quasarr 1.23.0__py3-none-any.whl → 1.24.0__py3-none-any.whl

quasarr/downloads/sources/dd.py

@@ -6,13 +6,19 @@ from quasarr.providers.log import info, debug
 from quasarr.providers.sessions.dd import create_and_persist_session, retrieve_and_validate_session
 
 
-def get_dd_download_links(shared_state, url, mirror, title): # signature must align with other download link functions!
+def get_dd_download_links(shared_state, url, mirror, title, password):
+    """
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    Returns plain download links from DD API.
+    """
+
     dd = shared_state.values["config"]("Hostnames").get("dd")
 
     dd_session = retrieve_and_validate_session(shared_state)
     if not dd_session:
         info(f"Could not retrieve valid session for {dd}")
-        return []
+        return {"links": []}
 
     links = []
 
@@ -35,9 +41,9 @@ def get_dd_download_links(shared_state, url, mirror, title): # signature must al
     try:
         release_list = []
         for page in range(0, 100, 20):
-            url = f'https://{dd}/index/search/keyword/{title}/qualities/{','.join(qualities)}/from/{page}/search'
+            api_url = f'https://{dd}/index/search/keyword/{title}/qualities/{",".join(qualities)}/from/{page}/search'
 
-            releases_on_page = dd_session.get(url, headers=headers, timeout=10).json()
+            releases_on_page = dd_session.get(api_url, headers=headers, timeout=10).json()
             if releases_on_page:
                 release_list.extend(releases_on_page)
 
@@ -46,7 +52,7 @@ def get_dd_download_links(shared_state, url, mirror, title): # signature must al
                 if release.get("fake"):
                     debug(f"Release {release.get('release')} marked as fake. Invalidating DD session...")
                     create_and_persist_session(shared_state)
-                    return []
+                    return {"links": []}
                 elif release.get("release") == title:
                     filtered_links = []
                     for link in release["links"]:
@@ -61,10 +67,11 @@ def get_dd_download_links(shared_state, url, mirror, title): # signature must al
                                 for existing_link in filtered_links
                         ):
                             debug(f"Skipping duplicate `.mkv` link from {link['hostname']}")
-                            continue  # Skip adding duplicate `.mkv` links from the same hostname
+                            continue
                         filtered_links.append(link)
 
-                    links = [link["url"] for link in filtered_links]
+                    # Build [[url, mirror], ...] format
+                    links = [[link["url"], link["hostname"]] for link in filtered_links]
                     break
             except Exception as e:
                 info(f"Error parsing DD download: {e}")
@@ -73,4 +80,4 @@ def get_dd_download_links(shared_state, url, mirror, title): # signature must al
     except Exception as e:
         info(f"Error loading DD download: {e}")
 
-    return links
+    return {"links": links}

quasarr/downloads/sources/dj.py

@@ -3,5 +3,14 @@
 # Project by https://github.com/rix1337
 
 
-def get_dj_download_links(shared_state, url, mirror, title):  # signature must align with other download link functions!
-    return [url]
+def get_dj_download_links(shared_state, url, mirror, title, password):
+    """
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    DJ source handler - the site itself acts as a protected crypter.
+    Returns the URL for CAPTCHA solving via userscript.
+    """
+
+    return {
+        "links": [[url, "junkies"]]
+    }

quasarr/downloads/sources/dl.py

@@ -17,24 +17,19 @@ def extract_password_from_post(soup, host):
     Extract password from forum post using multiple strategies.
     Returns empty string if no password found or if explicitly marked as 'no password'.
     """
-    # Get flattened text from the post - collapse whitespace to single spaces
     post_text = soup.get_text()
     post_text = re.sub(r'\s+', ' ', post_text).strip()
 
