Updated twitter login

twitter_login.py

@@ -1,103 +1,221 @@
 import argparse
-from playwright.sync_api import sync_playwright
+import json
-from tweety import Twitter
+import os
+import shutil
 import time
+from playwright.sync_api import sync_playwright
 
-def get_twitter_auth_token(username, password):
+SESSION_FILE_PERMISSIONS = 0o600
+
+
+def normalize_cookie_for_playwright(cookie: dict) -> dict:
+    """
+    Ensure cookie fields are compatible with Playwright's storage_state format.
+    Playwright requires 'domain' to start with a dot for cross-subdomain cookies,
+    and 'sameSite' must be one of 'Strict', 'Lax', or 'None'.
+    """
+    c = dict(cookie)
+
+    # Normalize domain: x.com → .x.com
+    domain = c.get("domain", "")
+    if domain and not domain.startswith("."):
+        c["domain"] = f".{domain}"
+
+    # Normalize sameSite to valid Playwright values
+    same_site = c.get("sameSite", "")
+    valid_same_site = {"Strict", "Lax", "None"}
+    if same_site not in valid_same_site:
+        c["sameSite"] = "None"
+
+    # Ensure required fields have defaults
+    c.setdefault("path", "/")
+    c.setdefault("httpOnly", False)
+    c.setdefault("secure", True)
+    c.setdefault("expires", -1)
+
+    return c
+
+
+def get_twitter_cookies(username: str, password: str) -> dict:
+    """
+    Automates Twitter login via Playwright and returns a Playwright-compatible
+    storage_state dict (cookies + origins/localStorage).
+    """
     with sync_playwright() as p:
         print("🚀 Launching headless browser...")
-        # Add arguments to bypass basic bot detection
+        browser = p.chromium.launch(
-        browser = p.chromium.launch(headless=True, args=["--disable-blink-features=AutomationControlled"])
+            headless=True,
+            args=["--disable-blink-features=AutomationControlled"]
+        )
         context = browser.new_context(
-            user_agent="Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+            user_agent=(
+                "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) "
+                "AppleWebKit/537.36 (KHTML, like Gecko) "
+                "Chrome/145.0.7632.6 Safari/537.36"
+            ),
+            viewport={"width": 1920, "height": 1080},
         )
         page = context.new_page()
 
         try:
             print("🌐 Navigating to X login...")
-            page.goto("https://x.com/i/flow/login")
+            page.goto("https://x.com/i/flow/login", wait_until="domcontentloaded")
 
-            # 1. Fill with username
+            # --- Username ---
-            print(f"👤 Entering username: {username}...")
+            print(f"👤 Entering username: {username[:10]}...")
-            # Using the exact autocomplete attribute from your HTML
             page.wait_for_selector('input[autocomplete="username"]', timeout=25000)
-            time.sleep(10)  # Wait a bit for any potential animations or dynamic content to load
-            # CLICK ON THE INPUT BEFORE POPULATING THE USERNAME
             page.click('input[autocomplete="username"]')
-            time.sleep(10)
             page.fill('input[autocomplete="username"]', username)
 
-            # Enter and wait for the next screen
+            print("➡️ Pressing Next...")
-            print("➡️ Pressing Enter to proceed...")
+            page.locator('button:has-text("Next")').first.click()
-            page.keyboard.press("Enter")
 
-            # Wait a moment for the screen transition animation
+            # --- Security challenge (email/phone) ---
-            time.sleep(10)
+            page.wait_for_selector(
+                'input[name="password"], '
+                'input[data-testid="ocfEnterTextTextInput"], '
+                'input[name="text"]',
+                timeout=15000,
+            )
+            time.sleep(1)
 
-            # 2. Fill with the password
+            if (
+                page.locator('input[data-testid="ocfEnterTextTextInput"]').is_visible()
+                or page.locator('input[name="text"]').is_visible()
+            ):
+                print("🛡️ Security challenge detected — this script needs --email arg.")
+                raise RuntimeError(
+                    "Security challenge appeared but no email/phone was provided. "
+                    "Re-run with --email your_email_or_phone"
+                )
+
+            # --- Password ---
             print("🔑 Entering password...")
-            # Using the exact name attribute from your HTML
+            page.wait_for_selector('input[name="password"]', timeout=15000)
-            page.wait_for_selector('input[name="password"]', timeout=25000)
-            time.sleep(10)  # Wait a bit for any potential animations or dynamic content to load
-            # CLICK ON THE INPUT BEFORE POPULATING THE PASSWORD
-            page.click('input[name="password"]')
             page.fill('input[name="password"]', password)
 
-            # 3. Click on Login
             print("🖱️ Clicking 'Log in'...")
-            # Targeting the exact span text you provided
             page.locator('span:has-text("Log in")').first.click()
 
