cython path_matcher added to reduce time on hot operations

This commit is contained in:
Илья Глазунов 2025-12-03 12:54:45 +03:00
parent 6c50a35aa3
commit 5d863bc97c
19 changed files with 1387 additions and 392 deletions

16
.gitignore vendored
View File

@ -13,4 +13,18 @@ static/*
.coverage
docs/
dist/
dist/
build/
# Cython generated files
*.c
*.so
*.pyd
*.html
*.egg-info/
# IDE
.idea/
.vscode/
*.swp
*.swo

View File

@ -1,4 +1,4 @@
.PHONY: help install build clean test lint format run dev-install dev-deps check release-patch release-minor release-major pipeline-check
.PHONY: help install build build-cython clean test lint format run dev-install dev-deps check release-patch release-minor release-major pipeline-check benchmark
PYTHON = python3
POETRY = poetry
@ -21,12 +21,14 @@ help:
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "install-package" "Installing package locally"
@echo ""
@echo "$(YELLOW)Building:$(NC)"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "build" "Building package"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "build" "Building package (with Cython)"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "build-cython" "Building Cython extensions only"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "clean" "Cleaning temporary files"
@echo ""
@echo "$(YELLOW)Testing:$(NC)"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "test" "Running tests"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "test-cov" "Running tests with coverage"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "benchmark" "Running performance benchmarks"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "lint" "Checking code with linters"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "format" "Formatting code"
@printf " $(YELLOW)%-20s$(CYAN) %s$(NC)\n" "check" "Lint and test"
@ -75,18 +77,27 @@ dev-deps:
@echo "$(GREEN)Installing additional tools...$(NC)"
$(POETRY) add --group dev pytest pytest-cov black isort mypy flake8
build: clean
build: clean build-cython
@echo "$(GREEN)Building package...$(NC)"
$(POETRY) build
build-cython:
@echo "$(GREEN)Building Cython extensions...$(NC)"
$(POETRY) run python scripts/build_cython.py build_ext --inplace || echo "$(YELLOW)Cython build skipped (optional)$(NC)"
clean:
@echo "$(GREEN)Cleaning temporary files...$(NC)"
rm -rf dist/
rm -rf build/
rm -rf *.egg-info/
find . -type d -name __pycache__ -exec rm -rf {} +
find . -type f -name "*.pyc" -delete
find . -type f -name "*.pyo" -delete
find . -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
find . -type f -name "*.pyc" -delete 2>/dev/null || true
find . -type f -name "*.pyo" -delete 2>/dev/null || true
@# Cython artifacts
find $(PACKAGE_NAME) -type f -name "*.c" -delete 2>/dev/null || true
find $(PACKAGE_NAME) -type f -name "*.so" -delete 2>/dev/null || true
find $(PACKAGE_NAME) -type f -name "*.pyd" -delete 2>/dev/null || true
find $(PACKAGE_NAME) -type f -name "*.html" -delete 2>/dev/null || true
test:
@echo "$(GREEN)Running tests...$(NC)"
@ -98,16 +109,20 @@ test-cov:
lint:
@echo "$(GREEN)Checking code with linters...$(NC)"
$(POETRY) run flake8 $(PACKAGE_NAME)/
$(POETRY) run flake8 $(PACKAGE_NAME)/ --exclude='*.pyx,*.pxd'
$(POETRY) run mypy $(PACKAGE_NAME)/
format:
@echo "$(GREEN)Formatting code...$(NC)"
$(POETRY) run black $(PACKAGE_NAME)/
$(POETRY) run isort $(PACKAGE_NAME)/
$(POETRY) run black $(PACKAGE_NAME)/ --exclude='\.pyx$$'
$(POETRY) run isort $(PACKAGE_NAME)/ --skip-glob='*.pyx'
check: lint test
benchmark: build-cython
@echo "$(GREEN)Running benchmarks...$(NC)"
$(POETRY) run python benchmarks/bench_path_matcher.py
run:
@echo "$(GREEN)Starting server in development mode...$(NC)"
$(POETRY) run python run.py --debug