-    # Strategy 1: Look for password label followed by the password value
-    # Pattern: "Passwort:" followed by optional separators, then the password
     password_pattern = r'(?:passwort|password|pass|pw)[\s:]+([a-zA-Z0-9._-]{2,50})'
     match = re.search(password_pattern, post_text, re.IGNORECASE)
 
     if match:
         password = match.group(1).strip()
-        # Skip if it looks like a section header or common word
         if not re.match(r'^(?:download|mirror|link|episode|info|mediainfo|spoiler|hier|click|klick|kein|none|no)',
                         password, re.IGNORECASE):
             debug(f"Found password: {password}")
             return password
 
-    # Strategy 2: Look for explicit "no password" indicators (only if no valid password found)
     no_password_patterns = [
         r'(?:passwort|password|pass|pw)[\s:]*(?:kein(?:es)?|none|no|nicht|not|nein|-|–|—)',
         r'(?:kein(?:es)?|none|no|nicht|not|nein)\s*(?:passwort|password|pass|pw)',
@@ -45,7 +40,6 @@ def extract_password_from_post(soup, host):
             debug("No password required (explicitly stated)")
             return ""
 
-    # Strategy 3: Default to hostname-based password
     default_password = f"www.{host}"
     debug(f"No password found, using default: {default_password}")
     return default_password
@@ -54,40 +48,53 @@ def extract_password_from_post(soup, host):
 def extract_mirror_name_from_link(link_element):
     """
     Extract the mirror/hoster name from the link text or nearby text.
-    Returns the extracted name or None.
     """
-    # Get the link text
     link_text = link_element.get_text(strip=True)
-
-    # Try to extract a meaningful name from the link text
-    # Look for text that looks like a hoster name (alphanumeric, may contain numbers/dashes)
-    # Filter out common non-hoster words
     common_non_hosters = {'download', 'mirror', 'link', 'hier', 'click', 'klick', 'code', 'spoiler'}
 
-    # Clean and extract potential mirror name
+    # Known hoster patterns for image detection
+    known_hosters = {
+        'rapidgator': ['rapidgator', 'rg'],
+        'ddownload': ['ddownload', 'ddl'],
+        'turbobit': ['turbobit'],
+        '1fichier': ['1fichier'],
+    }
+
     if link_text and len(link_text) > 2:
-        # Remove common symbols and whitespace
         cleaned = re.sub(r'[^\w\s-]', '', link_text).strip().lower()
-
-        # If it's a single word or hyphenated word and not in common non-hosters
         if cleaned and cleaned not in common_non_hosters:
-            # Extract the main part (first word if multiple)
             main_part = cleaned.split()[0] if ' ' in cleaned else cleaned
-            if len(main_part) > 2:  # Must be at least 3 characters
+            if 2 < len(main_part) < 30:
                 return main_part
 
-    # Check if there's a bold tag or nearby text in parent
     parent = link_element.parent
     if parent:
-        parent_text = parent.get_text(strip=True)
-        # Look for text before the link that might be the mirror name
         for sibling in link_element.previous_siblings:
-            if hasattr(sibling, 'get_text'):
-                sibling_text = sibling.get_text(strip=True).lower()
-                if sibling_text and len(sibling_text) > 2 and sibling_text not in common_non_hosters:
-                    cleaned = re.sub(r'[^\w\s-]', '', sibling_text).strip()
-                    if cleaned:
-                        return cleaned.split()[0] if ' ' in cleaned else cleaned
+            # Only process Tag elements, skip NavigableString (text nodes)
+            if not hasattr(sibling, 'name') or sibling.name is None:
+                continue
+
+            # Skip spoiler elements entirely
+            classes = sibling.get('class', [])
+            if classes and any('spoiler' in str(c).lower() for c in classes):
+                continue
+
+            # Check for images with hoster names in src/alt/data-url
+            img = sibling.find('img') if sibling.name != 'img' else sibling
+            if img:
+                img_identifiers = (img.get('src', '') + img.get('alt', '') + img.get('data-url', '')).lower()
+                for hoster, patterns in known_hosters.items():
+                    if any(pattern in img_identifiers for pattern in patterns):
+                        return hoster
+
+            sibling_text = sibling.get_text(strip=True).lower()
+            # Skip if text is too long - likely NFO content or other non-mirror text
+            if len(sibling_text) > 30:
+                continue
+            if sibling_text and len(sibling_text) > 2 and sibling_text not in common_non_hosters:
+                cleaned = re.sub(r'[^\w\s-]', '', sibling_text).strip()
+                if cleaned and 2 < len(cleaned) < 30:
+                    return cleaned.split()[0] if ' ' in cleaned else cleaned
 
