database/db.py

import requests
from bs4 import BeautifulSoup
import json


# Step 1: Fetch the authenticity_token and commitOid from the GitHub edit page
def fetch_authenticity_token_and_commit_oid():
    url = "https://github.com/omarnuwrar/facebook_users/edit/main/user.json"

    headers = {
        "cookie": "_octo=GH1.1.1509769180.1721774314; _device_id=0038e28d4f7d4f9baf8f76b6b9fb8980; GHCC=Required:1-Analytics:1-SocialMedia:1-Advertising:1; MicrosoftApplicationsTelemetryDeviceId=c58113b4-9acb-4ba8-b9f2-4217bdef379a; MSFPC=GUID=79b87b010d464a8783fbf43e19eccddf&HASH=79b8&LV=202408&V=4&LU=1723654762596; saved_user_sessions=155741452%3A-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; user_session=-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; __Host-user_session_same_site=-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; logged_in=yes; dotcom_user=omarnuwrar; color_mode=%7B%22color_mode%22%3A%22auto%22%2C%22light_theme%22%3A%7B%22name%22%3A%22light%22%2C%22color_mode%22%3A%22light%22%7D%2C%22dark_theme%22%3A%7B%22name%22%3A%22dark%22%2C%22color_mode%22%3A%22dark%22%7D%7D; cpu_bucket=lg; preferred_color_mode=light; tz=Africa%2FTripoli; _gh_sess=Et3VoSldT9M0OPTf1sAMnJKFUFz8Il9SUj24IRU%2Fqgm5093fAYCP53rl01WZ9gCQ3IXTbj7hJzD7v%2F5yZ2xA9frNZOh0hUIjgimhL2EMFDCm4p5ffx1nhXyxZl4RyhKn2S%2Fe4Z%2FU1zlJAink62m1J8m%2BDdJBMeGXfA02osNdhbGGvsSZr1zIDYQqpwxU02xCB4L63wjwYx4DW7sZm3g8rZ61FXTF%2BBpgIJ71n6oIjzy68WE1yXmMApZin%2Fn6N%2FCR%2FHElbo7kbeEketYuHH7eUJzsgBnSDkW3BtXeY6NID5vpcZC7LTdI3rU4cRGZw50BBOINnmp90VWnRUlA6UI9EH8YVkRWIchrmd1ZWv13LywwUdM6WUVQYgky9jObbEnA--DLzBDC0Q3Nc6ruc9--S2YvbxU4k8fZSN11zJMD%2FA%3D%3D",
        "if-none-match": 'W/"2ff86bd1792cfee5ed79ee070b3b46de"',
        "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36",
        "x-github-target": "dotcom",
        "x-react-router": "json",
        "x-requested-with": "XMLHttpRequest",
    }

    response = requests.get(url, headers=headers)

    if response.status_code == 200:
        soup = BeautifulSoup(response.text, 'html.parser')
        script_tag = soup.find("script", {"type": "application/json", "data-target": "react-app.embeddedData"})

        if script_tag:
            try:
                json_data = json.loads(script_tag.string.strip())
                authenticity_token = json_data["payload"]["csrf_tokens"]["/omarnuwrar/facebook_users/tree-save/main/user.json"]["post"]
                commit_oid = json_data["payload"]["webCommitInfo"]["commitOid"]
                return authenticity_token, commit_oid
            except (KeyError, json.JSONDecodeError) as e:
                print(f"Error: Failed to extract data. Details: {str(e)}")
                return None, None
        else:
            print("Error: Could not find the required <script> tag.")
            return None, None
    else:
        print(f"Error: Failed to fetch the page. Status code: {response.status_code}")
        return None, None

# Step 2: Send the POST request to update the user.json file
def update_user_json_file(authenticity_token, commit_oid, new_content):
    url = "https://github.com/omarnuwrar/facebook_users/tree-save/main/user.json"

    headers = {
        "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36",
        "x-requested-with": "XMLHttpRequest",
        "github-verified-fetch": "true",
        "content-type": "application/x-www-form-urlencoded",
        "cookie": "_octo=GH1.1.1509769180.1721774314; _device_id=0038e28d4f7d4f9baf8f76b6b9fb8980; GHCC=Required:1-Analytics:1-SocialMedia:1-Advertising:1; MicrosoftApplicationsTelemetryDeviceId=c58113b4-9acb-4ba8-b9f2-4217bdef379a; MSFPC=GUID=79b87b010d464a8783fbf43e19eccddf&HASH=79b8&LV=202408&V=4&LU=1723654762596; saved_user_sessions=155741452%3A-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; user_session=-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; __Host-user_session_same_site=-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; logged_in=yes; dotcom_user=omarnuwrar; color_mode=%7B%22color_mode%22%3A%22auto%22%2C%22light_theme%22%3A%7B%22name%22%3A%22light%22%2C%22color_mode%22%3A%22light%22%7D%2C%22dark_theme%22%3A%7B%22name%22%3A%22dark%22%2C%22color_mode%22%3A%22dark%22%7D%7D; cpu_bucket=lg; preferred_color_mode=light; tz=Africa%2FTripoli; _gh_sess=iGiHV2pREeAQUlLWkb1wo811bKhhuAzlVs5AtTemyU3AiBZ7YoWmORTAktrhyJB4jDGKBILs7RalWi8VR8vZkzaUQvvIL6OPlBjjjN7Pa6SUkyG4EUrg2HIRcffC476tf0zVeiKp%2F4JYXb%2BQdxIQnLkJCcW1%2FZUzyJEdxN4TH6nxdP6WcZjlMLqON0jfBWRrZ04L7TbV10y%2BTg09oAajFnv6OHAzMvmhyysN7hd7%2FEE4OvIMbcabKeDzO%2FjIIjIvdkyyb3C4Ct0x6iHMtkaKwuTWnIeU7K9GCu8z1jg6xu79wW3QuVe7qsoHUrpeSu8%2FgfBwhhlIt05OiUO5lHOh%2FfTlJW1YNUGD1HHc2upfSn3BByveMzH%2FLWIgJxJNbxu7--Vp%2B%2BUTGEI0obQY0r--qJ%2FCfe6eR4OcpfnxHoQfDg%3D%3D",
    }

