download pdf onedrive that block download

down_pdf_onedrive.py

import json
import re
import requests
from urllib.parse import urlparse, parse_qs

file = input("File url: ")


def custom_url_encode(input_string):
    encoded_string = ""

    for char in input_string:
        # Encode alphanumeric characters
        if char.isalnum():
            encoded_string += char
        else:
            # Encode non-alphanumeric characters
            encoded_string += "%" + "{:02X}".format(ord(char))

    return encoded_string


session = requests.Session()
session.headers.update({
    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.142.86 Safari/537.36"})
real_res = session.get(file)
url_parse = urlparse(real_res.url)
page_data_raw = re.search(r"var _spPageContextInfo=(.*?);", real_res.text).group(1)
page_data = json.loads(page_data_raw)
query = parse_qs(url_parse.query)
file_id = query["id"][0]
root_info = session.get(page_data["driveInfo"][".driveUrl"]).json()
a1_data = urlparse(root_info["webUrl"]).path

# to do custom encode
file_raw = file_id.rsplit('/', 1)
RootFolder = custom_url_encode(file_raw[0])
file_name = file_raw[1]
a1 = custom_url_encode(f"'{(a1_data)}'")

hd = {"accept": "application/json;odata=verbose", "content-type": "application/json;odata=verbose"}
body = {"parameters": {"__metadata": {"type": "SP.RenderListDataParameters"}, "RenderOptions": 5445383,
                       "AllowMultipleValueFilterForTaxonomyFields": True, "AddRequiredFields": True,
                       "RequireFolderColoringFields": True}}
path_sub_api = f"/_api/web/GetListUsingPath(DecodedUrl=@a1)/RenderListDataAsStream?@a1={a1}&RootFolder={RootFolder}"
rurl = page_data["webAbsoluteUrl"] + path_sub_api
full_info = session.post(rurl, headers=hd, data=json.dumps(body)).json()
env_json = full_info["ListSchema"]
env_json[".method"] = "passthrough"
for i in full_info["ListData"]["Row"]:
    env_json.update(i)
for k, v in env_json.copy().items():
    if k.startswith("."):
        env_json[k.replace(".", "")] = v

file_data = session.get(full_info["ListSchema"][".transformUrl"].replace(".", "").format(**env_json)).content
with open(file_name, "wb") as f:
    f.write(file_data)

requirements.txt

requests