     return None
 
@@ -95,12 +102,6 @@ def extract_mirror_name_from_link(link_element):
 def extract_links_and_password_from_post(post_content, host):
     """
     Extract download links and password from a forum post.
-    Only filecrypt and hide are supported - other link crypters will cause an error.
-
-    Returns:
-        tuple of (links, password) where:
-        - links: list of [url, mirror_name] pairs where mirror_name is the actual hoster
-        - password: extracted password string
     """
     links = []
     soup = BeautifulSoup(post_content, 'html.parser')
@@ -108,11 +109,9 @@ def extract_links_and_password_from_post(post_content, host):
     for link in soup.find_all('a', href=True):
         href = link.get('href')
 
-        # Skip internal forum links
         if href.startswith('/') or host in href:
             continue
 
-        # Check supported link crypters
         if re.search(r'filecrypt\.', href, re.IGNORECASE):
             crypter_type = "filecrypt"
         elif re.search(r'hide\.', href, re.IGNORECASE):
@@ -123,16 +122,11 @@ def extract_links_and_password_from_post(post_content, host):
             crypter_type = "tolink"
         else:
             debug(f"Unsupported link crypter/hoster found: {href}")
-            debug(f"Currently only filecrypt and hide are supported. Other crypters may be added later.")
             continue
 
-        # Extract mirror name from link text or nearby context
         mirror_name = extract_mirror_name_from_link(link)
-
-        # Use mirror name if found, otherwise fall back to crypter type
         identifier = mirror_name if mirror_name else crypter_type
 
-        # Avoid duplicates
         if [href, identifier] not in links:
             links.append([href, identifier])
             if mirror_name:
@@ -140,7 +134,6 @@ def extract_links_and_password_from_post(post_content, host):
             else:
                 debug(f"Found {crypter_type} link (no mirror name detected)")
 
-    # Only extract password if we found links
     password = ""
     if links:
         password = extract_password_from_post(soup, host)
@@ -148,52 +141,51 @@ def extract_links_and_password_from_post(post_content, host):
     return links, password
 
 
-def get_dl_download_links(shared_state, url, mirror, title):
+def get_dl_download_links(shared_state, url, mirror, title, password):
     """
-    Get download links from a thread.
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    DL source handler - extracts links and password from forum thread.
 
-    Returns:
-        tuple of (links, password) where:
-        - links: list of [url, mirror_name] pairs
-        - password: extracted password string
+    Note: The password parameter is unused intentionally - password must be extracted from the post.
     """
+
     host = shared_state.values["config"]("Hostnames").get(hostname)
 
     sess = retrieve_and_validate_session(shared_state)
     if not sess:
         info(f"Could not retrieve valid session for {host}")
-        return [], ""
+        return {"links": [], "password": ""}
 
     try:
         response = fetch_via_requests_session(shared_state, method="GET", target_url=url, timeout=30)
 
         if response.status_code != 200:
             info(f"Failed to load thread page: {url} (Status: {response.status_code})")
-            return [], ""
+            return {"links": [], "password": ""}
 
         soup = BeautifulSoup(response.text, 'html.parser')
 
         first_post = soup.select_one('article.message--post')
         if not first_post:
             info(f"Could not find first post in thread: {url}")
-            return [], ""
+            return {"links": [], "password": ""}
 
         post_content = first_post.select_one('div.bbWrapper')
         if not post_content:
             info(f"Could not find post content in thread: {url}")
-            return [], ""
+            return {"links": [], "password": ""}
 