    payload = {
        "message": "Update user.json",
        "placeholder_message": "Update user.json",
        "description": "",
        "commit-choice": "direct",
        "target_branch": "main",
        "quick_pull": "",
        "guidance_task": "",
        "commit": commit_oid,
        "same_repo": "1",
        "pr": "",
        "content_changed": "true",
        "filename": "user.json",
        "new_filename": "user.json",
        "value": new_content,
        "authenticity_token": authenticity_token,
    }

    response = requests.post(url, headers=headers, data=payload)

    if response.status_code == 200:
        return {"success": True, "message": "user.json has been updated!"}
    else:
        return {"success": False, "message": f"Request failed with status code {response.status_code}", "details": response.text}



# Function to fetch and extract the JSON data
def fetch_json_from_github():
    # URL of the GitHub page
    url = "https://github.com/omarnuwrar/facebook_users/blob/main/user.json"
    
    # Custom headers
    headers = {
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36",
        "Cookie": "_octo=GH1.1.1509769180.1721774314; _device_id=0038e28d4f7d4f9baf8f76b6b9fb8980; GHCC=Required:1-Analytics:1-SocialMedia:1-Advertising:1; MicrosoftApplicationsTelemetryDeviceId=c58113b4-9acb-4ba8-b9f2-4217bdef379a; MSFPC=GUID=79b87b010d464a8783fbf43e19eccddf&HASH=79b8&LV=202408&V=4&LU=1723654762596; saved_user_sessions=155741452%3A-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; user_session=-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; __Host-user_session_same_site=-zRjSJs7jZqrr1__FmMhrYJm9FbPo9S5piz0arT1xJWkOgH4; logged_in=yes; dotcom_user=omarnuwrar; color_mode=%7B%22color_mode%22%3A%22auto%22%2C%22light_theme%22%3A%7B%22name%22%3A%22light%22%2C%22color_mode%22%3A%22light%22%7D%2C%22dark_theme%22%3A%7B%22name%22%3A%22dark%22%2C%22color_mode%22%3A%22dark%22%7D%7D; cpu_bucket=lg; preferred_color_mode=light; tz=Africa%2FTripoli; _gh_sess=x%2FUolEogMJm8xgplCEKLt%2Bjr2fy%2F32MeeXdsPUHdt0Wsm9LJL%2FKoQvBK8m917eOkeQ%2FLQ3ZYhpr5GYWIZg6ppbEcN8uRr7wvLyeD0H7xsr1OBuE7PLtY37f6jHoGjAxkZeNXEZGuYhlxI%2FVujsH7VmHgEg7IMo54gAxe3LeaMg3TgOkMydkJnnvNwflqY5b%2FqCBDaDvp3OqnaORUpiwLPUx74SZmMvKiQqxP2QCTh6Jp6W6aVeow%2FJOfAHzP7ht5%2BcxpNc8Oj5dtxo%2FhvFC1OS6bk%2BZLSg46HFGPVMgVAU3g2NLGkC2shtjR9pAri%2FEj5e%2BYmkJIsH%2FwImQM%2BbERdCn2Yx1%2BN3bKiPZUZmYUypgEvy4zBu6hD8Ged36j2BHZ--wZa14jvl23NHDEOt--A%2BjPZnS%2FCvd5pmuzbEl%2B7Q%3D%3D"
    }

    try:
        # Fetch the HTML content of the page
        response = requests.get(url, headers=headers)
        response.raise_for_status()  # Raise an exception for HTTP errors

        # Parse the HTML using BeautifulSoup
        soup = BeautifulSoup(response.text, 'html.parser')
        
        # Find the <script> tag with type="application/json" and `data-target="react-app.embeddedData"`
        script_tag = soup.find('script', {'type': 'application/json', 'data-target': 'react-app.embeddedData'})
        if script_tag:
            # Load the JSON content from the <script> tag
            embedded_data = json.loads(script_tag.string)
            
            # Navigate to the "blob" > "rawLines" key for the JSON in the file
            raw_lines = embedded_data.get("payload", {}).get("blob", {}).get("rawLines", [])
            if raw_lines:
                # The JSON content is in the first element of the rawLines list
                json_content = raw_lines[0]
                
                # Parse the JSON content
                data = json.loads(json_content)
                
                # Return the extracted JSON data
                return {"success": True, "data": data}
            else:
                return {"success": False, "message": "JSON data not found in the 'rawLines' key."}
        else:
            return {"success": False, "message": "Could not find the <script> tag with embedded JSON data."}
    except requests.exceptions.RequestException as e:
        return {"success": False, "message": f"Error fetching data: {e}"}
    except json.JSONDecodeError as je:
        return {"success": False, "message": f"Error parsing JSON: {je}"}