starred/spotipy

Fork 0

mirror of https://github.com/spotipy-dev/spotipy.git synced 2026-04-27 00:25:54 +03:00

[GH-ISSUE #277] sp.audio_features() not working correctly. Spotify updated endpoints? #151

New issue

Closed

opened 2026-02-27 23:21:06 +03:00 by kerem · 8 comments

kerem commented

2026-02-27 23:21:06 +03:00

Owner

Originally created by @aalshukri on GitHub (Apr 22, 2018).
Original GitHub issue: https://github.com/spotipy-dev/spotipy/issues/277

Hi,

I have discovered that the function sp.audio_features() is not working correctly. This function currently returns nulls.

I think the problem relates to the endpoint in spotipy.

Minimal code example which returns nulls.
features = sp.audio_features(tid) print(json.dumps(features, indent=4))

I have also tried the the example code:
features = sp.audio_features(tids) delta = time.time() - start for feature in features: print(json.dumps(feature, indent=4)) print() analysis = sp._get(feature['analysis_url']) print(json.dumps(analysis, indent=4)) print() print ("features retrieved in %.2f seconds" % (delta,))
Reference: https://github.com/plamere/spotipy/blob/master/examples/audio_features.py

The above example failed because "sp._get(feature['analysis_url'])" does not return any JSON results (it returned nulls as mentioned) which causes an exception.

What I expected from this function was a list of features as shown below:
{ "danceability": 0.696, "energy": 0.905, "key": 2, "loudness": -2.743, "mode": 1, "speechiness": 0.103, "acousticness": 0.011, "instrumentalness": 0.000905, "liveness": 0.302, "valence": 0.625, "tempo": 114.944, "type": "audio_features", "id": "11dFghVXANMlKmJXsNCbNl", "uri": "spotify:track:11dFghVXANMlKmJXsNCbNl", "track_href": "https://api.spotify.com/v1/tracks/11dFghVXANMlKmJXsNCbNl", "analysis_url": "https://api.spotify.com/v1/audio-analysis/11dFghVXANMlKmJXsNCbNl", "duration_ms": 207960, "time_signature": 4 }
Reference: https://beta.developer.spotify.com/console/get-audio-features-track/

The functionality can be replicated using the spotify web console
GET https://api.spotify.com/v1/audio-features/{id}

Thanks for your help
Ash

Originally created by @aalshukri on GitHub (Apr 22, 2018). Original GitHub issue: https://github.com/spotipy-dev/spotipy/issues/277 Hi, I have discovered that the function sp.audio_features() is not working correctly. This function currently returns nulls. I think the problem relates to the endpoint in spotipy. Minimal code example which returns nulls. ` features = sp.audio_features(tid) print(json.dumps(features, indent=4)) ` I have also tried the the example code: ` features = sp.audio_features(tids) delta = time.time() - start for feature in features: print(json.dumps(feature, indent=4)) print() analysis = sp._get(feature['analysis_url']) print(json.dumps(analysis, indent=4)) print() print ("features retrieved in %.2f seconds" % (delta,)) ` Reference: https://github.com/plamere/spotipy/blob/master/examples/audio_features.py The above example failed because "sp._get(feature['analysis_url'])" does not return any JSON results (it returned nulls as mentioned) which causes an exception. What I expected from this function was a list of features as shown below: ` { "danceability": 0.696, "energy": 0.905, "key": 2, "loudness": -2.743, "mode": 1, "speechiness": 0.103, "acousticness": 0.011, "instrumentalness": 0.000905, "liveness": 0.302, "valence": 0.625, "tempo": 114.944, "type": "audio_features", "id": "11dFghVXANMlKmJXsNCbNl", "uri": "spotify:track:11dFghVXANMlKmJXsNCbNl", "track_href": "https://api.spotify.com/v1/tracks/11dFghVXANMlKmJXsNCbNl", "analysis_url": "https://api.spotify.com/v1/audio-analysis/11dFghVXANMlKmJXsNCbNl", "duration_ms": 207960, "time_signature": 4 } ` Reference: https://beta.developer.spotify.com/console/get-audio-features-track/ The functionality can be replicated using the spotify web console ` GET https://api.spotify.com/v1/audio-features/{id} ` Thanks for your help Ash

kerem

2026-02-27 23:21:06 +03:00

closed this issue
added the
bug
label

kerem commented

2026-02-27 23:21:07 +03:00

Author

Owner

@shatteringlass commented on GitHub (Aug 5, 2018):

I can confirm this. Possibly I'll submit a PR.