-        # Extract both links and password from the same post content
-        links, password = extract_links_and_password_from_post(str(post_content), host)
+        links, extracted_password = extract_links_and_password_from_post(str(post_content), host)
 
         if not links:
             info(f"No supported download links found in thread: {url}")
-            return [], ""
+            return {"links": [], "password": ""}
 
-        debug(f"Found {len(links)} download link(s) for: {title} (password: {password})")
-        return links, password
+        debug(f"Found {len(links)} download link(s) for: {title} (password: {extracted_password})")
+        return {"links": links, "password": extracted_password}
 
     except Exception as e:
         info(f"Error extracting download links from {url}: {e}")
         invalidate_session(shared_state)
-        return [], ""
+        return {"links": [], "password": ""}

quasarr/downloads/sources/dt.py

@@ -3,12 +3,35 @@
 # Project by https://github.com/rix1337
 
 import re
+from urllib.parse import urlparse
+
 import requests
 from bs4 import BeautifulSoup
+
 from quasarr.providers.log import info
 
 
-def get_dt_download_links(shared_state, url, mirror, title): # signature must align with other download link functions!
+def derive_mirror_from_url(url):
+    """Extract hoster name from URL hostname."""
+    try:
+        hostname = urlparse(url).netloc.lower()
+        if hostname.startswith('www.'):
+            hostname = hostname[4:]
+        parts = hostname.split('.')
+        if len(parts) >= 2:
+            return parts[-2]
+        return hostname
+    except:
+        return "unknown"
+
+
+def get_dt_download_links(shared_state, url, mirror, title, password):
+    """
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    DT source handler - returns plain download links.
+    """
+
     headers = {"User-Agent": shared_state.values["user_agent"]}
     session = requests.Session()
 
@@ -19,20 +42,19 @@ def get_dt_download_links(shared_state, url, mirror, title): # signature must al
         article = soup.find("article")
         if not article:
             info(f"Could not find article block on DT page for {title}")
-            return False
+            return None
+
         body = article.find("div", class_="card-body")
         if not body:
             info(f"Could not find download section for {title}")
-            return False
+            return None
 
-        # grab all <a href="…">
         anchors = body.find_all("a", href=True)
 
     except Exception as e:
         info(f"DT site has been updated. Grabbing download links for {title} not possible! ({e})")
-        return False
+        return None
 
-    # first do your normal filtering
     filtered = []
     for a in anchors:
         href = a["href"].strip()
@@ -45,22 +67,22 @@ def get_dt_download_links(shared_state, url, mirror, title): # signature must al
         if mirror and mirror not in href:
             continue
 
-        filtered.append(href)
+        mirror_name = derive_mirror_from_url(href)
+        filtered.append([href, mirror_name])
 
-    # if after filtering you got nothing, fall back to regex
+    # regex fallback if still empty
     if not filtered:
         text = body.get_text(separator="\n")
         urls = re.findall(r'https?://[^\s<>"\']+', text)
-        # de-dupe preserving order
         seen = set()
         for u in urls:
             u = u.strip()
             if u not in seen:
                 seen.add(u)
-                # apply same filters
                 low = u.lower()
                 if low.startswith(("http://", "https://")) and "imdb.com" not in low and "?ref=" not in low:
                     if not mirror or mirror in u:
-                        filtered.append(u)
+                        mirror_name = derive_mirror_from_url(u)
+                        filtered.append([u, mirror_name])
 
-    return filtered
+    return {"links": filtered} if filtered else None

quasarr/downloads/sources/dw.py

@@ -10,7 +10,13 @@ from bs4 import BeautifulSoup
 from quasarr.providers.log import info, debug
 
 
-def get_dw_download_links(shared_state, url, mirror, title): # signature must align with other download link functions!
+def get_dw_download_links(shared_state, url, mirror, title, password):
+    """
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    DW source handler - fetches protected download links from DW site.
+    """
+
     dw = shared_state.values["config"]("Hostnames").get("dw")
     ajax_url = "https://" + dw + "/wp-admin/admin-ajax.php"
 
