[iTunes] Add new extractor (Closes #2097) #9590
Conversation
TRox1972
changed the title
youtube_dl/extractor/itunes.py
Outdated
|
|
||
| webpage = self._download_webpage(sanitized_Request(self._html_search_regex( | ||
| r'<string>\s*(https?://itunes.apple.com/[^>]+)</string>', self._download_webpage( | ||
| request, display_id), 'iTunes url'), headers={'User-Agent': self._USER_AGENT}), display_id) |
There was a problem hiding this comment.
Avoid such cumbersome code. That's unreadable.
|
Seems without an iTunes User-Agent, the response from https://itunes.apple.com/us/itunes-u/uc-davis-symphony-orchestra/id403834767 just contains all what we want. If you're going to keep the current approach, use |
youtube_dl/extractor/itunes.py
Outdated
| # coding: utf-8 | ||
| from __future__ import unicode_literals | ||
|
|
||
| import datetime |
|
@yan12125 You're right. I'll change the approach to not spoof the user agent. |
| webpage = self._download_webpage(url, display_id) | ||
|
|
||
| video_infos = re.findall(r'var\s+__desc_popup_d_\d+\s*=\s*({[^><]+});', webpage) | ||
| html_entries = re.findall(r'<tr\s+[^>]*role="row"[^>]+>', webpage) |
There was a problem hiding this comment.
get_element_by_attribute() may be useful.
There was a problem hiding this comment.
It will get the content between <tr> and </tr>, but not the content inside the tags, like <tr preview-duration="485000">
|
Don't |
|
@yan12125 Sorry didn't know that. |
TRox1972
force-pushed
the
p18
branch
2 times, most recently
from
d4bf26f
to
5762b67
Compare
|
|
||
|
|
||
| class iTunesIE(InfoExtractor): | ||
| _VALID_URL = r'https?://itunes\.apple\.com/[a-z]{2}/[a-z0-9-]+/(?P<display_id>[a-z0-9-]+)?/(?:id)?(?P<id>[0-9]+)' |
There was a problem hiding this comment.
It should be the following as the (valid) URL variations of https://itunes.apple.com/us/itunes-u/uc-davis-symphony-orchestra/id403834767 don't match otherwise
_VALID_URL = r'https?://itunes\.apple\.com/[a-z]{2}?/?[a-z0-9-]+/?(?P<display_id>[a-z0-9-]+)?/(?:id)?(?P<id>[0-9]+)'
The extractor only works for free content, like most podcasts, i.e. it does not download 30-seconds previews of paid songs.