EDIT: strangely enough, the current implementation works for string arguments but not lists. I was able to fetch audio features for a comma separated string of track IDs. There seems to be nothing wrong in the else branch of the function code.

@shatteringlass commented on GitHub (Aug 5, 2018): I can confirm this. Possibly I'll submit a PR. EDIT: strangely enough, the current implementation works for string arguments but not lists. I was able to fetch audio features for a comma separated string of track IDs. There seems to be nothing wrong in the [else branch](https://github.com/plamere/spotipy/blob/4c2c1d763a3653aa225c4af848409ec31286a6bf/spotipy/client.py#L862) of the function code.

kerem commented

2026-02-27 23:21:07 +03:00

Author

Owner

@lyons7 commented on GitHub (Sep 19, 2018):

Did this ever get fixed? I couldn't even get the comma separated string of track IDs work-around to work!

@lyons7 commented on GitHub (Sep 19, 2018): Did this ever get fixed? I couldn't even get the comma separated string of track IDs work-around to work!

kerem commented

2026-02-27 23:21:07 +03:00

Author

Owner

@eribradl commented on GitHub (Oct 3, 2018):

I just ran into this same issue today, and it looks to me like audio_features('something') is expecting 'something' to be a list. Similarly, the response you get back is also a list.
@aalshukri To make your first example work, try:
features = sp.audio_features(tid)[0]

@eribradl commented on GitHub (Oct 3, 2018): I just ran into this same issue today, and it looks to me like audio_features('something') is expecting 'something' to be a list. Similarly, the response you get back is also a list. @aalshukri To make your first example work, try: features = sp.audio_features(tid)[0]

kerem commented

2026-02-27 23:21:07 +03:00

Author

Owner

@Brakjen commented on GitHub (May 7, 2019):

I am not getting this to work either. It returns a list of None. However, I can get the features if I set up my own get request using requests, and json to parse the output. The function below could be made more sophisticated to accept lists of track ids, and not a single id. Just pass your token and track id as kwargs.

import requests
from bs4 import BeautifulSoup as bs
import json

def get_track_features(token=None, track_id=None): 
    url = "https://api.spotify.com/v1/audio-features/"+track_id
    with requests.session() as sesh:
        header = {"Authorization": "Bearer {}".format(token)}
        response = sesh.get(url, headers=header)
        feature = str(bs(response.content, "html.parser"))
        return json.loads(feature)

@Brakjen commented on GitHub (May 7, 2019): I am not getting this to work either. It returns a list of None. However, I can get the features if I set up my own get request using requests, and json to parse the output. The function below could be made more sophisticated to accept lists of track ids, and not a single id. Just pass your token and track id as kwargs. ``` import requests from bs4 import BeautifulSoup as bs import json def get_track_features(token=None, track_id=None): url = "https://api.spotify.com/v1/audio-features/"+track_id with requests.session() as sesh: header = {"Authorization": "Bearer {}".format(token)} response = sesh.get(url, headers=header) feature = str(bs(response.content, "html.parser")) return json.loads(feature) ```

kerem commented

2026-02-27 23:21:07 +03:00

Author

Owner

@paarmann commented on GitHub (Feb 6, 2020):

Works as expected in spotipy=2.7.1

query_ids = ['id1', 'id2',...]
response = sp.audio_features(query_ids)

Returns:

[{'danceability': 0.817, 'energy': 0.781, 'key': 7, 'loudness': -4.178, 'mode': 0, 'speechiness': 0.112, 'acousticness': 0.00838, 'instrumentalness': 5.21e-06, 'liveness': 0.123, 'valence': 0.359, 'tempo': 104.0, 'type': 'audio_features', 'id': '07s9NNOT0sZQp7TyolLLgu', 'uri': 'spotify:track:07s9NNOT0sZQp7TyolLLgu', 'track_href': 'https://api.spotify.com/v1/tracks/07s9NNOT0sZQp7TyolLLgu', 'analysis_url': 'https://api.spotify.com/v1/audio-analysis/07s9NNOT0sZQp7TyolLLgu', 'duration_ms': 216707, 'time_signature': 4}, {'danceability': 0.659, ...

