forked from retoor/devplacepy
Compare commits
1
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
b1d4af7ca3 |
+1
-18
@@ -4,13 +4,12 @@ import asyncio
|
||||
import fcntl
|
||||
import logging
|
||||
import os
|
||||
import re
|
||||
import time
|
||||
from collections import defaultdict
|
||||
from contextlib import asynccontextmanager, contextmanager
|
||||
from pathlib import Path
|
||||
from fastapi import FastAPI, Request
|
||||
from fastapi.responses import HTMLResponse, RedirectResponse, Response
|
||||
from fastapi.responses import HTMLResponse, RedirectResponse
|
||||
from fastapi.staticfiles import StaticFiles
|
||||
from fastapi.exceptions import RequestValidationError
|
||||
from starlette.middleware.gzip import GZipMiddleware
|
||||
@@ -612,22 +611,6 @@ async def response_timing(request: Request, call_next):
|
||||
return response
|
||||
|
||||
|
||||
_KNOWN_CRAWLERS = re.compile(
|
||||
r"(google.*read.*aloud|googlebot|bingbot|slurp|duckduckbot|baiduspider|yandexbot)",
|
||||
re.IGNORECASE,
|
||||
)
|
||||
_AUTH_GATED_PATHS = frozenset({"/messages", "/notifications", "/game", "/admin"})
|
||||
|
||||
|
||||
@app.middleware("http")
|
||||
async def crawler_detection(request: Request, call_next):
|
||||
if request.method == "GET" and request.url.path in _AUTH_GATED_PATHS:
|
||||
ua = request.headers.get("user-agent", "")
|
||||
if _KNOWN_CRAWLERS.search(ua):
|
||||
return Response(status_code=204)
|
||||
return await call_next(request)
|
||||
|
||||
|
||||
app.add_middleware(GZipMiddleware, minimum_size=512, compresslevel=5)
|
||||
|
||||
|
||||
|
||||
@@ -17,7 +17,6 @@ async def robots_txt(request: Request):
|
||||
Disallow: /auth/
|
||||
Disallow: /messages/
|
||||
Disallow: /notifications/
|
||||
Disallow: /game/
|
||||
Disallow: /votes/
|
||||
Disallow: /avatar/
|
||||
Disallow: /follow/
|
||||
|
||||
@@ -328,3 +328,44 @@
|
||||
padding: 0.75rem;
|
||||
}
|
||||
}
|
||||
|
||||
/* ---- Markdown source/preview toggle ---- */
|
||||
.gist-markdown-toggle .gist-md-tabbar {
|
||||
display: flex;
|
||||
gap: 0;
|
||||
border-bottom: 1px solid var(--border);
|
||||
margin-bottom: 0;
|
||||
}
|
||||
|
||||
.gist-markdown-toggle .gist-md-tab {
|
||||
padding: 0.5em 1em;
|
||||
cursor: pointer;
|
||||
color: var(--text-muted);
|
||||
border: 1px solid transparent;
|
||||
border-bottom: none;
|
||||
border-radius: var(--radius) var(--radius) 0 0;
|
||||
background: transparent;
|
||||
font-size: 0.875rem;
|
||||
}
|
||||
|
||||
.gist-markdown-toggle .gist-md-tab.active {
|
||||
color: var(--text-primary);
|
||||
background: var(--bg-card);
|
||||
border-color: var(--border) var(--border) transparent;
|
||||
}
|
||||
|
||||
.gist-markdown-toggle .tab-pane {
|
||||
display: none;
|
||||
}
|
||||
|
||||
.gist-markdown-toggle .tab-pane.active {
|
||||
display: block;
|
||||
}
|
||||
|
||||
.gist-markdown-toggle .gist-markdown-preview {
|
||||
padding: 1em;
|
||||
border: 1px solid var(--border);
|
||||
border-radius: 0 0 var(--radius) var(--radius);
|
||||
min-height: 4em;
|
||||
background: var(--bg-card);
|
||||
}
|
||||
|
||||
@@ -35,6 +35,23 @@
|
||||
|
||||
{% if gist['language'] == 'markdown_rendered' %}
|
||||
<div class="gist-rendered-content rendered-content">{{ render_content(gist['source_code'], author_is_admin=is_admin(author)) }}</div>
|
||||
{% elif gist['language'] == 'markdown' %}
|
||||
<div class="gist-code-block gist-markdown-toggle" data-tabs>
|
||||
<nav class="gist-md-tabbar" role="tablist">
