From 24eb7c257807aceef713f6d01824cf600a9d6232 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Sat, 29 Apr 2017 15:55:08 +0800 Subject: [PATCH] [xtube] Fix extraction with non-standard JSON 'sources' Closes #12734 Thanks @paulguy for the fix! --- ChangeLog | 6 ++++++ youtube_dl/extractor/xtube.py | 23 +++++++++++++++++++++-- 2 files changed, 27 insertions(+), 2 deletions(-) diff --git a/ChangeLog b/ChangeLog index cd49ac42d..aa7543325 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,9 @@ +version + +Extractors +* [xtube] Fix extraction for older FLV videos (#12734) + + version 2017.04.28 Core diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py index 5584674a0..bea9b87ad 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -6,6 +6,7 @@ import re from .common import InfoExtractor from ..utils import ( int_or_none, + js_to_json, orderedSet, parse_duration, sanitized_Request, @@ -37,6 +38,22 @@ class XTubeIE(InfoExtractor): 'comment_count': int, 'age_limit': 18, } + }, { + # FLV videos with duplicated formats + 'url': 'http://www.xtube.com/video-watch/A-Super-Run-Part-1-YT-9299752', + 'md5': 'a406963eb349dd43692ec54631efd88b', + 'info_dict': { + 'id': '9299752', + 'display_id': 'A-Super-Run-Part-1-YT', + 'ext': 'flv', + 'title': 'A Super Run - Part 1 (YT)', + 'description': 'md5:ca0d47afff4a9b2942e4b41aa970fd93', + 'uploader': 'tshirtguy59', + 'duration': 579, + 'view_count': int, + 'comment_count': int, + 'age_limit': 18, + }, }, { # new URL schema 'url': 'http://www.xtube.com/video-watch/strange-erotica-625837', @@ -68,8 +85,9 @@ class XTubeIE(InfoExtractor): }) sources = self._parse_json(self._search_regex( - r'(["\'])sources\1\s*:\s*(?P{.+?}),', - webpage, 'sources', group='sources'), video_id) + r'(["\'])?sources\1?\s*:\s*(?P{.+?}),', + webpage, 'sources', group='sources'), video_id, + transform_source=js_to_json) formats = [] for format_id, format_url in sources.items(): @@ -78,6 +96,7 @@ class XTubeIE(InfoExtractor): 'format_id': format_id, 'height': int_or_none(format_id), }) + self._remove_duplicate_formats(formats) self._sort_formats(formats) title = self._search_regex(