View File

@ -0,0 +1,215 @@
#!/usr/bin/env python3
"""
Benchmark script for path_matcher performance comparison.
Compares:
- Pure Python implementation
- Cython implementation (if available)
- Original MountedApp from asgi_mount.py
Usage:
python benchmarks/bench_path_matcher.py
"""
import time
import statistics
from typing import Callable, List, Tuple
from pyserve._path_matcher_py import (
FastMountedPath as PyFastMountedPath,
FastMountManager as PyFastMountManager,
path_matches_prefix as py_path_matches_prefix,
)
try:
from pyserve._path_matcher import (
FastMountedPath as CyFastMountedPath,
FastMountManager as CyFastMountManager,
path_matches_prefix as cy_path_matches_prefix,
)
CYTHON_AVAILABLE = True
except ImportError:
CYTHON_AVAILABLE = False
print("Cython module not compiled. Run: python setup_cython.py build_ext --inplace\n")
from pyserve.asgi_mount import MountedApp
def benchmark(func: Callable, iterations: int = 100000) -> Tuple[float, float]:
times = []
for _ in range(1000):
func()
for _ in range(iterations):
start = time.perf_counter_ns()
func()
end = time.perf_counter_ns()
times.append(end - start)
return statistics.mean(times), statistics.stdev(times)
def format_time(ns: float) -> str:
if ns < 1000:
return f"{ns:.1f} ns"
elif ns < 1_000_000:
return f"{ns/1000:.2f} µs"
else:
return f"{ns/1_000_000:.2f} ms"
def run_benchmarks():
print("=" * 70)
print("PATH MATCHER BENCHMARK")
print("=" * 70)
print()
# Test paths
mount_path = "/api/v1"
test_paths = [
"/api/v1/users/123/posts", # Matching - long
"/api/v1", # Matching - exact
"/api/v2/users", # Not matching - similar prefix
"/other/path", # Not matching - completely different
]
iterations = 100000
# =========================================================================
# Benchmark 1: Single path matching
# =========================================================================
print("BENCHMARK 1: Single Path Matching")
print("-" * 70)
print(f" Mount path: {mount_path}")
print(f" Iterations: {iterations:,}")
print()
results = {}
# Original MountedApp
original_mount = MountedApp(mount_path, app=None, name="test") # type: ignore
for test_path in test_paths:
print(f" Test path: {test_path}")
# Original
mean, std = benchmark(lambda: original_mount.matches(test_path), iterations)
results[("Original", test_path)] = mean
print(f" Original MountedApp: {format_time(mean):>12} ± {format_time(std)}")
# Pure Python
py_mount = PyFastMountedPath(mount_path)
mean, std = benchmark(lambda: py_mount.matches(test_path), iterations)
results[("Python", test_path)] = mean
print(f" Pure Python: {format_time(mean):>12} ± {format_time(std)}")
# Cython (if available)
if CYTHON_AVAILABLE:
cy_mount = CyFastMountedPath(mount_path)
mean, std = benchmark(lambda: cy_mount.matches(test_path), iterations)
results[("Cython", test_path)] = mean
print(f" Cython: {format_time(mean):>12} ± {format_time(std)}")
print()
# =========================================================================
# Benchmark 2: Mount Manager lookup
# =========================================================================
print()
