fix langlinks.py

import tqdm
import re
import json
import requests
from huggingface_hub import login
from huggingface_hub import upload_file
from datasets import Dataset

from google.colab import userdata

# تسجيل الدخول إلى Hugging Face (استبدل "YOUR_ACCESS_TOKEN" بالتوكن الخاص بك)
login(userdata.get('HF_API'))

# تحميل الملف JSON من الرابط مباشرة
json_url = "https://huggingface.co/Ibrahemqasim/enwiki_to_arwiki_categories/resolve/main/langlinks.json"
response = requests.get(json_url)
data = response.json()

# تحميل الملف JSON من الرابط مباشرة
json_url2 = "https://huggingface.co/Ibrahemqasim/enwiki_to_arwiki_categories/resolve/main/countries.json"
response2 = requests.get(json_url2)
countries = response2.json()

# تحويل القاموس إلى قائمة من القواميس [{ "en": "value", "ar": "value" }, ...]

to_work = [
    "langlinks",
    "filtered_data",
    "cats_2000_contry",
    "cats_2000",
]

data_lists = {
    "langlinks" : {},
    "filtered_data" : {},
    "cats_2000_contry" : {},
    "cats_2000" : {},
}


# نرتب الدول حسب عدد الفراغات (تنازليًا)
sorted_keys = sorted(countries.keys(), key=lambda x: -x.count(' '))

# نبني تعبير regex
regex_pattern = r'\b(' + '|'.join(map(re.escape, sorted_keys)) + r')\b'

YEAR_PATTERN = "{YEAR}"
COUNTRY_PATTERN = "{COUNTRY}"

for tab in tqdm.tqdm(data):
    # ---
    key = tab["en"]
    value = tab["ar"]
    # ---
    # "Category:1. FC Köln non-playing staff"
    # remove " from start and end
    # ---
    if key.startswith('"') and key.endswith('"'):
        key = key[1:-1]
    # ----
    # remove (:") from start and remove (",) from end
    # :"cc",
    if value.startswith(':"') and value.endswith('",'):
        value = value[2:-2]
    # ----
    # data_lists["langlinks"].append({"en": key, "ar": value, "count": 0})
    if key in data_lists["langlinks"]:
        data_lists["langlinks"][key]["count"] += 1
    else:
        data_lists["langlinks"][key] = {"ar": value, "count": 0}
    # ----
    # Add if key and value has 4 digits and they are the same
    reg_year = r"(\d+[–-]\d+|\d{4})"
    # ---
    key_digits = re.search(reg_year, key)
    value_digits = re.search(reg_year, value)
    # ----
    if key_digits and value_digits and key_digits.group() == value_digits.group():
        # data_lists["filtered_data"].append({"en": key, "ar": value, "count": 0})
        if key in data_lists["filtered_data"]:
            data_lists["filtered_data"][key]["count"] += 1
        else:
            data_lists["filtered_data"][key] = {"ar": value, "count": 0}
        # ---
        key2 = key.replace(key_digits.group(), YEAR_PATTERN)
        value2 = value.replace(value_digits.group(), YEAR_PATTERN)
        # ---
        # data_lists["cats_2000"].append({"en": key2, "ar": value2, "count": 0})
        # ---
        if key2 in data_lists["cats_2000"]:
            data_lists["cats_2000"][key2]["count"] += 1
        else:
            data_lists["cats_2000"][key2] = {"ar": value2, "count": 0}
        # ----
        # البحث عن اسم الدولة في key2
        match = re.search(regex_pattern, key2)
        # ----
        if match:
            en_country = match.group(1)
            ar_country = countries.get(en_country)
            # ---
            if ar_country and ar_country in value2:
                key3 = re.sub(rf'\b{re.escape(en_country)}\b', COUNTRY_PATTERN, key2)
                value3 = re.sub(rf'\b{re.escape(ar_country)}\b', COUNTRY_PATTERN, value2)
                # ---
                if COUNTRY_PATTERN in key3 and COUNTRY_PATTERN in value3:
                    # ---
                    if key3 in data_lists["cats_2000_contry"]:
                        data_lists["cats_2000_contry"][key3]["count"] += 1
                    else:
                        data_lists["cats_2000_contry"][key3] = {"ar": value3, "count": 0}
        # ----
# ----
print(f"{len(data_lists['cats_2000_contry'])=}")
print(f"{len(data_lists['cats_2000'])=}")

print(f"all data len: {len(data):,}.")

datasets_list = {
    "langlinks" : "categories_en2ar",
    "filtered_data" : "categories_en2ar_with_years",
    "cats_2000_contry" : "categories_en2ar-cats_2000_contry",
    "cats_2000" : "categories_en2ar-cats_2000",
}

# for x, data_list in data_lists.items():
for x in to_work:
    data_list = data_lists.get(x)
    # ---
    data_list = [{"en": key, "ar": value["ar"], "count": value["count"]} for key, value in data_list.items()]

    # حفظ القاموس المصحح في ملف JSON
    with open(f"{x}.json", "w", encoding="utf-8") as f:
        json.dump(data_list, f, ensure_ascii=False, indent=4)

    print("______________")
    print(f"file: {x} uploaded successfully!")
    print(f"len of {x} : {len(data_list)}.")
    # ---
    # continue
    # ---
    upload_file(
        path_or_fileobj=f"{x}.json",  # اسم الملف الذي تم حفظه
        path_in_repo=f"{x}.json",  # المسار داخل المستودع
        repo_id="Ibrahemqasim/enwiki_to_arwiki_categories",  # معرف المستودع
        # repo_type="dataset",  # نوع المستودع (نستخدم dataset للملفات)
    )

    print("____________________________")

    set_name = datasets_list.get(x)

    if set_name:
        # إنشاء Dataset
        dataset = Dataset.from_list(data_list)

        # رفع Dataset إلى Hugging Face
        dataset.push_to_hub(f"Ibrahemqasim/{set_name}")