|
||||
<button type="button" class="gist-md-tab" data-tab="source" role="tab">Source</button>
|
||||
<button type="button" class="gist-md-tab" data-tab="preview" role="tab">Preview</button>
|
||||
</nav>
|
||||
<div data-tab-pane="source" class="tab-pane active" role="tabpanel">
|
||||
<div class="gist-code-header">
|
||||
<span class="gist-language-name">{{ language_name(gist['language']) }}</span>
|
||||
<button type="button" class="gist-copy-btn" data-copy="gist-code-content">Copy</button>
|
||||
</div>
|
||||
<pre><code id="gist-code-content" class="language-markdown">{{ gist['source_code'] }}</code></pre>
|
||||
</div>
|
||||
<div data-tab-pane="preview" class="tab-pane gist-markdown-preview rendered-content" role="tabpanel">
|
||||
{{ render_content(gist['source_code'], author_is_admin=is_admin(author)) }}
|
||||
</div>
|
||||
</div>
|
||||
{% else %}
|
||||
<div class="gist-code-block">
|
||||
<div class="gist-code-header">
|
||||
|
||||
@@ -196,8 +196,7 @@ def test_robots_txt_exists(app_server):
|
||||
assert "Sitemap:" in r.text
|
||||
|
||||
|
||||
def test_robots_disallows_auth_gated_paths(app_server):
|
||||
def test_robots_disallows_admin_and_uploads(app_server):
|
||||
r = requests.get(f"{BASE_URL}/robots.txt")
|
||||
required = ["/auth/", "/game/", "/messages/", "/notifications/", "/admin/", "/uploads/"]
|
||||
for path in required:
|
||||
assert f"Disallow: {path}" in r.text, f"Missing Disallow: {path}"
|
||||
assert "Disallow: /admin/" in r.text
|
||||
assert "Disallow: /uploads/" in r.text
|
||||
|
||||
@@ -1,50 +0,0 @@
|
||||
# retoor <retoor@molodetz.nl>
|
||||
|
||||
import requests
|
||||
from tests.conftest import BASE_URL
|
||||
|
||||
|
||||
def test_bot_blocked_on_auth_path(app_server):
|
||||
headers = {"User-Agent": "Google-Read-Aloud"}
|
||||
r = requests.get(f"{BASE_URL}/messages", headers=headers)
|
||||
assert r.status_code == 204
|
||||
assert r.content == b""
|
||||
|
||||
|
||||
def test_bot_blocked_on_game_path(app_server):
|
||||
headers = {"User-Agent": "Google-Read-Aloud"}
|
||||
r = requests.get(f"{BASE_URL}/game", headers=headers)
|
||||
assert r.status_code == 204
|
||||
assert r.content == b""
|
||||
|
||||
|
||||
def test_bot_blocked_on_admin_path(app_server):
|
||||
headers = {"User-Agent": "Google-Read-Aloud"}
|
||||
r = requests.get(f"{BASE_URL}/admin", headers=headers)
|
||||
assert r.status_code == 204
|
||||
assert r.content == b""
|
||||
|
||||
|
||||
def test_bot_not_blocked_on_public_path(app_server):
|
||||
headers = {"User-Agent": "Google-Read-Aloud"}
|
||||
r = requests.get(f"{BASE_URL}/", headers=headers)
|
||||
assert r.status_code == 200
|
||||
|
||||
|
||||
def test_normal_user_not_blocked(app_server):
|
||||
r = requests.get(f"{BASE_URL}/messages", allow_redirects=False)
|
||||
assert r.status_code == 303
|
||||
|
||||
|
||||
def test_known_crawler_user_agent_variants(app_server):
|
||||
ua_list = [
|
||||
"Googlebot",
|
||||
"Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)",
|
||||
"bingbot/2.0; +http://www.bing.com/bingbot.htm",
|
||||
"Mozilla/5.0 (compatible; DuckDuckBot-Https/1.1; ...)",
|
||||
"Mozilla/5.0 (compatible; Baiduspider/2.0; +http://www.baidu.com/search/spider.html)",
|
||||
]
|
||||
for ua in ua_list:
|
||||
headers = {"User-Agent": ua}
|
||||
r = requests.get(f"{BASE_URL}/admin", headers=headers)
|
||||
assert r.status_code == 204, f"Expected 204 for UA: {ua}"
|
||||
Reference in New Issue
Block a user