The result list may contain Null values if there are no audio features available for the id at this position.

@paarmann commented on GitHub (Feb 6, 2020): Works as expected in spotipy=2.7.1 ``` query_ids = ['id1', 'id2',...] response = sp.audio_features(query_ids) ``` Returns: ``` [{'danceability': 0.817, 'energy': 0.781, 'key': 7, 'loudness': -4.178, 'mode': 0, 'speechiness': 0.112, 'acousticness': 0.00838, 'instrumentalness': 5.21e-06, 'liveness': 0.123, 'valence': 0.359, 'tempo': 104.0, 'type': 'audio_features', 'id': '07s9NNOT0sZQp7TyolLLgu', 'uri': 'spotify:track:07s9NNOT0sZQp7TyolLLgu', 'track_href': 'https://api.spotify.com/v1/tracks/07s9NNOT0sZQp7TyolLLgu', 'analysis_url': 'https://api.spotify.com/v1/audio-analysis/07s9NNOT0sZQp7TyolLLgu', 'duration_ms': 216707, 'time_signature': 4}, {'danceability': 0.659, ... ``` The result list may contain Null values if there are no audio features available for the id at this position.

kerem commented

2026-02-27 23:21:07 +03:00

Author

Owner

@stephanebruckert commented on GitHub (Feb 9, 2020):

Thanks for the feedback @paarmann, gonna close this one

@stephanebruckert commented on GitHub (Feb 9, 2020): Thanks for the feedback @paarmann, gonna close this one

kerem commented

2026-02-27 23:21:07 +03:00

Author

Owner

@anushkacodez commented on GitHub (Feb 13, 2025):

Works as expected in spotipy=2.7.1

query_ids = ['id1', 'id2',...]
response = sp.audio_features(query_ids)

Returns:

[{'danceability': 0.817, 'energy': 0.781, 'key': 7, 'loudness': -4.178, 'mode': 0, 'speechiness': 0.112, 'acousticness': 0.00838, 'instrumentalness': 5.21e-06, 'liveness': 0.123, 'valence': 0.359, 'tempo': 104.0, 'type': 'audio_features', 'id': '07s9NNOT0sZQp7TyolLLgu', 'uri': 'spotify:track:07s9NNOT0sZQp7TyolLLgu', 'track_href': 'https://api.spotify.com/v1/tracks/07s9NNOT0sZQp7TyolLLgu', 'analysis_url': 'https://api.spotify.com/v1/audio-analysis/07s9NNOT0sZQp7TyolLLgu', 'duration_ms': 216707, 'time_signature': 4}, {'danceability': 0.659, ...

The result list may contain Null values if there are no audio features available for the id at this position.

Hi this doesn't work anymore. Is there any other way to get audio features in python?

@anushkacodez commented on GitHub (Feb 13, 2025): > Works as expected in spotipy=2.7.1 > > ``` > query_ids = ['id1', 'id2',...] > response = sp.audio_features(query_ids) > ``` > > Returns: > > ``` > [{'danceability': 0.817, 'energy': 0.781, 'key': 7, 'loudness': -4.178, 'mode': 0, 'speechiness': 0.112, 'acousticness': 0.00838, 'instrumentalness': 5.21e-06, 'liveness': 0.123, 'valence': 0.359, 'tempo': 104.0, 'type': 'audio_features', 'id': '07s9NNOT0sZQp7TyolLLgu', 'uri': 'spotify:track:07s9NNOT0sZQp7TyolLLgu', 'track_href': 'https://api.spotify.com/v1/tracks/07s9NNOT0sZQp7TyolLLgu', 'analysis_url': 'https://api.spotify.com/v1/audio-analysis/07s9NNOT0sZQp7TyolLLgu', 'duration_ms': 216707, 'time_signature': 4}, {'danceability': 0.659, ... > ``` > > The result list may contain Null values if there are no audio features available for the id at this position. Hi this doesn't work anymore. Is there any other way to get audio features in python?

kerem commented

2026-02-27 23:21:07 +03:00

Author

Owner

@dieser-niko commented on GitHub (Feb 13, 2025):

No, you're somewhat late to the party. Please check #1172

@dieser-niko commented on GitHub (Feb 13, 2025): No, you're somewhat late to the party. Please check #1172