From 0bf96f0213b8b8d12fdee78141ea6d592aad2f43 Mon Sep 17 00:00:00 2001
From: fchinembiri <fchinembiri24@gmail.com>
Date: Fri, 10 Jul 2026 20:40:39 +0200
Subject: [PATCH] fix(musicseerr): implement session-based conversion
 triggering and polling in Tubidy scraper

---
 apps/musicseerr/tasks.py | 68 ++++++++++++++++++++++++++++++++--------
 1 file changed, 55 insertions(+), 13 deletions(-)

diff --git a/apps/musicseerr/tasks.py b/apps/musicseerr/tasks.py
index 1f49211..fe3c90f 100644
--- a/apps/musicseerr/tasks.py
+++ b/apps/musicseerr/tasks.py
@@ -148,6 +148,7 @@ def search_tubidy(query: str, proxy_url: str) -> list:
 def get_tubidy_download_link(content_id: str, proxy_url: str) -> str:
     """
     Queries watch.php for the given content ID to find the direct MP3 download URL.
+    Supports initiating and polling the conversion process if the video is not yet processed.
     """
     proxies = {"http": proxy_url, "https": proxy_url} if proxy_url else None
     headers = {
@@ -166,23 +167,64 @@ def get_tubidy_download_link(content_id: str, proxy_url: str) -> str:
     
     logger.info(f"Fetching watch.php for content ID '{content_id}'")
     try:
-        response = cf_requests.get(watch_url, headers=headers, params=params, proxies=proxies, impersonate="chrome", timeout=15)
+        session = cf_requests.Session()
+        if proxies:
+            session.proxies = proxies
+        session.headers.update(headers)
+        
+        response = session.get(watch_url, params=params, impersonate="chrome", timeout=15)
         response.raise_for_status()
         
-        soup = BeautifulSoup(response.text, "html.parser")
-        for li in soup.find_all("li", class_=lambda x: x and "list-group-item" in x and "big" in x):
-            a_tag = li.find("a")
-            if a_tag:
+        def extract_link(html: str) -> str:
+            soup = BeautifulSoup(html, "html.parser")
+            # 1. Search in list-group-item big tags
+            for li in soup.find_all("li", class_=lambda x: x and "list-group-item" in x and "big" in x):
+                a_tag = li.find("a")
+                if a_tag:
+                    text = a_tag.text.lower()
+                    href = a_tag.get("href")
+                    if href and "download" in text:
+                        return urljoin(watch_url, href)
+            # 2. General fallback search for direct download/mp3 links
+            for a_tag in soup.find_all("a", href=True):
                 text = a_tag.text.lower()
-                href = a_tag.get("href")
-                if href and "download" in text:
-                    return href
-                    
-        # Fallback to general scraping of the watch.php page if structure differs
+                href = a_tag["href"]
+                if "download" in text and "act=process" not in href and "act=down" not in href:
+                    return urljoin(watch_url, href)
+                if (".mp3" in href.lower() or "video.mp3.tubidy.cool" in href.lower()) and "act=process" not in href:
+                    return urljoin(watch_url, href)
+            return ""
+            
+        link = extract_link(response.text)
+        if link:
+            return link
+            
+        # If no link found, check if there is a process link
+        soup = BeautifulSoup(response.text, "html.parser")
+        process_link = None
         for a_tag in soup.find_all("a", href=True):
-            text = a_tag.text.lower()
-            if "download" in text and "act=process" not in a_tag["href"]:
-                return a_tag["href"]
+            if "act=process" in a_tag["href"]:
+                process_link = urljoin(watch_url, a_tag["href"])
+                break
+                
+        if process_link:
+            logger.info(f"Triggering Tubidy processing step for content ID '{content_id}' via {process_link}")
+            proc_resp = session.get(process_link, impersonate="chrome", timeout=15)
+            proc_resp.raise_for_status()
+            
+            # Wait for conversion (typically takes 5-10s on Tubidy server)
+            import time
+            wait_time = 10
+            logger.info(f"Waiting {wait_time} seconds for conversion to finish...")
+            time.sleep(wait_time)
+            
+            # Re-fetch the download page
+            response = session.get(watch_url, params=params, impersonate="chrome", timeout=15)
+            response.raise_for_status()
+            link = extract_link(response.text)
+            if link:
+                return link
+                
     except Exception as e:
         logger.error(f"Failed to fetch download link for {content_id}: {e}")