-            print("⏳ Waiting for login to complete...")
+            # --- Poll for auth_token + ct0 ---
-            # Wait for the redirect to the home page
+            print("🍪 Waiting for auth_token + ct0 cookies...")
-            page.wait_for_url("**/home", timeout=25000)
-            
-            # 4. Extract Cookies
-            cookies = context.cookies()
             auth_token = None
-            for cookie in cookies:
+            ct0 = None
-                if cookie['name'] == 'auth_token':
+            for _ in range(40):
-                    auth_token = cookie['value']
+                cookies_list = context.cookies()
+                auth_token = next((c["value"] for c in cookies_list if c["name"] == "auth_token"), None)
+                ct0 = next((c["value"] for c in cookies_list if c["name"] == "ct0"), None)
+                if auth_token and ct0:
+                    print("✅ Both cookies found!")
                     break
+                page.wait_for_timeout(1000)
+            else:
+                raise TimeoutError("auth_token/ct0 cookies never appeared after 40 seconds.")
+
+            # --- Wait for page to fully settle before evaluate() ---
+            print("⏳ Waiting for page to stabilize...")
+            page.wait_for_load_state("domcontentloaded", timeout=15000)
+            try:
+                page.wait_for_selector('[data-testid="primaryColumn"]', timeout=20000)
+            except Exception:
+                print("⚠️ primaryColumn not found — page may still be usable.")
+
+            # --- Grab localStorage (non-critical) ---
+            try:
+                local_storage = page.evaluate(
+                    "() => Object.entries(localStorage).map(([name, value]) => ({ name, value }))"
+                )
+            except Exception as ls_err:
+                print(f"⚠️ Could not extract localStorage (non-critical): {ls_err}")
+                local_storage = []
+
+            # --- Re-fetch final cookie list and normalize for Playwright ---
+            raw_cookies = context.cookies()
+            normalized_cookies = [normalize_cookie_for_playwright(c) for c in raw_cookies]
+
+            session_data = {
+                "cookies": normalized_cookies,
+                "origins": [
+                    {
+                        "origin": "https://x.com",
+                        "localStorage": local_storage,
+                    }
+                ],
+            }
+
+            print(f"✅ auth_token: {auth_token[:10]}...")
+            print(f"✅ ct0:        {ct0[:10]}...")
 
             browser.close()
-            return auth_token
+            return session_data
 
         except Exception as e:
-            # IF IT FAILS, TAKE A SCREENSHOT!
             print(f"❌ Error encountered. Taking a screenshot...")
-            page.screenshot(path="error_screenshot.png")
+            try:
+                page.screenshot(path="error_screenshot.png")
+                print("📸 Saved: error_screenshot.png")
+            except Exception:
+                pass
             browser.close()
             raise e
 
+
+def save_session(session_data: dict, path: str):
+    """Save Playwright storage_state JSON to disk with restricted permissions."""
+    os.makedirs(os.path.dirname(os.path.abspath(path)), exist_ok=True)
+    temp_path = f"{path}.tmp"
+    with open(temp_path, "w", encoding="utf-8") as f:
+        json.dump(session_data, f, indent=2)
+    os.replace(temp_path, path)
+    try:
+        os.chmod(path, SESSION_FILE_PERMISSIONS)
+    except Exception as e:
+        print(f"⚠️ Could not set file permissions on {path}: {e}")
+    print(f"💾 Session saved to: {path}")
+
+
+def clear_session(path: str):
+    """Delete a stale session file."""
+    if os.path.exists(path):
+        os.remove(path)
+        print(f"🧹 Removed stale session file: {path}")
+    else:
+        print(f"ℹ️ No session file found at {path} — nothing to clear.")
+
+
 def main():
-    # Set up argument parsing
+    parser = argparse.ArgumentParser(
-    parser = argparse.ArgumentParser(description="Automate Twitter login and save session token.")
+        description="Automate Twitter login and save Playwright session state."
-    parser.add_argument("username", help="Your Twitter username")
+    )
-    parser.add_argument("password", help="Your Twitter password")
+    parser.add_argument("username", help="Twitter username or handle")
+    parser.add_argument("password", help="Twitter password")
+    parser.add_argument(
+        "--email",
+        default="",
+        help="Twitter email or phone (required if X shows a security challenge)",
+    )
+    parser.add_argument(
+        "--output",
+        default="twitter_browser_state.json",  # ← matches twitter2bsky.py exactly
+        help="Output path for the Playwright storage state JSON",
+    )
+    parser.add_argument(
+        "--clear-session",
+        action="store_true",
+        help="Delete any existing session file before starting",
+    )
 
     args = parser.parse_args()
 
-    try:
+    if args.clear_session:
-        # Pass the arguments from the terminal to the function
+        clear_session(args.output)
-        token = get_twitter_auth_token(args.username, args.password)
 
-        if token:
+    session_data = get_twitter_cookies(args.username, args.password)
-            print("🔑 Successfully extracted auth_token!")
+    save_session(session_data, args.output)
 
-            # Initialize the tweety session file
+    print(f"\n✅ Done! Session saved to '{args.output}'")
-            app = Twitter("my_account_session")
+    print(f"   twitter2bsky.py will automatically pick it up on next run.")
 
-            print("💾 Saving session to tweety-ns...")
-            app.load_auth_token(token)
-            
-            print("✅ Success! Your session has been saved.")
-        else:
-            print("❌ Failed to find auth_token cookie.")
-            
-    except Exception as e:
-        print(f"❌ An error occurred: {e}")
-        print("📸 Check the 'error_screenshot.png' file in your folder to see what X showed the bot!")
 
 if __name__ == "__main__":
     main()