Support unicode urls filtering

bot/exts/filtering/_filter_lists/domain.py

@@ -1,5 +1,6 @@
 import re
 import typing
+import urllib.parse
 
 from bot.exts.filtering._filter_context import Event, FilterContext
 from bot.exts.filtering._filter_lists.filter_list import FilterList, ListType
@@ -51,7 +52,7 @@ async def actions_for(
             return None, [], {}
 
         text = clean_input(text)
-        urls = {match.group(1).lower().rstrip("/") for match in URL_RE.finditer(text)}
+        urls = {urllib.parse.unquote(match.group(1)).lower().rstrip("/") for match in URL_RE.finditer(text)}
         new_ctx = ctx.replace(content=urls)
 
         triggers = await self[ListType.DENY].filter_list_result(new_ctx)

bot/exts/filtering/_filters/domain.py

@@ -1,4 +1,5 @@
 import re
+import urllib.parse
 from typing import ClassVar
 from urllib.parse import urlparse
 
@@ -36,11 +37,11 @@ class DomainFilter(Filter):
 
     async def triggered_on(self, ctx: FilterContext) -> bool:
         """Searches for a domain within a given context."""
-        domain = tldextract.extract(self.content).registered_domain.lower()
+        domain = tldextract.extract(self.content).top_domain_under_public_suffix.lower()
 
         for found_url in ctx.content:
             extract = tldextract.extract(found_url)
-            if self.content.lower() in found_url and extract.registered_domain == domain:
+            if self.content.lower() in found_url and extract.top_domain_under_public_suffix == domain:
                 if self.extra_fields.only_subdomains:
                     if not extract.subdomain and not urlparse(f"https://{found_url}").path:
                         return False
@@ -59,4 +60,5 @@ async def process_input(cls, content: str, description: str) -> tuple[str, str]:
         match = URL_RE.fullmatch(content)
         if not match or not match.group(1):
             raise BadArgument(f"`{content}` is not a URL.")
-        return match.group(1), description
+        unquoted_content = urllib.parse.unquote(match.group(1))
+        return unquoted_content, description