Downloading captions always returns a 403

问题:

When I call the captions.download endpoint with an ID that we retrieve from the captions.list endpoint, it always returns a 403. For example:

https://www.youtube.com/watch?v=1HRwpwOj4aA

I call captions.list with:

GET https://www.googleapis.com/youtube/v3/captions?part=id&videoId=1HRwpwOj4aA&key={YOUR_API_KEY}

This is response:

cache-control:  private, max-age=0, must-revalidate, no-transform
content-encoding:  gzip
content-length:  236
content-type:  application/json; charset=UTF-8
date:  Sat, 23 May 2015 17:55:57 GMT
etag:  "dhbhlDw5j8dK10GxeV_UG6RSReM/Rztb3ln4Zb6O07vb7_KSZi2y1NM"
expires:  Sat, 23 May 2015 17:55:57 GMT
server:  GSE
vary:  Origin, X-Origin

{
 "kind": "youtube#captionListResponse",
 "etag": "\"dhbhlDw5j8dK10GxeV_UG6RSReM/Rztb3ln4Zb6O07vb7_KSZi2y1NM\"",
 "items": [
  {
   "kind": "youtube#caption",
   "etag": "\"dhbhlDw5j8dK10GxeV_UG6RSReM/pwH-4wtyQJz0U3l57fA8uKm4e1I\"",
   "id": "kHlUsiuNS4TjB25loauZNXGrjK91I1tEdNyOpTRNA78="
  }
 ]
}

When I use the above id to call captions.download:

GET https://www.googleapis.com/youtube/v3/captions/kHlUsiuNS4TjB25loauZNXGrjK91I1tEdNyOpTRNA78%3D?key={YOUR_API_KEY}

This is response:

403 Forbidden
cache-control:  private, max-age=0
content-encoding:  gzip
content-length:  29
content-type:  text/html; charset=UTF-8
date:  Sat, 23 May 2015 17:59:05 GMT
expires:  Sat, 23 May 2015 17:59:05 GMT
server:  GSE
vary:  Origin, X-Origin

Forbidden

Any ideas what could be happening here?

回答1:

From the YouTube API docs:

403 Forbidden: The permissions associated with the request are not sufficient to download the caption track. The request might not be properly authorized, or the video order might not have enabled third-party contributions for this caption.

回答2:

what you wrote above about "only works for videos your google account owns" is not my experience. I just successfully ran captions.download on a video (about dog training) which I definitely do not own - do not even have a dog. However, I have tested the exact same code on the video mentioned here on stackoverflow and get a 403 error.

So no it doesn't always return a 403 sometimes it returns a 200! Try it with the dog video mentioned above:

python captions.py --videoid="=jBN2_YuTclU" --action="download" --captionid='8S2GjnNfitU5HHoLyTeLxq_W1dP29YRFC8E8vFBUtws='

with the code you probably already have here.

It will need your client_secrets.json downloaded from the Google credentials page somewhere and a missing file youtube-v3-api-captions.json which you can get from here. The code launches a browser where you log in for OAuth2 authorisation.

Still, there must be a reason why it works for some videos and not others. @Abhishek might have it above. The wrong comment has been upvoted there. Nothing in the output of captions.list for a video that allows captions downloads and a video that does not is obviously different which would explains why one works and the other does not. If anyone can supply which {'key':'value'} pair in the youtube api controls this would be helpful.

回答3:

Status 403 Forbidden means that nobody has the right to access that URL. You shouldn't receive that message if you have the wrong API key, for example; that should give Status 401 Unauthorised. I'd check the URL carefully.

回答4:

Instead caption download API that sometimes returns 403 (if video not have enabled third-party contributions for this caption) you can use youtube.com/api/timedtext