mirror of
https://codeberg.org/polarisfm/youtube-dl
synced 2024-11-26 02:14:32 +01:00
Prefer API to scraping HTML when possible
This commit is contained in:
geauxlo
GitHub
parent
4d44560896
commit
762fe4e670
151
youtube_dl/extractor/screencast.py
Normal file
151
youtube_dl/extractor/screencast.py
Normal file
@ -0,0 +1,151 @@
|
||||
# coding: utf-8
|
||||
from __future__ import unicode_literals
|
||||
|
||||
import re
|
||||
|
||||
from .common import InfoExtractor
|
||||
from ..compat import (
|
||||
compat_parse_qs,
|
||||
compat_urllib_request,
|
||||
)
|
||||
from ..utils import (
|
||||
ExtractorError,
|
||||
)
|
||||
|
||||
|
||||
class ScreencastIE(InfoExtractor):
|
||||
_VALID_URL = r'https?://(?:www\.)?screencast\.com/t/(?P<id>[a-zA-Z0-9]+)'
|
||||
_API_URL = 'https://www.screencast.com/api/external/oembed?url=%s&format=json'
|
||||
|
||||
_TESTS = [{
|
||||
'url': 'http://www.screencast.com/t/3ZEjQXlT',
|
||||
'md5': '917df1c13798a3e96211dd1561fded83',
|
||||
'info_dict': {
|
||||
'id': '3ZEjQXlT',
|
||||
'ext': 'm4v',
|
||||
'title': 'Color Measurement with Ocean Optics Spectrometers',
|
||||
'description': 'md5:240369cde69d8bed61349a199c5fb153',
|
||||
'thumbnail': r're:^https?://.*\.(?:gif|jpg)$',
|
||||
}
|
||||
}, {
|
||||
'url': 'http://www.screencast.com/t/V2uXehPJa1ZI',
|
||||
'md5': 'e8e4b375a7660a9e7e35c33973410d34',
|
||||
'info_dict': {
|
||||
'id': 'V2uXehPJa1ZI',
|
||||
'ext': 'mov',
|
||||
'title': 'The Amadeus Spectrometer',
|
||||
'description': 're:^In this video, our friends at.*To learn more about Amadeus, visit',
|
||||
'thumbnail': r're:^https?://.*\.(?:gif|jpg)$',
|
||||
}
|
||||
}, {
|
||||
'url': 'http://www.screencast.com/t/aAB3iowa',
|
||||
'md5': 'dedb2734ed00c9755761ccaee88527cd',
|
||||
'info_dict': {
|
||||
'id': 'aAB3iowa',
|
||||
'ext': 'mp4',
|
||||
'title': 'Google Earth Export',
|
||||
'description': 'Provides a demo of a CommunityViz export to Google Earth, one of the 3D viewing options.',
|
||||
'thumbnail': r're:^https?://.*\.(?:gif|jpg)$',
|
||||
}
|
||||
}, {
|
||||
'url': 'http://www.screencast.com/t/X3ddTrYh',
|
||||
'md5': '669ee55ff9c51988b4ebc0877cc8b159',
|
||||
'info_dict': {
|
||||
'id': 'X3ddTrYh',
|
||||
'ext': 'wmv',
|
||||
'title': 'Toolkit 6 User Group Webinar (2014-03-04) - Default Judgment and First Impression',
|
||||
'description': 'md5:7b9f393bc92af02326a5c5889639eab0',
|
||||
'thumbnail': r're:^https?://.*\.(?:gif|jpg)$',
|
||||
}
|
||||
}, {
|
||||
'url': 'http://screencast.com/t/aAB3iowa',
|
||||
'only_matching': True,
|
||||
}]
|
||||
|
||||
def _real_extract(self, url):
|
||||
video_id = self._match_id(url)
|
||||
|
||||
# The JSON given by the API has a thumbnail URL,
|
||||
# but it's inferior to the webpage's thumbnail.
|
||||
# It also has no video description, so we
|
||||
# definitely still need to get the webpage.
|
||||
|
||||
info = self._download_json(
|
||||
self._API_URL % url, video_id,
|
||||
'Downloading video info JSON')
|
||||
|
||||
video_url = info.get('url')
|
||||
if video_url != None:
|
||||
video_url_raw = compat_urllib_request.quote(video_url)
|
||||
video_url = re.sub(r'^(?P<proto>https|http)%3A',
|
||||
lambda match: '%s:' % match.group('proto'),
|
||||
video_url_raw)
|
||||
|
||||
title = info.get('title')
|
||||
webpage = self._download_webpage(url, video_id)
|
||||
|
||||
if video_url == None:
|
||||
video_url = self._html_search_regex(
|
||||
r'<embed name="Video".*?src="([^"]+)"', webpage,
|
||||
'QuickTime embed', default=None)
|
||||
|
||||
if video_url == None:
|
||||
flash_vars_s = self._html_search_regex(
|
||||
r'<param name="flashVars" value="([^"]+)"', webpage, 'flash vars',
|
||||
default=None)
|
||||
if not flash_vars_s:
|
||||
flash_vars_s = self._html_search_regex(
|
||||
r'<param name="initParams" value="([^"]+)"', webpage, 'flash vars',
|
||||
default=None)
|
||||
if flash_vars_s:
|
||||
flash_vars_s = flash_vars_s.replace(',', '&')
|
||||
if flash_vars_s:
|
||||
flash_vars = compat_parse_qs(flash_vars_s)
|
||||
video_url_raw = compat_urllib_request.quote(
|
||||
flash_vars['content'][0])
|
||||
video_url = re.sub(r'^(?P<proto>http|https)%3A',
|
||||
lambda match: '%s:' % match.group('proto'),
|
||||
video_url_raw)
|
||||
|
||||
if video_url == None:
|
||||
video_meta = self._html_search_meta(
|
||||
'og:video', webpage, default=None)
|
||||
if video_meta:
|
||||
video_url = self._search_regex(
|
||||
r'src=(.*?)(?:$|&)', video_meta,
|
||||
'meta tag video URL', default=None)
|
||||
|
||||
if video_url == None:
|
||||
video_url = self._html_search_regex(
|
||||
r'MediaContentUrl["\']\s*:(["\'])(?P<url>(?:(?!\1).)+)\1',
|
||||
webpage, 'video url', default=None, group='url')
|
||||
|
||||
if video_url == None:
|
||||
video_url = self._html_search_meta(
|
||||
'og:video', webpage, default=None)
|
||||
|
||||
if video_url == None:
|
||||
raise ExtractorError('Cannot find video')
|
||||
|
||||
if title == None:
|
||||
title = self._og_search_title(webpage, default=None)
|
||||
|
||||
if title == None:
|
||||
title = self._html_search_regex(
|
||||
[r'<b>Title:</b> ([^<]+)</div>',
|
||||
r'class="tabSeperator">></span><span class="tabText">(.+?)<',
|
||||
r'<title>([^<]+)</title>'],
|
||||
webpage, 'title')
|
||||
|
||||
thumbnail = self._og_search_thumbnail(webpage)
|
||||
description = self._og_search_description(webpage, default=None)
|
||||
if description == None:
|
||||
description = self._html_search_meta('description', webpage)
|
||||
|
||||
return {
|
||||
'id': video_id,
|
||||
'url': video_url,
|
||||
'title': title,
|
||||
'description': description,
|
||||
'thumbnail': thumbnail,
|
||||
}
|
||||
Loading…
Reference in New Issue
Block a user