client.py

import requests
import json
import tarfile
import io
import time
import os

from config import *

def list_notebooks():
    res = requests.get(
                API_BASE_URL + "/notes",
                cookies = COOKIES,
                headers = {"User-Agent": USER_AGENT},
                proxies=PROXIES,
                verify=not DEBUG,
            )

    if res.status_code != 200:
        raise Exception("Unable to list notes")
    
    res_json = json.loads(res.text)
    for i, item in enumerate(res_json["itemsList"]):
        print(i, item["title"], item["id"])

    return res_json["itemsList"]

def get_render_token(notebook_id):
    get_params = f"notebookId={notebook_id}&marketplaceId={MARKETPLACE_ID}"
    res = requests.get(
                BASE_URL + "/openNotebook?" + get_params,
                cookies = COOKIES,
                headers = {"User-Agent": USER_AGENT},
                proxies=PROXIES,
                verify=not DEBUG,
            )

    if res.status_code != 200:
        raise Exception("Unable to open notebook")
    
    res_json = json.loads(res.text)
    return res_json

def is_notebook_updated(notebook, notebook_id):
    SAVE_DIR = os.path.join(NOTEBOOK_DIR, notebook_id)
    metadata_file_path = os.path.join(SAVE_DIR, "metadata")

    if not os.path.isfile(metadata_file_path):
        return True
    
    try:
        cached_metadata = json.load(open(metadata_file_path, 'r'))
    except:
        raise Exception("Error loading cached metadata")

    if notebook["metadata"]["modificationTime"] > cached_metadata["modificationTime"]:
        return True
    else:
        return False


def download_notebook(notebook, notebook_id, start_page=0, end_page=-1):

    if not is_notebook_updated(notebook, notebook_id):
        return True

    SAVE_DIR = os.path.join(NOTEBOOK_DIR, notebook_id) 

    if not os.path.isdir(SAVE_DIR):
        os.mkdir(SAVE_DIR)

    if end_page == -1:
        end_page = notebook["metadata"]["totalPages"] - 1

    get_params = f"startPage={start_page}&endPage={end_page}&width={IMAGE_WIDTH}&height={IMAGE_HEIGHT}&dpi={IMAGE_DPI}"
    res = requests.get(
                BASE_URL + "/renderPage?" + get_params,
                cookies = COOKIES,
                headers = {
                    "User-Agent": USER_AGENT,
                    "X-Amzn-Karamel-Notebook-Rendering-Token": notebook["renderingToken"],
                },
                proxies=PROXIES,
                verify=not DEBUG,
            )

    if res.status_code != 200:
        raise Exception("Unable to open notebook")

    tardata = io.BytesIO(res.content)

    extracted_files = {}

    with tarfile.open(fileobj=tardata) as tar:
        for member in tar.getmembers():
            if member.isfile():  # Skip directories
                extracted_files[member.name] = tar.extractfile(member).read()
    
    # clear out old data
    listing = os.listdir(SAVE_DIR)
    for item in listing:
        if item.endswith(".png") or item == "metadata":
            os.remove(os.path.join(SAVE_DIR, item))

    image_filenames = [name for name in extracted_files.keys() if name.endswith(".png")]
    for image_filename in image_filenames:
        with open(os.path.join(SAVE_DIR, image_filename), "wb") as f:
            f.write(extracted_files[image_filename])
    
    json.dump(notebook["metadata"], open(os.path.join(SAVE_DIR, "metadata"), "w"))
    

if __name__ == "__main__":
    notebooks = list_notebooks()
    selection = input(">")
    notebook_id = notebooks[int(selection)]["id"]
    notebook = get_render_token(notebook_id)
    try:
        download_notebook(notebook, notebook_id, 0, -1)
    except:
        download_notebook(notebook, notebook_id, 0, -1)