[radiocloud.jp] Add new extractor #12725
Conversation
youtube_dl/extractor/radiocloud.py
Outdated
| break | ||
|
|
||
| if not element: | ||
| raise ExtractorError('Could not find details of id {}'.format(video_id)) |
youtube_dl/extractor/radiocloud.py
Outdated
| return None | ||
|
|
||
| element = None | ||
| for e in get_elements_by_class("contents_box", webpage): |
youtube_dl/extractor/radiocloud.py
Outdated
| note='Downloading player', | ||
| errnote='Unable to download player') | ||
|
|
||
| file_url = self._search_regex(r'var\s*source\s*=\s*"(.+?)"', webpage, 'url') |
| 'ext': 'm4a', | ||
| 'title': '「オープニング」 ', | ||
| 'description': '「“フィギュアスケートは人生”引退会見で浅田真央さんは何を語ったのか？」スポーツライター・青嶋ひろのさんが解説！', | ||
| } |
There was a problem hiding this comment.
Interesting, it seems like for people without an account the page loads content a week or two into the past. For people who are logged in with an account the page loads content a month into the past.
I'll try to work out if I can find a way to load user restricted content without authentication. If that fails, I'll implement authentication (but I'll make it into its own commit instead of amending the current commit like for the other fixes).
I'm not sure if there is any access to content older than a month. I'll fetch the extracted URL for a recording about to expire and see if I can still access it after it expires.
If I can't find a way to download content which the website does not give direct access to even for logged in users, what should I do for tests which have an "expiration date" so to speak?
|
@dstftw Sorry for the long wait. I've been a bit busy and was for a while happy with my own solution. I've made the requested changes. Upon further inspection there is an "end_date" or expiry date on each upload. I did a search across all uploads on the website (with a separate script) and found an upload which expires in 2038. When 2038 is approaching, feel free to ask for a new test :P . So far I've made do with my own scripts but it would be really neat to get this in youtube_dl so I can use this feature directly with mpv. |
I have
One of the following applies
The purpose of this pull request
Description
This is a new extractor for https://radiocloud.jp/.
This is a audio-only website, if that is not within the scope of this project, my apologies, please close the pull request.
Additionally I can also confirm this is not a website which is based around copyright infringement, from the little Japanese that I know it appears to be the official website for some Japanese radio station.
The website restricts access to older recordings to users without an account, however this restriction is not enforced by anything more than a semi-transparent div preventing clicks, therefore authentication was not implemented in the extractor as it is unnecessary.
The website doesn't have any obvious way of getting the URL for a specific recording, to get one you need to view the RSS feed for a radio program and then use one of the URLs provided there.
The extractor currently makes no attempt to create a playlist of all recordings found on the archive page of a program as there could be quite a few recordings and I personally don't see it as a useful feature, but it could be done.
The extractor also makes no attempt to guess which recording you wish to download if the URL provided does not refer to a specific content_id.
A note about some regex:
I think this regex is a bit iffy, but I couldn't think of a better way.
Thank you for your time.
Edit: removed note about .strip() - I was mistaken.