Use Protobuf in my Stack Overflow answer to retrieve video captions

[Benjamin-Loison]

The concerned Stack Overflow answer 70013529.

As requested on Discord.

Concerning the first part, maybe I simplified by trying to remove characters one by one and now it is not correctly encoded (such that I am not able to formalize the associated encoding) but correctly decoded on Google servers.

Being aware of what parameters are actually necessary would be nice to minimize requests. Related to #256.

To verify correctness can use:

jq '.actions[0].updateEngagementPanelAction.content.transcriptRenderer.content.transcriptSearchPanelRenderer.body.transcriptSegmentListRenderer.initialSegments[].transcriptSegmentRenderer.snippet.runs[0].text'

Related to Benjamin_Loison/protobuf_google/issues/1.

As blackboxprotobuf does not have this issue, let us switch to Python.

[Benjamin-Loison]

import requests
import blackboxprotobuf
import base64

def getBase64Protobuf(message, typedef):
    data = blackboxprotobuf.encode_message(message, typedef)
    return base64.b64encode(data).decode('ascii')

requestsParams = {
    'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'
}
url = 'https://www.youtube.com/youtubei/v1/get_transcript'
headers = {
    'Content-Type': 'application/json'
}

message = {
    "1": "lo0X2ZdElQ4",
    "2": "CgASAnJ1GgA="
}

typedef = {
    "1": {
        "type": "string"
    },
    "2": {
        "type": "string"
    }
}

params = getBase64Protobuf(message, typedef)

data = {
    'context': {
        'client': {
            'clientName': 'WEB',
            'clientVersion': '2.20240313.05.00'
        }
    },
    'params': params
}

data = requests.post(url, params = requestsParams, headers = headers, json = data).json()
print('развитием' in str(data))

[Benjamin-Loison]

Concerning the first part, as requested on Discord:

curl https://www.youtube.com/youtubei/v1/get_transcript -H 'Content-Type: application/json' --data-raw '{"context": {"client": {"clientName": "WEB", "clientVersion": "2.20240408.05.00"}}, "params": "CgtxUVY2Z3V2Rm1NMBISQ2dOaGMzSVNBbVZ1R2dBJTNEGAEqM2VuZ2FnZW1lbnQtcGFuZWwtc2VhcmNoYWJsZS10cmFuc2NyaXB0LXNlYXJjaC1wYW5lbDABOAFAAQ%3D%3D"}'

import requests
import blackboxprotobuf
import base64

def getBase64Protobuf(message, typedef):
    data = blackboxprotobuf.encode_message(message, typedef)
    return base64.b64encode(data).decode('ascii')

message = {
    '1': 'asr',
    '2': 'en',
}

typedef = {
    '1': {
        'type': 'string'
    },
    '2': {
        'type': 'string'
    },
}

two = getBase64Protobuf(message, typedef)

message = {
    '1': 'qQV6guvFmM0',
    '2': two,
}

typedef = {
    '1': {
        'type': 'string'
    },
    '2': {
        'type': 'string'
    },
}

params = getBase64Protobuf(message, typedef)

url = 'https://www.youtube.com/youtubei/v1/get_transcript'
headers = {
    'Content-Type': 'application/json'
}
data = {
    'context': {
        'client': {
            'clientName': 'WEB',
            'clientVersion': '2.20240313'
        }
    },
    'params': params
}

data = requests.post(url, headers = headers, json = data).json()
print('this is Will Smith and I\'m waiting for' in str(data))