fix(musicseerr): enhance Tubidy search with query variations and censorship handling

apps/musicseerr/tasks.py

@@ -96,6 +96,7 @@ def scrape_tubidy_link(url: str, proxy_url: str) -> str:
 def search_tubidy(query: str, proxy_url: str) -> list:
     """
     Searches Tubidy for the query and returns a list of results with ID, title, and link.
+    Includes robust fallback variations if the initial search returns no results.
     """
     proxies = {"http": proxy_url, "https": proxy_url} if proxy_url else None
     headers = {
@@ -115,31 +116,83 @@ def search_tubidy(query: str, proxy_url: str) -> list:
         response = cf_requests.get(search_url, headers=headers, params=params, proxies=proxies, impersonate="chrome", timeout=15)
         response.raise_for_status()
         
-        soup = BeautifulSoup(response.text, "html.parser")
-        results = []
+        def parse_results(html: str) -> list:
+            soup = BeautifulSoup(html, "html.parser")
+            results_list = []
+            for media_body in soup.find_all("div", class_="media-body"):
+                a_tag = media_body.find("a")
+                if a_tag:
+                    href = a_tag.get("href")
+                    if href:
+                        title = a_tag.get("aria-label") or a_tag.text.strip()
+                        match = re.search(r'/watch/([^/]+)', href)
+                        content_id = match.group(1) if match else None
+                        link = href
+                        if href.startswith("//"):
+                            link = f"https:{href}"
+                        elif href.startswith("/"):
+                            link = f"{endpoint}{href}"
                         
-        for media_body in soup.find_all("div", class_="media-body"):
-            a_tag = media_body.find("a")
-            if a_tag:
-                href = a_tag.get("href")
-                if href:
-                    title = a_tag.get("aria-label") or a_tag.text.strip()
-                    # Extract ID from /watch/content_id
-                    match = re.search(r'/watch/([^/]+)', href)
-                    content_id = match.group(1) if match else None
-                    link = href
-                    if href.startswith("//"):
-                        link = f"https:{href}"
-                    elif href.startswith("/"):
-                        link = f"{endpoint}{href}"
+                        if content_id:
+                            results_list.append({
+                                "id": content_id,
+                                "title": title,
+                                "link": link
+                            })
+            return results_list
 
-                    if content_id:
-                        results.append({
-                            "id": content_id,
-                            "title": title,
-                            "link": link
-                        })
-        return results
+        results = parse_results(response.text)
+        if results:
+            return results
+
+        # Fallback query variations if first search yielded no results
+        variations = []
+        
+        # Variation 1: Clean basic symbols and parentheticals
+        clean1 = re.sub(r'[\(\[][^\)\]]*[\)\]]', '', query)
+        clean1 = clean1.replace("-", " ")
+        clean1 = re.sub(r'\s+', ' ', clean1).strip()
+        variations.append(clean1)
+        
+        # Variation 2: Clean and also remove feat/ft
+        clean2 = re.sub(r'(?i)\b(feat|ft)\.?\b.*$', '', clean1)
+        clean2 = re.sub(r'[^\w\s]', '', clean2)
+        clean2 = re.sub(r'\s+', ' ', clean2).strip()
+        variations.append(clean2)
+        
+        # Variation 3: Handle censored words (e.g. fucked -> f*cked, fuck -> f*ck)
+        if "fucked" in clean2.lower() or "fuck" in clean2.lower():
+            clean3 = re.sub(r'(?i)\bfucked\b', 'f*cked', clean2)
+            clean3 = re.sub(r'(?i)\bfuck\b', 'f*ck', clean3)
+            variations.append(clean3)
+            
+        # Variation 4: Try just the artist and the first 2 words of the title
+        parts = query.split(" - ", 1)
+        if len(parts) == 2:
+            artist = re.sub(r'[^\w\s]', '', parts[0]).strip()
+            title_words = re.sub(r'[^\w\s]', '', parts[1]).strip().split()
+            if len(title_words) > 1:
+                clean4 = f"{artist} {' '.join(title_words[:1])}"
+                variations.append(clean4)
+                
+        # Try each variation in order
+        for var in variations:
+            var = var.strip()
+            if not var or var == query:
+                continue
+            logger.info(f"Retrying Tubidy search with variation: '{var}'")
+            params["q"] = var
+            try:
+                response = cf_requests.get(search_url, headers=headers, params=params, proxies=proxies, impersonate="chrome", timeout=15)
+                response.raise_for_status()
+                new_results = parse_results(response.text)
+                if new_results:
+                    logger.info(f"Found {len(new_results)} results for variation '{var}'")
+                    return new_results
+            except Exception as e:
+                logger.warning(f"Variation search for '{var}' failed: {e}")
+
+        return []
     except Exception as e:
         logger.error(f"Failed to search Tubidy: {e}")
         return []