@@ -26,7 +32,7 @@ def get_dw_download_links(shared_state, url, mirror, title): # signature must al
         download_buttons = content.find_all("button", {"class": "show_link"})
     except:
         info(f"DW site has been updated. Grabbing download links for {title} not possible!")
-        return False
+        return {"links": []}
 
     download_links = []
     try:
@@ -62,4 +68,4 @@ def get_dw_download_links(shared_state, url, mirror, title): # signature must al
         info(f"DW site has been updated. Parsing download links for {title} not possible!")
         pass
 
-    return download_links
+    return {"links": download_links}

quasarr/downloads/sources/he.py

@@ -13,7 +13,13 @@ from quasarr.providers.log import info, debug
 hostname = "he"
 
 
-def get_he_download_links(shared_state, url, mirror, title):
+def get_he_download_links(shared_state, url, mirror, title, password):
+    """
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    HE source handler - fetches plain download links from HE pages.
+    """
+
     headers = {
         'User-Agent': shared_state.values["user_agent"],
     }
@@ -25,7 +31,7 @@ def get_he_download_links(shared_state, url, mirror, title):
         soup = BeautifulSoup(resp.text, 'html.parser')
     except Exception as e:
         info(f"{hostname}: could not fetch release for {title}: {e}")
-        return False
+        return {"links": [], "imdb_id": None}
 
     imdb_id = None
     try:
@@ -46,7 +52,7 @@ def get_he_download_links(shared_state, url, mirror, title):
     for retries in range(10):
         form = soup.find('form', id=re.compile(r'content-protector-access-form'))
         if not form:
-            return False
+            return {"links": [], "imdb_id": None}
 
         action = form.get('action') or url
         action_url = urljoin(resp.url, action)
@@ -104,7 +110,7 @@ def get_he_download_links(shared_state, url, mirror, title):
 
     if not links:
         info(f"No external download links found on {hostname} page for {title}")
-        return False
+        return {"links": [], "imdb_id": None}
 
     return {
         "links": links,

quasarr/downloads/sources/mb.py

@@ -10,7 +10,13 @@ from bs4 import BeautifulSoup
 from quasarr.providers.log import info, debug
 
 
-def get_mb_download_links(shared_state, url, mirror, title): # signature must align with other download link functions!
+def get_mb_download_links(shared_state, url, mirror, title, password):
+    """
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    MB source handler - fetches protected download links from MB pages.
+    """
+
     headers = {
         'User-Agent': shared_state.values["user_agent"],
     }
@@ -20,7 +26,7 @@ def get_mb_download_links(shared_state, url, mirror, title): # signature must al
         response.raise_for_status()
     except Exception as e:
         info(f"Failed to fetch page for {title or url}: {e}")
-        return False
+        return {"links": []}
 
     soup = BeautifulSoup(response.text, "html.parser")
 
@@ -42,6 +48,6 @@ def get_mb_download_links(shared_state, url, mirror, title): # signature must al
 
     if not download_links:
         info(f"No download links found for {title}. Site structure may have changed. - {url}")
-        return False
+        return {"links": []}
 
-    return download_links
+    return {"links": download_links}

quasarr/downloads/sources/nk.py

@@ -11,7 +11,13 @@ hostname = "nk"
 supported_mirrors = ["rapidgator", "ddownload"]
 
 
-def get_nk_download_links(shared_state, url, mirror, title):
+def get_nk_download_links(shared_state, url, mirror, title, password):
+    """
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    NK source handler - fetches protected download links from NK pages.
+    """
+
     host = shared_state.values["config"]("Hostnames").get(hostname)
     headers = {
         'User-Agent': shared_state.values["user_agent"],
@@ -24,7 +30,7 @@ def get_nk_download_links(shared_state, url, mirror, title):
         soup = BeautifulSoup(resp.text, 'html.parser')
     except Exception as e:
         info(f"{hostname}: could not fetch release page for {title}: {e}")
-        return False
+        return {"links": []}
 
     anchors = soup.select('a.btn-orange')
     candidates = []
@@ -51,4 +57,4 @@ def get_nk_download_links(shared_state, url, mirror, title):
     if not candidates:
         info(f"No external download links found on {hostname} page for {title}")
 