print("BENCHMARK 2: Mount Manager Lookup (10 mounts)")
print("-" * 70)
# Setup managers with 10 mounts
mount_paths = [f"/api/v{i}" for i in range(10)]
py_manager = PyFastMountManager()
for p in mount_paths:
py_manager.add_mount(PyFastMountedPath(p, name=p))
if CYTHON_AVAILABLE:
cy_manager = CyFastMountManager()
for p in mount_paths:
cy_manager.add_mount(CyFastMountedPath(p, name=p))
test_lookups = [
"/api/v5/users/123", # Middle mount
"/api/v0/items", # First mount (longest)
"/api/v9/data", # Last mount
"/other/not/found", # No match
]
for test_path in test_lookups:
print(f" Lookup path: {test_path}")
# Pure Python
mean, std = benchmark(lambda: py_manager.get_mount(test_path), iterations)
print(f" Pure Python: {format_time(mean):>12} ± {format_time(std)}")
# Cython
if CYTHON_AVAILABLE:
mean, std = benchmark(lambda: cy_manager.get_mount(test_path), iterations)
print(f" Cython: {format_time(mean):>12} ± {format_time(std)}")
print()
# =========================================================================
# Benchmark 3: Combined match + modify
# =========================================================================
print()
print("BENCHMARK 3: Combined Match + Modify Path")
print("-" * 70)
from pyserve._path_matcher_py import match_and_modify_path as py_match_modify
if CYTHON_AVAILABLE:
from pyserve._path_matcher import match_and_modify_path as cy_match_modify
test_path = "/api/v1/users/123/posts"
print(f" Test path: {test_path}")
print(f" Mount path: {mount_path}")
print()
# Original (separate calls)
def original_match_modify():
if original_mount.matches(test_path):
return original_mount.get_modified_path(test_path)
return None
mean, std = benchmark(original_match_modify, iterations)
print(f" Original (2 calls): {format_time(mean):>12} ± {format_time(std)}")
# Pure Python combined
mean, std = benchmark(lambda: py_match_modify(test_path, mount_path), iterations)
print(f" Pure Python (combined): {format_time(mean):>12} ± {format_time(std)}")
# Cython combined
if CYTHON_AVAILABLE:
mean, std = benchmark(lambda: cy_match_modify(test_path, mount_path), iterations)
print(f" Cython (combined): {format_time(mean):>12} ± {format_time(std)}")
# =========================================================================
# Summary
# =========================================================================
print()
print("=" * 70)
print("SUMMARY")
print("=" * 70)
if CYTHON_AVAILABLE:
print("Cython module is available and was benchmarked")
else:
print("Cython module not available - only Pure Python was benchmarked")
print(" To build Cython module:")
print(" 1. Install Cython: pip install cython")
print(" 2. Build: python setup_cython.py build_ext --inplace")
print()
print("The optimized path matcher provides:")
print(" - Pre-computed path length and trailing slash")
print(" - Boundary-aware prefix matching (prevents /api matching /api-v2)")
print(" - Combined match+modify operation to reduce function calls")
print(" - Longest-prefix-first ordering in MountManager")
if __name__ == "__main__":
run_benchmarks()

