From 84e425bd3b10c49ae28b70d9998ef5ec2bf8868f Mon Sep 17 00:00:00 2001
From: Mathieu Virbel <mat@meltingrocks.com>
Date: Fri, 10 Nov 2023 19:06:18 +0100
Subject: [PATCH] server: fix slow profanity filter

---
 server/reflector/processors/types.py |  7 +++-
 server/reflector/redis_cache.py      | 50 ++++++++++++++++++++++++++++
 server/reflector/settings.py         |  1 +
 3 files changed, 57 insertions(+), 1 deletion(-)
 create mode 100644 server/reflector/redis_cache.py

diff --git a/server/reflector/processors/types.py b/server/reflector/processors/types.py
index 312f5433..93e565df 100644
--- a/server/reflector/processors/types.py
+++ b/server/reflector/processors/types.py
@@ -5,6 +5,7 @@ from pathlib import Path
 
 from profanityfilter import ProfanityFilter
 from pydantic import BaseModel, PrivateAttr
+from reflector.redis_cache import redis_cache
 
 PUNC_RE = re.compile(r"[.;:?!…]")
 
@@ -68,10 +69,14 @@ class Transcript(BaseModel):
         # Uncensored text
         return "".join([word.text for word in self.words])
 
+    @redis_cache(prefix="profanity", duration=3600 * 24 * 7)
+    def _get_censored_text(self, text: str):
+        return profanity_filter.censor(text).strip()
+
     @property
     def text(self):
         # Censored text
-        return profanity_filter.censor(self.raw_text).strip()
+        return self._get_censored_text(self.raw_text)
 
     @property
     def human_timestamp(self):
diff --git a/server/reflector/redis_cache.py b/server/reflector/redis_cache.py
new file mode 100644
index 00000000..c31471cf
--- /dev/null
+++ b/server/reflector/redis_cache.py
@@ -0,0 +1,50 @@
+import functools
+import json
+
+import redis
+from reflector.settings import settings
+
+redis_clients = {}
+
+
+def get_redis_client(db=0):
+    """
+    Get a Redis client for the specified database.
+    """
+    if db not in redis_clients:
+        redis_clients[db] = redis.StrictRedis(
+            host=settings.REDIS_HOST,
+            port=settings.REDIS_PORT,
+            db=db,
+        )
+    return redis_clients[db]
+
+
+def redis_cache(prefix="cache", duration=3600, db=settings.REDIS_CACHE_DB, argidx=1):
+    """
+    Cache the result of a function in Redis.
+    """
+
+    def decorator(func):
+        @functools.wraps(func)
+        def wrapper(*args, **kwargs):
+            # Check if the first argument is a string
+            if len(args) < (argidx + 1) or not isinstance(args[argidx], str):
+                return func(*args, **kwargs)
+
+            # Compute the cache key based on the arguments and prefix
+            cache_key = prefix + ":" + args[argidx]
+            redis_client = get_redis_client(db=db)
+            cached_result = redis_client.get(cache_key)
+
+            if cached_result:
+                return json.loads(cached_result.decode("utf-8"))
+
+            # If the result is not cached, call the original function
+            result = func(*args, **kwargs)
+            redis_client.setex(cache_key, duration, json.dumps(result))
+            return result
+
+        return wrapper
+
+    return decorator
diff --git a/server/reflector/settings.py b/server/reflector/settings.py
index 7ce91195..013ab9b7 100644
--- a/server/reflector/settings.py
+++ b/server/reflector/settings.py
@@ -124,6 +124,7 @@ class Settings(BaseSettings):
     # Redis
     REDIS_HOST: str = "localhost"
     REDIS_PORT: int = 6379
+    REDIS_CACHE_DB: int = 2
 
     # Secret key
     SECRET_KEY: str = "changeme-f02f86fd8b3e4fd892c6043e5a298e21"