mirror of
https://github.com/simonw/datasette.git
synced 2026-10-31 23:47:38 +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
79 lines
2.9 KiB
Diff
79 lines
2.9 KiB
Diff
diff --git a/datasette/utils/__init__.py b/datasette/utils/__init__.py
|
|
index 669d04c..7843b9b 100644
|
|
--- a/datasette/utils/__init__.py
|
|
+++ b/datasette/utils/__init__.py
|
|
@@ -1779,6 +1779,22 @@ def sha256_file(filepath, chunk_size=4096):
|
|
return hasher.hexdigest()
|
|
|
|
|
|
+_sha256_cache = {}
|
|
+
|
|
+
|
|
+def cached_sha256_file(filepath, chunk_size=4096):
|
|
+ # Re-hash only if the file has changed. st_ctime_ns cannot be set by
|
|
+ # tools that preserve mtime (cp -p, tar), st_ino catches replaced files
|
|
+ stat = os.stat(filepath)
|
|
+ key = (stat.st_ino, stat.st_size, stat.st_mtime_ns, stat.st_ctime_ns)
|
|
+ cached = _sha256_cache.get(filepath)
|
|
+ if cached is not None and cached[0] == key:
|
|
+ return cached[1]
|
|
+ digest = sha256_file(filepath, chunk_size=chunk_size)
|
|
+ _sha256_cache[filepath] = (key, digest)
|
|
+ return digest
|
|
+
|
|
+
|
|
async def calculate_etag(filepath, chunk_size=4096):
|
|
if filepath in _etag_cache:
|
|
return _etag_cache[filepath]
|
|
diff --git a/datasette/utils/asgi.py b/datasette/utils/asgi.py
|
|
index 2d4a6cf..64d47be 100644
|
|
--- a/datasette/utils/asgi.py
|
|
+++ b/datasette/utils/asgi.py
|
|
@@ -9,7 +9,12 @@ from urllib.parse import parse_qs, parse_qsl, urlunparse
|
|
import aiofiles
|
|
import aiofiles.os
|
|
|
|
-from datasette.utils import MultiParams, calculate_etag, error_body, sha256_file
|
|
+from datasette.utils import (
|
|
+ MultiParams,
|
|
+ cached_sha256_file,
|
|
+ calculate_etag,
|
|
+ error_body,
|
|
+)
|
|
from datasette.utils.multipart import (
|
|
DEFAULT_MAX_FIELD_SIZE,
|
|
DEFAULT_MAX_FIELDS,
|
|
@@ -496,7 +501,8 @@ def asgi_static(root_path, chunk_size=4096, headers=None, content_type=None):
|
|
hash_value = request.args.get("_hash")
|
|
if (
|
|
hash_value
|
|
- and hash_value == sha256_file(full_path, chunk_size=chunk_size)[:12]
|
|
+ and hash_value
|
|
+ == cached_sha256_file(full_path, chunk_size=chunk_size)[:12]
|
|
):
|
|
headers["Cache-Control"] = HASHED_STATIC_CACHE_CONTROL
|
|
etag = await calculate_etag(full_path, chunk_size=chunk_size)
|
|
diff --git a/tests/test_utils.py b/tests/test_utils.py
|
|
index 68d4a50..2de1d51 100644
|
|
--- a/tests/test_utils.py
|
|
+++ b/tests/test_utils.py
|
|
@@ -923,6 +923,19 @@ def test_sha256_file(tmp_path):
|
|
assert utils.sha256_file(path, chunk_size=2) == hashlib.sha256(b"hello").hexdigest()
|
|
|
|
|
|
+def test_cached_sha256_file(tmp_path):
|
|
+ path = tmp_path / "test.txt"
|
|
+ path.write_text("hello")
|
|
+ assert utils.cached_sha256_file(path) == hashlib.sha256(b"hello").hexdigest()
|
|
+ key, _ = utils._sha256_cache[path]
|
|
+ utils._sha256_cache[path] = (key, "cached")
|
|
+ assert utils.cached_sha256_file(path) == "cached"
|
|
+ # Modifying the file invalidates the cached hash
|
|
+ path.write_text("hello!")
|
|
+ assert utils.cached_sha256_file(path) == hashlib.sha256(b"hello!").hexdigest()
|
|
+ utils._sha256_cache.clear()
|
|
+
|
|
+
|
|
@pytest.mark.asyncio
|
|
async def test_calculate_etag(tmp_path):
|
|
path = tmp_path / "test.txt"
|