[espn] Fix extraction (closes #11041)

This commit is contained in:
Sergey M․ 2016-11-08 00:29:12 +07:00
parent 97726317ac
commit ebc7ab1e23
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D

View File

@ -1,38 +1,117 @@
from __future__ import unicode_literals from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import remove_end from ..compat import compat_str
from ..utils import (
determine_ext,
int_or_none,
unified_timestamp,
)
class ESPNIE(InfoExtractor): class ESPNIE(InfoExtractor):
_VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/(?:[^/]+/)*(?P<id>[^/]+)' _VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/video/clip(?:\?.*?\bid=|/_/id/)(?P<id>\d+)'
_TESTS = [{ _TESTS = [{
'url': 'http://espn.go.com/video/clip?id=10365079', 'url': 'http://espn.go.com/video/clip?id=10365079',
'md5': '60e5d097a523e767d06479335d1bdc58',
'info_dict': { 'info_dict': {
'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG', 'id': '10365079',
'ext': 'mp4', 'ext': 'mp4',
'title': '30 for 30 Shorts: Judging Jewell', 'title': '30 for 30 Shorts: Judging Jewell',
'description': None, 'description': 'md5:39370c2e016cb4ecf498ffe75bef7f0f',
'timestamp': 1390936111,
'upload_date': '20140128',
}, },
'params': { 'params': {
'skip_download': True, 'skip_download': True,
}, },
'add_ie': ['OoyalaExternal'],
}, { }, {
# intl video, from http://www.espnfc.us/video/mls-highlights/150/video/2743663/must-see-moments-best-of-the-mls-season # intl video, from http://www.espnfc.us/video/mls-highlights/150/video/2743663/must-see-moments-best-of-the-mls-season
'url': 'http://espn.go.com/video/clip?id=2743663', 'url': 'http://espn.go.com/video/clip?id=2743663',
'md5': 'f4ac89b59afc7e2d7dbb049523df6768',
'info_dict': { 'info_dict': {
'id': '50NDFkeTqRHB0nXBOK-RGdSG5YQPuxHg', 'id': '2743663',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Must-See Moments: Best of the MLS season', 'title': 'Must-See Moments: Best of the MLS season',
'description': 'md5:4c2d7232beaea572632bec41004f0aeb',
'timestamp': 1449446454,
'upload_date': '20151207',
}, },
'params': { 'params': {
'skip_download': True, 'skip_download': True,
}, },
'add_ie': ['OoyalaExternal'], 'expected_warnings': ['Unable to download f4m manifest'],
}, { }, {
'url': 'http://www.espn.com/video/clip?id=10365079',
'only_matching': True,
}, {
'url': 'http://www.espn.com/video/clip/_/id/17989860',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
clip = self._download_json(
'http://api-app.espn.com/v1/video/clips/%s' % video_id,
video_id)['videos'][0]
title = clip['headline']
format_urls = set()
formats = []
def traverse_source(source, base_source_id=None):
for source_id, source in source.items():
if isinstance(source, compat_str):
extract_source(source, base_source_id)
elif isinstance(source, dict):
traverse_source(
source,
'%s-%s' % (base_source_id, source_id)
if base_source_id else source_id)
def extract_source(source_url, source_id=None):
if source_url in format_urls:
return
format_urls.add(source_url)
ext = determine_ext(source_url)
if ext == 'smil':
formats.extend(self._extract_smil_formats(
source_url, video_id, fatal=False))
elif ext == 'f4m':
formats.extend(self._extract_f4m_formats(
source_url, video_id, f4m_id=source_id, fatal=False))
elif ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
source_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id=source_id, fatal=False))
else:
formats.append({
'url': source_url,
'format_id': source_id,
})
traverse_source(clip['links']['source'])
self._sort_formats(formats)
description = clip.get('caption') or clip.get('description')
thumbnail = clip.get('thumbnail')
duration = int_or_none(clip.get('duration'))
timestamp = unified_timestamp(clip.get('originalPublishDate'))
return {
'id': video_id,
'title': title,
'description': description,
'thumbnail': thumbnail,
'timestamp': timestamp,
'duration': duration,
'formats': formats,
}
class ESPNArticleIE(InfoExtractor):
_VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/(?:[^/]+/)*(?P<id>[^/]+)'
_TESTS = [{
'url': 'https://espn.go.com/video/iframe/twitter/?cms=espn&id=10365079', 'url': 'https://espn.go.com/video/iframe/twitter/?cms=espn&id=10365079',
'only_matching': True, 'only_matching': True,
}, { }, {
@ -47,11 +126,12 @@ class ESPNIE(InfoExtractor):
}, { }, {
'url': 'http://espn.go.com/nba/playoffs/2015/story/_/id/12887571/john-wall-washington-wizards-no-swelling-left-hand-wrist-game-5-return', 'url': 'http://espn.go.com/nba/playoffs/2015/story/_/id/12887571/john-wall-washington-wizards-no-swelling-left-hand-wrist-game-5-return',
'only_matching': True, 'only_matching': True,
}, {
'url': 'http://www.espn.com/video/clip?id=10365079',
'only_matching': True,
}] }]
@classmethod
def suitable(cls, url):
return False if ESPNIE.suitable(url) else super(ESPNArticleIE, cls).suitable(url)
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
@ -61,23 +141,5 @@ def _real_extract(self, url):
r'class=(["\']).*?video-play-button.*?\1[^>]+data-id=["\'](?P<id>\d+)', r'class=(["\']).*?video-play-button.*?\1[^>]+data-id=["\'](?P<id>\d+)',
webpage, 'video id', group='id') webpage, 'video id', group='id')
cms = 'espn' return self.url_result(
if 'data-source="intl"' in webpage: 'http://espn.go.com/video/clip?id=%s' % video_id, ESPNIE.ie_key())
cms = 'intl'
player_url = 'https://espn.go.com/video/iframe/twitter/?id=%s&cms=%s' % (video_id, cms)
player = self._download_webpage(
player_url, video_id)
pcode = self._search_regex(
r'["\']pcode=([^"\']+)["\']', player, 'pcode')
title = remove_end(
self._og_search_title(webpage),
'- ESPN Video').strip()
return {
'_type': 'url_transparent',
'url': 'ooyalaexternal:%s:%s:%s' % (cms, video_id, pcode),
'ie_key': 'OoyalaExternal',
'title': title,
}