Update app.py

app.py

@@ -6,39 +6,44 @@ app = FastAPI()
 # IMPORTANT: You need to confirm the text/selector of the download button.
 # Based on typical Pixeldrain layout, let's assume the button has the text "Download" or "Download File".
 # We'll use a robust selector that looks for the button element containing the text "Download".
+# --- app.py code ---
+
+# ... imports ...
+
 DOWNLOAD_BUTTON_SELECTOR = 'button:has-text("Download")' 
-# If the text is different (e.g., "Get File"), you MUST update the selector above.
+# Make sure to confirm the button text: "Download", "Download File", etc.
 
 def get_pixeldrain_info(pixeldrain_url: str):
     try:
         with sync_playwright() as p:
-            # Launch Chromium browser with slower speed for stability
+            # 1. Add User-Agent spoofing to the context creation
+            user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+            
             browser = p.chromium.launch(headless=True, slow_mo=50) 
-            page = browser.new_page()
+            # Create a new context with a realistic User-Agent
+            context = browser.new_context(user_agent=user_agent)
+            page = context.new_page()
 
-            # 1. Navigate to the Pixeldrain page
-            page.goto(pixeldrain_url, wait_until="networkidle")
-            print("Successfully navigated to Pixeldrain page.")
-            
-            # --- We need to capture the network traffic to find the final direct link ---
+            # 2. Navigate to the Pixeldrain page with a less strict wait condition
+            # Changed 'networkidle' to 'domcontentloaded' (or just 'load')
+            page.goto(pixeldrain_url, wait_until="domcontentloaded", timeout=30000) 
+            print("Successfully navigated to Pixeldrain page (DOM loaded).")
             
-            # Use a Python variable to store the final link we are looking for
+            # --- Network Listener (Keep this logic) ---
             final_download_url = None
             
-            # 2. Set up a listener to capture the network response
             def handle_response(response):
                 nonlocal final_download_url
-                # Look for the final redirect URL pattern (which often contains /api/file/...)
-                # The final content download link will look like https://pd.pixeldrain.com/api/file/HASH/FILENAME
+                # Look for the final redirect URL pattern (https://pd.pixeldrain.com/api/file/HASH/FILENAME)
                 if "/api/file/" in response.url and response.status == 200:
                     print(f"Captured final link via network: {response.url}")
                     final_download_url = response.url
-                    # Stop the network listener once the link is found (optional optimization)
                     
             page.on("response", handle_response)
             
             # 3. Wait for the download button to be visible and click it (to trigger s("download"))
             try:
+                # Give it a bit more time to appear after DOMContentLoaded
                 page.wait_for_selector(DOWNLOAD_BUTTON_SELECTOR, timeout=15000)
                 print(f"Found button with selector: {DOWNLOAD_BUTTON_SELECTOR}. Clicking...")
                 page.click(DOWNLOAD_BUTTON_SELECTOR)
@@ -46,29 +51,18 @@ def get_pixeldrain_info(pixeldrain_url: str):
                 raise ValueError(f"Timeout: Could not find or click the button with selector: {DOWNLOAD_BUTTON_SELECTOR}")
                 
             # 4. Wait for the final URL to be captured by the network listener
-            # We wait up to 15 seconds for the final URL variable to be set
-            page.wait_for_timeout(15000) # Give it time to execute JS and fire network requests
+            page.wait_for_timeout(15000) 
             
-            if not final_download_url:
-                # If the URL wasn't found in network, try to find it in the DOM (e.g., a hidden link that appears)
-                try:
-                    # Search for any link that looks like a final download link in the DOM
-                    dom_link = page.query_selector('a[href*="/api/file/"]')
-                    if dom_link:
-                        final_download_url = dom_link.get_attribute('href')
-                except:
-                    pass
-                
+            # ... (rest of the URL check/cookie extraction remains the same) ...
+
             if not final_download_url:
                 raise ValueError("Could not capture the final tokenized download URL after clicking.")
 
             # 5. Extract cookies and user-agent
             cookies = page.context.cookies()
             cookie_string = "; ".join([f"{c['name']}={c['value']}" for c in cookies])
-            user_agent = page.evaluate('() => navigator.userAgent')
-            
-            # Extract filename from the URL (best guess)
-            filename = final_download_url.split('/')[-1]
+            # Use the explicit User-Agent we set earlier
+            # user_agent is already defined at the start of the function
 
             browser.close()
 
@@ -76,13 +70,15 @@ def get_pixeldrain_info(pixeldrain_url: str):
                 "download_url": final_download_url,
                 "cookies": cookie_string,
                 "user_agent": user_agent,
-                "filename": filename 
+                "filename": filename # This comes from the URL split
             }
 
     except Exception as e:
         print(f"Error during Playwright execution: {e}")
+        # Re-raise the exception details for debugging, but return None to signal failure
         return None
 
+# ... (FastAPI routes remain the same) ...
 # FastAPI routes remain the same
 @app.get("/resolve_pixeldrain")
 def resolve_url(url: str):