#109 : Add adult content filtering functionality

db/config.py

@@ -19,6 +19,9 @@ class Settings(BaseSettings):
     prowlarr_api_key: str | None = None
     prowlarr_search_interval_hour: int = 24
     prowlarr_immediate_max_process: int = 10
+    adult_content_regex_keywords: str = (
+        r"\b(18\+|adult|porn|sex|xxx|nude|naked|erotic|sexy|18\s*plus|18\s*\+)\b"
+    )
 
     class Config:
         env_file = ".env"

scrapers/prowlarr.py

@@ -18,6 +18,7 @@ from scrapers.helpers import (
     UA_HEADER,
 )
 from utils.network import CircuitBreaker, batch_process_with_circuit_breaker
+from utils.parser import is_contain_18_plus_keywords
 from utils.torrent import extract_torrent_metadata
 
 
@@ -216,6 +217,12 @@ async def get_torrent_data_from_prowlarr(download_url: str) -> tuple[dict, bool]
 
 async def prowlarr_data_parser(meta_data: dict) -> tuple[dict, bool]:
     """Parse prowlarr data."""
+    if is_contain_18_plus_keywords(meta_data.get("title")):
+        logging.warning(
+            f"Skipping {meta_data.get('title')} due to adult content keywords."
+        )
+        return {}, False
+
     if meta_data.get("indexer") in [
         "Torlock",
         "YourBittorrent",

scrapers/torrentio.py

@@ -15,7 +15,7 @@ from scrapers.helpers import (
     update_torrent_series_streams_metadata,
     update_torrent_movie_streams_metadata,
 )
-from utils.parser import convert_size_to_bytes
+from utils.parser import convert_size_to_bytes, is_contain_18_plus_keywords
 from utils.validation_helper import is_video_file
 
 
@@ -123,6 +123,10 @@ async def store_and_parse_movie_stream_data(
     streams = []
     info_hashes = []
     for stream in stream_data:
+        if is_contain_18_plus_keywords(stream["title"]):
+            logging.warning(f"Stream contains 18+ keywords: {stream['title']}")
+            continue
+
         parsed_data = parse_stream_title(stream)
         if not parsed_data["seeders"]:
             continue
@@ -176,6 +180,10 @@ async def store_and_parse_series_stream_data(
     streams = []
     info_hashes = []
     for stream in stream_data:
+        if is_contain_18_plus_keywords(stream["title"]):
+            logging.warning(f"Stream contains 18+ keywords: {stream['title']}")
+            continue
+
         parsed_data = parse_stream_title(stream)
         if not parsed_data["seeders"]:
             continue

utils/parser.py

@@ -42,6 +42,10 @@ from streaming_providers.torbox.utils import (
 )
 
 ia = Cinemagoer()
+ADULT_CONTENT_KEYWORDS = re.compile(
+    settings.adult_content_regex_keywords,
+    re.IGNORECASE,
+)
 
 
 async def filter_and_sort_streams(
@@ -325,3 +329,10 @@ def generate_manifest(manifest: dict, user_data: UserData) -> dict:
 
     manifest["resources"] = resources
     return manifest
+
+
+def is_contain_18_plus_keywords(title: str) -> bool:
+    """
+    Check if the title contains 18+ keywords to filter out adult content.
+    """
+    return ADULT_CONTENT_KEYWORDS.search(title) is not None

utils/torrent.py

@@ -17,6 +17,8 @@ from anyio.streams.memory import MemoryObjectSendStream
 from demagnetize.core import Demagnetizer
 from torf import Magnet
 
+from utils.parser import is_contain_18_plus_keywords
+
 # remove logging from demagnetize
 logging.getLogger("demagnetize").setLevel(logging.CRITICAL)
 
@@ -71,6 +73,12 @@ def extract_torrent_metadata(content: bytes) -> dict:
             tracker[0].decode() for tracker in torrent_data.get(b"announce-list", [])
         ]
         torrent_name = info.get(b"name", b"").decode() or file_data[0]["filename"]
+        if is_contain_18_plus_keywords(torrent_name):
+            logging.warning(
+                f"Torrent name contains 18+ keywords: {torrent_name}. Skipping"
+            )
+            return {}
+
         largest_file = max(file_data, key=lambda x: x["size"])
 
         return {