From 65c586632ae421b45c084630f448e67f6c7a7a02 Mon Sep 17 00:00:00 2001
From: Arjun Satarkar <me@arjunsatarkar.net>
Date: Wed, 17 Jul 2024 15:55:04 +0530
Subject: wplink: support multiple links per message, improve efficiency

Added some sanity checks, a cache, and used HEAD rather than GET
requests. Checked the respone status code.
---
 wplink/info.json |  2 +-
 wplink/wplink.py | 47 ++++++++++++++++++++++++++++++++---------------
 2 files changed, 33 insertions(+), 16 deletions(-)

(limited to 'wplink')

diff --git a/wplink/info.json b/wplink/info.json
index bd066ba..297e4e0 100644
--- a/wplink/info.json
+++ b/wplink/info.json
@@ -1,4 +1,4 @@
 {
     "author": ["Arjun Satarkar"],
-    "requirements": ["aiohttp"]
+    "requirements": ["aiohttp", "async-lru"]
 }
diff --git a/wplink/wplink.py b/wplink/wplink.py
index 8c27166..0cdd003 100644
--- a/wplink/wplink.py
+++ b/wplink/wplink.py
@@ -1,6 +1,8 @@
 import aiohttp
+import async_lru
 import discord
 from redbot.core import commands
+import logging
 import re
 import urllib.parse
 
@@ -11,25 +13,40 @@ class WPLink(commands.Cog):
 
     @commands.Cog.listener()
     async def on_message(self, message: discord.Message):
-        wikilink_pattern = r"\[\[(.+)\]\]"
-        match = re.search(wikilink_pattern, message.content)
-        if match is not None:
-            title = match.group(1)
+        WIKILINK_PATTERN = r"\[\[(.+?)\]\]"
+        MAX_LINKS_PER_MESSAGE = 6
+        # Per https://www.mediawiki.org/wiki/Page_title_size_limitations
+        MAX_TITLE_LEN = 255
+
+        titles = re.findall(WIKILINK_PATTERN, message.content)
+        titles = titles[:MAX_LINKS_PER_MESSAGE]
+
+        formatted_page_urls = []
+        for title in titles:
+            if len(title) > MAX_TITLE_LEN:
+                continue
             page_url = await self.look_up_page(title)
             if page_url is not None:
-                await message.reply(
-                    f"<{page_url}>", allowed_mentions=discord.AllowedMentions.none()
-                )
+                formatted_page_urls.append(f"<{page_url}>")
+
+        if formatted_page_urls:
+            await message.reply(
+                ", ".join(formatted_page_urls),
+                allowed_mentions=discord.AllowedMentions.none(),
+            )
 
+    @async_lru.alru_cache(maxsize=512)
     async def look_up_page(self, title: str) -> str | None:
+        logging.info("Looking up page title %s", title)
+        MAX_URL_SIZE = 400
         query_url = f"https://en.wikipedia.org/wiki/Special:Search?search={urllib.parse.quote(title)}&go=Go"
         async with aiohttp.ClientSession() as session:
-            async with session.get(query_url) as response:
+            async with session.head(query_url, allow_redirects=True) as response:
+                if response.status != 200:
+                    return None
                 result_url = str(response.url)
-                return (
-                    result_url
-                    if not result_url.startswith(
-                        "https://en.wikipedia.org/wiki/Special:Search?"
-                    )
-                    else None
-                )
+                if len(result_url) > MAX_URL_SIZE or result_url.startswith(
+                    "https://en.wikipedia.org/wiki/Special:Search?"
+                ):
+                    return None
+                return result_url
-- 
cgit v1.2.3-57-g22cb