-    return candidates
+    return {"links": candidates}

quasarr/downloads/sources/nx.py

@@ -3,14 +3,28 @@
 # Project by https://github.com/rix1337
 
 import re
+from urllib.parse import urlparse
 
 import requests
-from bs4 import BeautifulSoup
 
 from quasarr.providers.log import info
 from quasarr.providers.sessions.nx import retrieve_and_validate_session
 
 
+def derive_mirror_from_url(url):
+    """Extract hoster name from URL hostname."""
+    try:
+        hostname = urlparse(url).netloc.lower()
+        if hostname.startswith('www.'):
+            hostname = hostname[4:]
+        parts = hostname.split('.')
+        if len(parts) >= 2:
+            return parts[-2]
+        return hostname
+    except:
+        return "unknown"
+
+
 def get_filer_folder_links_via_api(shared_state, url):
     try:
         headers = {
@@ -20,7 +34,7 @@ def get_filer_folder_links_via_api(shared_state, url):
 
         m = re.search(r"/folder/([A-Za-z0-9]+)", url)
         if not m:
-            return url  # not a folder URL
+            return url
 
         folder_hash = m.group(1)
         api_url = f"https://filer.net/api/folder/{folder_hash}"
@@ -33,7 +47,6 @@ def get_filer_folder_links_via_api(shared_state, url):
         files = data.get("files", [])
         links = []
 
-        # Build download URLs from their file hashes
         for f in files:
             file_hash = f.get("hash")
             if not file_hash:
@@ -41,14 +54,19 @@ def get_filer_folder_links_via_api(shared_state, url):
             dl_url = f"https://filer.net/get/{file_hash}"
             links.append(dl_url)
 
-        # Return extracted links or fallback
         return links if links else url
 
     except:
         return url
 
 
-def get_nx_download_links(shared_state, url, mirror, title): # signature must align with other download link functions!
+def get_nx_download_links(shared_state, url, mirror, title, password):
+    """
+    KEEP THE SIGNATURE EVEN IF SOME PARAMETERS ARE UNUSED!
+
+    NX source handler - auto-decrypts via site API and returns plain download links.
+    """
+
     nx = shared_state.values["config"]("Hostnames").get("nx")
 
     if f"{nx}/release/" not in url:
@@ -57,7 +75,7 @@ def get_nx_download_links(shared_state, url, mirror, title): # signature must al
     nx_session = retrieve_and_validate_session(shared_state)
     if not nx_session:
         info(f"Could not retrieve valid session for {nx}")
-        return []
+        return {"links": []}
 
     headers = {
         'User-Agent': shared_state.values["user_agent"],
@@ -81,13 +99,13 @@ def get_nx_download_links(shared_state, url, mirror, title): # signature must al
         except:
             info("Invalid response decrypting " + str(title) + " URL: " + str(url))
             shared_state.values["database"]("sessions").delete("nx")
-            return []
+            return {"links": []}
 
     if payload and any(key in payload for key in ("err", "error")):
         error_msg = payload.get("err") or payload.get("error")
         info(f"Error decrypting {title!r} URL: {url!r} - {error_msg}")
         shared_state.values["database"]("sessions").delete("nx")
-        return []
+        return {"links": []}
 
     try:
         decrypted_url = payload['link'][0]['url']
@@ -96,10 +114,13 @@ def get_nx_download_links(shared_state, url, mirror, title): # signature must al
                 urls = get_filer_folder_links_via_api(shared_state, decrypted_url)
             else:
                 urls = [decrypted_url]
-            return urls
+
+            # Convert to [[url, mirror], ...] format
+            links = [[u, derive_mirror_from_url(u)] for u in urls]
+            return {"links": links}
     except:
         pass
 
     info("Something went wrong decrypting " + str(title) + " URL: " + str(url))
     shared_state.values["database"]("sessions").delete("nx")
-    return []
+    return {"links": []}