View File

@ -12,6 +12,16 @@ warn_unused_ignores = True
warn_no_return = True
warn_unreachable = True
strict_equality = True
exclude = (?x)(
^pyserve/_path_matcher\.pyx$
)
[mypy-tests.*]
disallow_untyped_defs = False
[mypy-pyserve._path_matcher]
ignore_missing_imports = True
follow_imports = skip
[mypy-django.*]
ignore_missing_imports = True

72
poetry.lock generated
View File

@ -274,6 +274,55 @@ files = [
[package.extras]
toml = ["tomli ; python_full_version <= \"3.11.0a6\""]
[[package]]
name = "cython"
version = "3.2.2"
description = "The Cython compiler for writing C extensions in the Python language."
optional = false
python-versions = ">=3.8"
groups = ["dev"]
files = [
{file = "cython-3.2.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:b5afac4e77e71a9010dc7fd3191ced00f9b12b494dd7525c140781054ce63a73"},
{file = "cython-3.2.2-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9cd2ede6af225499ad22888dbfb13b92d71fc1016f401ee637559a5831b177c2"},
{file = "cython-3.2.2-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8c9265b3e84ae2d999b7c3165c683e366bbbbbe4346468055ca2366fe013f2df"},
{file = "cython-3.2.2-cp310-cp310-win_amd64.whl", hash = "sha256:d7b3447b2005dffc5f276d420a480d2b57d15091242652d410b6a46fb00ed251"},
{file = "cython-3.2.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:d140c2701cbb8cf960300cf1b67f3b4fa9d294d32e51b85f329bff56936a82fd"},
{file = "cython-3.2.2-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:50bbaabee733fd2780985e459fc20f655e02def83e8eff10220ad88455a34622"},
{file = "cython-3.2.2-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a9509f1e9c41c86b790cff745bb31927bbc861662a3b462596d71d3d2a578abb"},
{file = "cython-3.2.2-cp311-cp311-win_amd64.whl", hash = "sha256:034ab96cb8bc8e7432bc27491f8d66f51e435b1eb21ddc03aa844be8f21ad847"},
{file = "cython-3.2.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:692a41c8fe06fb2dc55ca2c8d71c80c469fd16fe69486ed99f3b3cbb2d3af83f"},
{file = "cython-3.2.2-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:098590c1dc309f8a0406ade031963a95a87714296b425539f9920aebf924560d"},
{file = "cython-3.2.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a3898c076e9c458bcb3e4936187919fda5f5365fe4c567d35d2b003444b6f3fe"},
{file = "cython-3.2.2-cp312-cp312-win_amd64.whl", hash = "sha256:2b910b89a2a71004064c5e890b9512a251eda63fae252caa0feb9835057035f9"},
{file = "cython-3.2.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:aa24cd0bdab27ca099b2467806c684404add597c1108e07ddf7b6471653c85d7"},
{file = "cython-3.2.2-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:60f4aa425e1ff98abf8d965ae7020f06dd2cbc01dbd945137d2f9cca4ff0524a"},
{file = "cython-3.2.2-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a473df474ba89e9fee81ee82b31062a267f9e598096b222783477e56d02ad12c"},
{file = "cython-3.2.2-cp313-cp313-win_amd64.whl", hash = "sha256:b4df52101209817fde7284cf779156f79142fb639b1d7840f11680ff4bb30604"},
