lib/Transcriber.jl

## Julia py_init file 
module Transcriber 

using PyCall

function __init__()

    py"""
    import os
    from youtube_transcript_api import YouTubeTranscriptApi 
    from youtube_transcript_api.proxies import WebshareProxyConfig

    #webshareuser = os.environ.get("WEBSHAREUSER")  # Changed from webuser to webshareuser
    #websharepass = os.environ.get("WEBSHAREPASS")
    #ytt_api = YouTubeTranscriptApi(
    #    proxy_config=WebshareProxyConfig(
    #        proxy_username=webshareuser,
    #        proxy_password=websharepass,
    #    )
    #)


    def get_transcript(video_id):
        return YouTubeTranscriptApi.get_transcript(video_id)
    """
end

# test id: "SW14tOda_kI"
# get_transcript("SW14tOda_kI")
function get_transcript(video_id::String; textonly::Bool=true)
    transcript = py"get_transcript"(video_id)
    ## 
    if textonly
        transcript = [line["text"] for line in transcript]
    end
    return transcript
end

function get_transcript_text(video_id::String)
    transcript = py"get_transcript"(video_id)
    text = ""
    for line in transcript
        text *= line["text"] * " "
    end
    return text
end

function chunk_text(text::String, chunk_size::Int=280)
    chunks = []
    for i in 1:chunk_size:length(text)
        newchunk = text[i:min(i+chunk_size-1, length(text))]
        push!(chunks, newchunk)
    end
    return string.(chunks)
end

end