Added smart invite filtering and dumb domain filtering.

4 files changed, 89 insertions, 12 deletions

diff --git a/bot/__main__.py b/bot/__main__.py
index 4429c2a0d..3ecf8cb18 100644
--- a/bot/__main__.py
+++ b/bot/__main__.py
@@ -41,6 +41,7 @@ bot.load_extension("bot.cogs.logging")
 bot.load_extension("bot.cogs.modlog")
 bot.load_extension("bot.cogs.security")
 bot.load_extension("bot.cogs.events")
+bot.load_extension("bot.cogs.filtering")
 
 # Commands, etc
 bot.load_extension("bot.cogs.bigbrother")
diff --git a/bot/cogs/filtering.py b/bot/cogs/filtering.py
index 23b0d9e96..eee1dbd16 100644
--- a/bot/cogs/filtering.py
+++ b/bot/cogs/filtering.py
@@ -4,27 +4,71 @@ import re
 from discord import Message
 from discord.ext.commands import Bot
 
-from bot.constants import Channels
+from bot.constants import Channels, Filter
 
 log = logging.getLogger(__name__)
 
+INVITE_RE = (
+    r"(?:discord(?:[\.,]|dot)gg|"                     # Could be discord.gg/
+    r"discord(?:[\.,]|dot)com(?:\/|slash)invite|"     # or discord.com/invite/
+    r"discordapp(?:[\.,]|dot)com(?:\/|slash)invite|"  # or discordapp.com/invite/
+    r"discord(?:[\.,]|dot)me|"                        # or discord.me
+    r"discord(?:[\.,]|dot)io"                         # or discord.io.
+    r")(?:[\/]|slash)"                                # / or slash
+    r"([a-zA-Z0-9]+)"                                 # the invite code itself
+)
+
+URL_RE = "(https?://[^\s]+)"
+ZALGO_RE = r"[\u0300-\u036F\u0489]"
+
 
 class Filtering:
     """
     Filtering out invites, blacklisting domains,
-    and preventing certain expressions"""
+    and preventing certain expressions
+    """
 
     def __init__(self, bot: Bot):
         self.bot = bot
 
     async def on_message(self, msg: Message):
 
-        has_zalgo = await self._filter_zalgo(msg.content)
+        if msg.channel.id == Channels.devtest and not msg.author.bot:
 
-        if has_zalgo:
-            self.bot.get_channel(Channels.modlog).send(
-                content="ZALGO!"
-            )
+            has_zalgo = await self._has_zalgo(msg.content)
+            has_invites = await self._has_invites(msg.content)
+            has_urls = await self._has_urls(msg.content)
+
+            if has_zalgo:
+                await self.bot.get_channel(msg.channel.id).send(
+                    content="ZALGO!"
+                )
+
+            if has_invites:
+                await self.bot.get_channel(msg.channel.id).send(
+                    content="INVITES!"
+                )
+
+            if has_urls:
+                await self.bot.get_channel(msg.channel.id).send(
+                    content="EVIL ILLEGAL HITLER DOMAINS!"
+                )
+
+    @staticmethod
+    async def _has_urls(text):
+        """
+        Returns True if the text contains one of
+        the blacklisted URLs from the config file.
+        """
+
+        if not re.search(URL_RE, text):
+            return False
+
+        for url in Filter.domain_blacklist:
+            if url in text:
+                return True
+
+        return False
 
     @staticmethod
     async def _has_zalgo(text):
@@ -34,7 +78,36 @@ class Filtering:
         Zalgo range is \u0300 – \u036F and \u0489.
         """
 
-        return bool(re.search(r"[\u0300-\u036F\u0489]", text))
+        return bool(re.search(ZALGO_RE, text))
+
+    @staticmethod
+    async def _has_invites(text):
+        """
+        Returns True if the text contains an invite which
+        is not on the guild_invite_whitelist in config.yml.
+
+        Also catches a lot of common ways to try to cheat the system.
+        """
+
+        # Remove spaces to prevent cases like
+        # d i s c o r d . c o m / i n v i t e / p y t h o n
+        text = text.replace(" ", "")
+
+        # Remove backslashes to prevent escape character aroundfuckery like
+        # discord\.gg/gdudes-pony-farm
+        text = text.replace("\\", "")
+
+        invites = re.findall(INVITE_RE, text)
+        for invite in invites:
+
+            filter_invite = (
+                invite not in Filter.guild_invite_whitelist
+                and invite.lower() not in Filter.vanity_url_whitelist
+            )
+
+            if filter_invite:
+                return True
+        return False
 
 
 def setup(bot: Bot):
diff --git a/bot/constants.py b/bot/constants.py
index 834c14fd8..597c7c4a7 100644
--- a/bot/constants.py
+++ b/bot/constants.py
@@ -191,8 +191,8 @@ class Bot(metaclass=YAMLGetter):
     token: str
 
 
-class Filtering(metaclass=YAMLGetter):
-    section = "filtering"
+class Filter(metaclass=YAMLGetter):
+    section = "filter"
 
     filter_zalgo: bool
     filter_invites: bool
@@ -202,6 +202,7 @@ class Filtering(metaclass=YAMLGetter):
     domain_blacklist: List[str]
     watched_expressions: List[str]
     guild_invite_whitelist: List[str]
+    vanity_url_whitelist: List[str]
 
     channel_whitelist: List[int]
     role_whitelist: List[int]
diff --git a/config-default.yml b/config-default.yml
index eeeb5fb34..06117ef05 100644
--- a/config-default.yml
+++ b/config-default.yml
@@ -90,7 +90,7 @@ guild:
         helpers:                          267630620367257601
 
 
-filtering:
+filter:
 
     # What do we filter?
     filter_zalgo:      true
@@ -111,6 +111,9 @@ filtering:
         - 010z0Kw1A9ql5c1Qe  # Programming: Meme Edition
         - XBGetGp            # STEM
 
+    vanity_url_whitelist:
+        - python             # Python Discord
+
     # Censor doesn't apply to these
     channel_whitelist:
         - *ADMINS
@@ -126,7 +129,6 @@ filtering:
         - *MOD_ROLE
         - *OWNER_ROLE
 
-
 keys:
     deploy_bot:  !ENV "DEPLOY_BOT_KEY"
     deploy_site: !ENV "DEPLOY_SITE"