{file = "cython-3.2.2-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:177faf4d61e9f2d4d2db61194ac9ec16d3fe3041c1b6830f871a01935319eeb3"},
{file = "cython-3.2.2-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8db28aef793c81dc69383b619ca508668998aaf099cd839d3cbae85184cce744"},
{file = "cython-3.2.2-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3de43a5786033a27fae1c882feb5ff0d023c38b83356e6800c1be0bcd6cf9f11"},
{file = "cython-3.2.2-cp314-cp314-win_amd64.whl", hash = "sha256:fed44d0ab2d36f1b0301c770b0dafec23bcb9700d58e7769cd6d9136b3304c11"},
{file = "cython-3.2.2-cp38-cp38-macosx_11_0_arm64.whl", hash = "sha256:e7200309b81f4066cf36a96efeec646716ca74afd73d159045169263db891133"},
{file = "cython-3.2.2-cp38-cp38-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8e72ee88a9a5381d30a6da116a3c8352730b9b038a49ed9bc5c3d0ed6d69b06c"},
{file = "cython-3.2.2-cp38-cp38-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0e35ff0f1bb3a7a5c40afb8fb540e4178b6551909f10748bf39d323f8140ccf3"},
{file = "cython-3.2.2-cp38-cp38-win_amd64.whl", hash = "sha256:b223c1f84c3420c24f6a4858e979524bd35a79437a5839e29d41201c87ed119d"},
{file = "cython-3.2.2-cp39-abi3-macosx_10_9_x86_64.whl", hash = "sha256:a6387e3ad31342443916db9a419509935fddd8d4cbac34aab9c895ae55326a56"},
{file = "cython-3.2.2-cp39-abi3-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:436eb562d0affbc0b959f62f3f9c1ed251b9499e4f29c1d19514ae859894b6bf"},
{file = "cython-3.2.2-cp39-abi3-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:f560ff3aea5b5df93853ec7bf1a1e9623d6d511f4192f197559aca18fca43392"},
{file = "cython-3.2.2-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:d8c93fe128b58942832b1fcac96e48f93c2c69b569eff0d38d30fb5995fecfa0"},
{file = "cython-3.2.2-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:b4fe499eed7cd70b2aa4e096b9ce2588f5e6fdf049b46d40a5e55efcde6e4904"},
{file = "cython-3.2.2-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:14432d7f207245a3c35556155873f494784169297b28978a6204f1c60d31553e"},
{file = "cython-3.2.2-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:820c4a99dbf6b3e6c0300be42b4040b501eff0e1feeb80cfa52c48a346fb0df2"},
{file = "cython-3.2.2-cp39-abi3-win32.whl", hash = "sha256:826cad0ad43ab05a26e873b5d625f64d458dc739ec6fdeecab848b60a91c4252"},
{file = "cython-3.2.2-cp39-abi3-win_arm64.whl", hash = "sha256:5f818d40bbcf17e2089e2de7840f0de1c0ca527acf9b044aba79d5f5d8a5bdba"},
{file = "cython-3.2.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:ff07e784ea748225bbdea07fec0ac451379e9e41a0a84cb57b36db19dd01ae71"},
{file = "cython-3.2.2-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:aff11412ed5fc78bd8b148621f4d1034fcad6cfcba468c20cd9f327b4f61ec3e"},
{file = "cython-3.2.2-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ca18d9d53c0e2f0c9347478b37532b46e0dc34c704e052ab1b0d8b21a290fc0f"},
