From 3b7ecfaa696028007d379c7dcfb7d1c346ebf455 Mon Sep 17 00:00:00 2001
From: Minecon724 <git@m724.eu>
Date: Fri, 17 Jan 2025 15:16:51 +0100
Subject: [PATCH] Sufficient update

---
 .gitignore                 |  1 +
 Dockerfile                 | 12 ++++++++++++
 README.md                  | 19 +++++++++++++++----
 requirements.txt           |  2 +-
 main.py => src/__main__.py | 31 ++++++++++++++++++++-----------
 5 files changed, 49 insertions(+), 16 deletions(-)
 create mode 100644 Dockerfile
 rename main.py => src/__main__.py (86%)

diff --git a/.gitignore b/.gitignore
index 16a179c..6dbc493 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,3 @@
 venv/
 cache/
+__pycache__/
\ No newline at end of file
diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000..73d7c26
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,12 @@
+FROM python:3.13-alpine
+
+COPY requirements.txt ./
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY src ./src
+EXPOSE 8000
+
+RUN useradd app
+USER app
+
+CMD ["hypercorn", "src.__main__:app", "--host", "0.0.0.0"]
\ No newline at end of file
diff --git a/README.md b/README.md
index 3b6f039..21f6ee4 100644
--- a/README.md
+++ b/README.md
@@ -1,7 +1,18 @@
-Change `base_url` and run it as you would run a quart app
+Change `BASE_URL` environment variable to your URL!
 
-This is a MITM between you and wikipedia, basically it impersonates wikipedia.
+Production:
 
-I don't know if that's legal, but I don't think you're going to get sued.
+```bash
+BASE_URL=wikipedia.localhost:8000 hypercorn src.__main__:app
+```
 
-It's easier to maintain this way. If somebody turns out to have a problem, I will do something about it
+Debug:
+
+```bash
+BASE_URL=wikipedia.localhost:8000 python3 src/
+```
+
+
+## Privacy
+
+Nothing is intentionally logged, however, it's possible to tell what pages were accessed by the instance and when, due to cache. And leaking happens.
diff --git a/requirements.txt b/requirements.txt
index 666a289..d24addd 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,2 +1,2 @@
-quart
 aiohttp[speedups]
+quart
\ No newline at end of file
diff --git a/main.py b/src/__main__.py
similarity index 86%
rename from main.py
rename to src/__main__.py
index dfb4967..c5e3314 100644
--- a/main.py
+++ b/src/__main__.py
@@ -3,17 +3,22 @@ import aiohttp, aiofiles
 import aiofiles.os
 from time import time
 from binascii import crc32
+from os import getenv
 import re
 
 app = Quart(__name__, static_folder=None)
-base_url = 'wikipedia.localhost:5000'
+base_proto = 'http'
+base_url = 'wikipedia.localhost:8000'
 
-def fix_url(text):
+print("Base URL is " + base_proto + "://" + base_url)
+
+def fix_url(text: str) -> str:
     pattern = r'((https:|http:)?//([\w-]+\.)+)(wikipedia|wikimedia)\.org'
+
     def repl(match):
         protocol = match.group(2)
         if protocol:
-            return 'http://' + match.group(3) + base_url
+            return base_proto + '://' + match.group(3) + base_url
         else:
             return '//' + match.group(3) + base_url
     return re.sub(pattern, repl, text)
@@ -22,7 +27,10 @@ def fix_url(text):
 @app.route('/<path:path>')
 async def proxy(path):
     host = request.headers.get('Host')
-    dep = host.split('.')[0]
+
+    dep = host[:-len(base_url)]
+    if dep != '':
+        dep = dep[:-1] # remove dot
 
     if not host.endswith(base_url):
         return '', 200
@@ -83,19 +91,21 @@ async def proxy(path):
             headers = {k.lower(): v for k,v in headers if k.lower() in ['cache-control', 'age', 'content-language', 'content-type', 'last-modified', 'date', 'x-content-type-options', 'location']}
             content = await response.content.read()
 
-            if 'location' in headers:
-                print(headers['location'])
-                print(request.url_root)
+            if 'location' in headers: # if a redirect
                 headers['location'] = fix_url(headers['location'])
 
-            if any([headers['content-type'].startswith(t) for t in ['text/html', 'text/javascript', 'application']]):
+            try:
                 content = fix_url(content.decode()).encode()
+            except: # if file is not a text file
+                pass
 
             headers['content-length'] = str(len(content))
+            headers['x-rp-cache'] = 'MISS'
 
             await aiofiles.os.makedirs(cache_path, exist_ok=True)
+            
             async with aiofiles.open(cache_path + '/_file', mode='wb') as f:
-                await f.write(int.to_bytes(int(time()-int(headers['age'])), 8))
+                await f.write(int.to_bytes(int(time() - int(headers['age'])), 8))
                 await f.write(int.to_bytes(response.status, 2))
                 await f.write(int.to_bytes(len(headers)))
                 for k, v in headers.items():
@@ -107,11 +117,10 @@ async def proxy(path):
                     await f.write(int.to_bytes(len(value_encoded)))
                     await f.write(value_encoded)
                 await f.write(content)
-                print('cache written')
            
             r = Response(content, status=response.status, headers=headers)
             return r
 
 if __name__ == "__main__":
     app.debug = True
-    app.run()
+    app.run(port=8000)