[Bug] API ignoring field parameter

[sean89503]

Describe the bug
When using the api it is stated in the doc it will support the fields API

Steps to Reproduce
in python

import` requests
from urllib.parse import parse_qs

# Define the API endpoint
api_endpoint = "https://invidious.us/api/v1/channels/UC8i8e7likh-EYMq1bZ0ttHw"

# Define the desired fields
desired_fields = ["fields", "author", "authorId"]

# Construct the query string
query_string = "&".join([f"{field}=" for field in desired_fields])

# Append the query string to the API endpoint
url = f"{api_endpoint}?{query_string}"

# Make the GET request
response = requests.get(url)

# Check if the request was successful (status code 200)
if response.status_code == 200:
    # Parse the JSON response
    data = response.json()
    # Print the relevant fields
    print("Author:", data.get("author"))
    print("ALL LOOKS:", data.get("authorId"))
else:
    print("Error:", response.status_code)

please change invidious.us as that's a redirect site
-->

Screenshots

[unixfox]

fields feature were removed in #4276

doc is outdated :/ @SamantazFox

[absidue]

@unixfox There is already a PR open to update the docs, all you need to do is review it :)

[sean89503]

is there a way to get only a few items, most of the info I want is in /api/v1/CHANNELS/videos but the description of the videos are cut off in /api/v1/CHANNELS/videos so I have to go to /api/v1/videos/video_id to get it but I have to download everything.

[SamantazFox]

is there a way to get only a few items

@sean89503 We removed that fields feature because it was slowing down our API due to extra processing on the server side. Even if we only returned a single field, invidious had to parse the full data from Youtube, generate a full response, then drop the other fields.

In addition, the full response is small enough (~60kB or less than 10kB compressed) for it to not be a problem.

the description of the videos are cut off in /api/v1/CHANNELS/videos so I have to go to /api/v1/videos/video_id to get it but I have to download everything.

Blame Youtube for that. The full description is only available from the /api/v1/videos endpoint. All other places get a truncated (and un-formatted) version.