{file = "cython-3.2.2-cp39-cp39-win_amd64.whl", hash = "sha256:86b1d39a1ea974dd16fe3bcef0df7b64dadd0bd38d05a339f287b48d37cb109f"},
{file = "cython-3.2.2-py3-none-any.whl", hash = "sha256:13b99ecb9482aff6a6c12d1ca6feef6940c507af909914b49f568de74fa965fb"},
{file = "cython-3.2.2.tar.gz", hash = "sha256:c3add3d483acc73129a61d105389344d792c17e7c1cee24863f16416bd071634"},
]
[[package]]
name = "django"
version = "5.2.9"
@ -1131,6 +1180,27 @@ files = [
{file = "pyyaml-6.0.2.tar.gz", hash = "sha256:d584d9ec91ad65861cc08d42e834324ef890a082e591037abe114850ff7bbc3e"},
]
[[package]]
name = "setuptools"
version = "80.9.0"
description = "Easily download, build, install, upgrade, and uninstall Python packages"
optional = false
python-versions = ">=3.9"
groups = ["dev"]
files = [
{file = "setuptools-80.9.0-py3-none-any.whl", hash = "sha256:062d34222ad13e0cc312a4c02d73f059e86a4acbfbdea8f8f76b28c99f306922"},
{file = "setuptools-80.9.0.tar.gz", hash = "sha256:f36b47402ecde768dbfafc46e8e4207b4360c654f1f3bb84475f0a28628fb19c"},
]
[package.extras]
check = ["pytest-checkdocs (>=2.4)", "pytest-ruff (>=0.2.1) ; sys_platform != \"cygwin\"", "ruff (>=0.8.0) ; sys_platform != \"cygwin\""]
core = ["importlib_metadata (>=6) ; python_version < \"3.10\"", "jaraco.functools (>=4)", "jaraco.text (>=3.7)", "more_itertools", "more_itertools (>=8.8)", "packaging (>=24.2)", "platformdirs (>=4.2.2)", "tomli (>=2.0.1) ; python_version < \"3.11\"", "wheel (>=0.43.0)"]
cover = ["pytest-cov"]
doc = ["furo", "jaraco.packaging (>=9.3)", "jaraco.tidelift (>=1.4)", "pygments-github-lexers (==0.0.5)", "pyproject-hooks (!=1.1)", "rst.linker (>=1.9)", "sphinx (>=3.5)", "sphinx-favicon", "sphinx-inline-tabs", "sphinx-lint", "sphinx-notfound-page (>=1,<2)", "sphinx-reredirects", "sphinxcontrib-towncrier", "towncrier (<24.7)"]
enabler = ["pytest-enabler (>=2.2)"]
test = ["build[virtualenv] (>=1.0.3)", "filelock (>=3.4.0)", "ini2toml[lite] (>=0.14)", "jaraco.develop (>=7.21) ; python_version >= \"3.9\" and sys_platform != \"cygwin\"", "jaraco.envs (>=2.2)", "jaraco.path (>=3.7.2)", "jaraco.test (>=5.5)", "packaging (>=24.2)", "pip (>=19.1)", "pyproject-hooks (!=1.1)", "pytest (>=6,!=8.1.*)", "pytest-home (>=0.5)", "pytest-perf ; sys_platform != \"cygwin\"", "pytest-subprocess", "pytest-timeout", "pytest-xdist (>=3)", "tomli-w (>=1.0.0)", "virtualenv (>=13.0.0)", "wheel (>=0.44.0)"]
type = ["importlib_metadata (>=7.0.2) ; python_version < \"3.10\"", "jaraco.develop (>=7.21) ; sys_platform != \"cygwin\"", "mypy (==1.14.*)", "pytest-mypy"]
[[package]]
name = "sniffio"
version = "1.3.1"
@ -1551,4 +1621,4 @@ wsgi = ["a2wsgi"]
[metadata]
lock-version = "2.1"
python-versions = ">=3.12"
content-hash = "a08668c23222843b27b3977933c93b261328f43e90f22f35212c6c6f6030e3dc"
content-hash = "32ebf260f6792987cb4236fe29ad3329374e063504d507b5a0319684e24a30a8"

