mirror of
https://github.com/simonw/datasette.git
synced 2026-11-02 16:37:38 +01:00
d1dbb40362Tools for profiling Datasette's request/response cycle in-process with Python 3.15's profiling.sampling, plus the patches proposed by the study (one per optimisation, and ALL-combined.diff). The patches are not applied to the code. A report with measurements follows. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01F8mEgRS8swZfYqwqrVMeo5
109 lines
4.1 KiB
Diff
109 lines
4.1 KiB
Diff
diff --git a/datasette/utils/asgi.py b/datasette/utils/asgi.py
|
|
index 2d4a6cf..980d003 100644
|
|
--- a/datasette/utils/asgi.py
|
|
+++ b/datasette/utils/asgi.py
|
|
@@ -1,5 +1,6 @@
|
|
import asyncio
|
|
import json
|
|
+import os
|
|
import re
|
|
from http.cookies import Morsel, SimpleCookie
|
|
from mimetypes import guess_type
|
|
@@ -470,15 +471,34 @@ HASHED_STATIC_CACHE_CONTROL = "max-age=31536000, immutable, public"
|
|
def asgi_static(root_path, chunk_size=4096, headers=None, content_type=None):
|
|
root_path = Path(root_path)
|
|
static_headers = {}
|
|
+ # (dev, inode) of the resolved root directory, and its resolved path
|
|
+ resolved_root = [None, None]
|
|
|
|
if headers:
|
|
static_headers = headers.copy()
|
|
|
|
+ def resolve_root():
|
|
+ # Resolving walks every path component, so reuse the last result for
|
|
+ # as long as the root still points at the same directory - a symlinked
|
|
+ # root that is re-pointed while Datasette is running is resolved again
|
|
+ try:
|
|
+ stat = os.stat(root_path)
|
|
+ except OSError:
|
|
+ return root_path.resolve()
|
|
+ if resolved_root[0] != (stat.st_dev, stat.st_ino):
|
|
+ resolved = root_path.resolve()
|
|
+ try:
|
|
+ stat = os.stat(resolved)
|
|
+ except OSError:
|
|
+ return resolved
|
|
+ resolved_root[:] = [(stat.st_dev, stat.st_ino), resolved]
|
|
+ return resolved_root[1]
|
|
+
|
|
async def inner_static(request, send):
|
|
path = request.scope["url_route"]["kwargs"]["path"]
|
|
headers = static_headers.copy()
|
|
try:
|
|
- full_path = (root_path / path).resolve().absolute()
|
|
+ full_path = (root_path / path).resolve()
|
|
except FileNotFoundError:
|
|
await asgi_send_html(send, "404: Directory not found", 404)
|
|
return
|
|
@@ -487,7 +507,7 @@ def asgi_static(root_path, chunk_size=4096, headers=None, content_type=None):
|
|
return
|
|
# Ensure full_path is within root_path to avoid weird "../" tricks
|
|
try:
|
|
- full_path.relative_to(root_path.resolve())
|
|
+ full_path.relative_to(resolve_root())
|
|
except ValueError:
|
|
await asgi_send_html(send, "404: Path not inside root path", 404)
|
|
return
|
|
diff --git a/tests/test_html.py b/tests/test_html.py
|
|
index 4dbcf5c..e2d3e54 100644
|
|
--- a/tests/test_html.py
|
|
+++ b/tests/test_html.py
|
|
@@ -167,6 +167,48 @@ def test_static_mounts():
|
|
assert response.status_code == 404
|
|
|
|
|
|
+@pytest.mark.asyncio
|
|
+async def test_static_mounts_symlinked_root_repointed(tmp_path):
|
|
+ # A symlinked --static root that is switched to a new directory while
|
|
+ # Datasette is running serves the new directory, and only that directory
|
|
+ (tmp_path / "v1").mkdir()
|
|
+ (tmp_path / "v2").mkdir()
|
|
+ (tmp_path / "v1" / "a.txt").write_text("one")
|
|
+ (tmp_path / "v1" / "old.txt").write_text("old")
|
|
+ (tmp_path / "v2" / "a.txt").write_text("two")
|
|
+ (tmp_path / "current").symlink_to("v1")
|
|
+ ds = Datasette(static_mounts=[("assets", str(tmp_path / "current"))])
|
|
+ app = ds.app()
|
|
+
|
|
+ async def get(path):
|
|
+ messages = []
|
|
+
|
|
+ async def receive():
|
|
+ return {"type": "http.request", "body": b"", "more_body": False}
|
|
+
|
|
+ async def send(message):
|
|
+ messages.append(message)
|
|
+
|
|
+ scope = {
|
|
+ "type": "http",
|
|
+ "http_version": "1.1",
|
|
+ "method": "GET",
|
|
+ "path": path,
|
|
+ "raw_path": path.encode(),
|
|
+ "query_string": b"",
|
|
+ "headers": [],
|
|
+ }
|
|
+ await app(scope, receive, send)
|
|
+ body = b"".join(m.get("body", b"") for m in messages[1:])
|
|
+ return messages[0]["status"], body
|
|
+
|
|
+ assert await get("/assets/a.txt") == (200, b"one")
|
|
+ (tmp_path / "next").symlink_to("v2")
|
|
+ (tmp_path / "next").replace(tmp_path / "current")
|
|
+ assert await get("/assets/a.txt") == (200, b"two")
|
|
+ assert (await get("/assets/../v1/old.txt"))[0] == 404
|
|
+
|
|
+
|
|
def test_static_mounts_hash_cache_control():
|
|
mount_path = pathlib.Path(__file__).parent
|
|
with make_app_client(static_mounts=[("custom-static", str(mount_path))]) as client:
|