilm=181290.html', + 'url': 'http://www.allocine.fr/video/player_gen_cmedia=19544709&cfilm=181290.html', 'md5': '101250fb127ef9ca3d73186ff22a47ce', 'info_dict': { 'id': '19544709', + 'display_id': '19544709', 'ext': 'mp4', 'title': 'Dragons 2 - Bande annonce finale VF', - 'description': 'md5:601d15393ac40f249648ef000720e7e3', + 'description': 'md5:6cdd2d7c2687d4c6aafe80a35e17267a', 'thumbnail': 're:http://.*\.jpg', }, }, { @@ -52,43 +50,30 @@ class AllocineIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - typ = mobj.group('typ') - display_id = mobj.group('id') + display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - if typ == 'film': - video_id = self._search_regex(r'href="/video/player_gen_cmedia=([0-9]+).+"', webpage, 'video id') - else: - player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player', default=None) - if player: - player_data = json.loads(player) - video_id = compat_str(player_data['refMedia']) - else: - model = self._search_regex(r'data-model="([^"]+)">', webpage, 'data model') - model_data = self._parse_json(unescapeHTML(model), display_id) - video_id = compat_str(model_data['id']) + model = self._html_search_regex( + r'data-model="([^"]+)"', webpage, 'data model') + model_data = self._parse_json(model, display_id) - xml = self._download_xml('http://www.allocine.fr/ws/AcVisiondataV4.ashx?media=%s' % video_id, display_id) - - video = xpath_element(xml, './/AcVisionVideo').attrib quality = qualities(['ld', 'md', 'hd']) formats = [] - for k, v in video.items(): - if re.match(r'.+_path', k): - format_id = k.split('_')[0] - formats.append({ - 'format_id': format_id, - 'quality': quality(format_id), - 'url': v, - }) + for video_url in model_data['sources'].values(): + video_id, format_id = url_basename(video_url).split('_')[:2] + formats.append({ + 'format_id': format_id, + 'quality': quality(format_id), + 'url': video_url, + }) self._sort_formats(formats) return { 'id': video_id, - 'title': video['videoTitle'], + 'display_id': display_id, + 'title': model_data['title'], 'thumbnail': self._og_search_thumbnail(webpage), 'formats': formats, 'description': self._og_search_description(webpage),