View File

@ -52,7 +52,7 @@ all-frameworks = [
[build-system]
requires = ["poetry-core>=2.0.0,<3.0.0"]
requires = ["poetry-core>=2.0.0,<3.0.0", "setuptools", "cython>=3.0.0"]
build-backend = "poetry.core.masonry.api"
[tool.black]
@ -95,4 +95,6 @@ isort = "^6.0.1"
mypy = "^1.17.1"
flake8 = "^7.3.0"
pytest-asyncio = "^1.3.0"
cython = "^3.0.0"
setuptools = "^80.0.0"

View File

@ -5,17 +5,17 @@ PyServe - HTTP web server written on Python
__version__ = "0.8.0"
__author__ = "Ilya Glazunov"
from .server import PyServeServer
from .config import Config
from .asgi_mount import (
ASGIAppLoader,
ASGIMountManager,
MountedApp,
create_django_app,
create_fastapi_app,
create_flask_app,
create_django_app,
create_starlette_app,
)
from .config import Config
from .server import PyServeServer
__all__ = [
"PyServeServer",

225
pyserve/_path_matcher.pyx Normal file
View File

@ -0,0 +1,225 @@
# cython: language_level=3
# cython: boundscheck=False
# cython: wraparound=False
# cython: cdivision=True
"""
Fast path matching module for PyServe.
This Cython module provides optimized path matching operations
for ASGI mount routing, significantly reducing overhead on hot paths.
"""
from cpython.object cimport PyObject
cdef class FastMountedPath:
cdef:
str _path
str _path_with_slash
Py_ssize_t _path_len
bint _is_root
public str name
public bint strip_path
def __cinit__(self):
self._path = ""
self._path_with_slash = "/"
self._path_len = 0
self._is_root = 1
self.name = ""
self.strip_path = 1
def __init__(self, str path, str name="", bint strip_path=True):
cdef Py_ssize_t path_len
path_len = len(path)
if path_len > 1 and path[path_len - 1] == '/':
path = path[:path_len - 1]
self._path = path
self._path_len = len(path)
self._is_root = 1 if (path == "" or path == "/") else 0
self._path_with_slash = path + "/" if self._is_root == 0 else "/"
self.name = name if name else path
self.strip_path = 1 if strip_path else 0
@property
def path(self) -> str:
return self._path
cpdef bint matches(self, str request_path):
cdef Py_ssize_t req_len
if self._is_root:
return 1
req_len = len(request_path)
if req_len < self._path_len:
return 0
if req_len == self._path_len:
return 1 if request_path == self._path else 0
if request_path[self._path_len] == '/':
return 1 if request_path[:self._path_len] == self._path else 0
return 0
cpdef str get_modified_path(self, str original_path):
cdef str new_path
if not self.strip_path:
return original_path
if self._is_root:
return original_path
new_path = original_path[self._path_len:]
if not new_path:
return "/"
return new_path
def __repr__(self):
return f"FastMountedPath(path={self._path!r}, name={self.name!r})"
def _get_path_len_neg(mount):
return -len(mount.path)
cdef class FastMountManager:
cdef:
list _mounts
int _mount_count
def __cinit__(self):
self._mounts = []
self._mount_count = 0
def __init__(self):
self._mounts = []
self._mount_count = 0
cpdef void add_mount(self, FastMountedPath mount):
self._mounts.append(mount)
self._mounts = sorted(self._mounts, key=_get_path_len_neg, reverse=False)
self._mount_count = len(self._mounts)
cpdef FastMountedPath get_mount(self, str request_path):
cdef:
int i
FastMountedPath mount
for i in range(self._mount_count):
mount = <FastMountedPath>self._mounts[i]
if mount.matches(request_path):
return mount
return None
cpdef bint remove_mount(self, str path):
cdef:
int i
Py_ssize_t path_len
FastMountedPath mount
path_len = len(path)
if path_len > 1 and path[path_len - 1] == '/':
path = path[:path_len - 1]
for i in range(self._mount_count):
mount = <FastMountedPath>self._mounts[i]
if mount._path == path:
del self._mounts[i]
self._mount_count -= 1
return 1
return 0
@property
def mounts(self) -> list:
return list(self._mounts)
@property
def mount_count(self) -> int:
return self._mount_count
cpdef list list_mounts(self):
cdef:
list result = []
FastMountedPath mount
for mount in self._mounts:
result.append({
"path": mount._path,
"name": mount.name,
"strip_path": mount.strip_path,
})
return result
cpdef bint path_matches_prefix(str request_path, str mount_path):
cdef:
Py_ssize_t mount_len = len(mount_path)
Py_ssize_t req_len = len(request_path)
if mount_len == 0 or mount_path == "/":
return 1
if req_len < mount_len:
return 0
if req_len == mount_len:
return 1 if request_path == mount_path else 0
if request_path[mount_len] == '/':
return 1 if request_path[:mount_len] == mount_path else 0
return 0
cpdef str strip_path_prefix(str original_path, str mount_path):
cdef:
Py_ssize_t mount_len = len(mount_path)
str result
if mount_len == 0 or mount_path == "/":
return original_path
result = original_path[mount_len:]
if not result:
return "/"
return result
cpdef tuple match_and_modify_path(str request_path, str mount_path, bint strip_path=True):
cdef:
Py_ssize_t mount_len = len(mount_path)
Py_ssize_t req_len = len(request_path)
bint is_root = 1 if (mount_len == 0 or mount_path == "/") else 0
str modified
if is_root:
return (True, request_path if strip_path else request_path)
if req_len < mount_len:
return (False, None)
if req_len == mount_len:
if request_path == mount_path:
return (True, "/" if strip_path else request_path)
return (False, None)
if request_path[mount_len] == '/' and request_path[:mount_len] == mount_path:
if strip_path:
modified = request_path[mount_len:]
return (True, modified if modified else "/")
return (True, request_path)
return (False, None)

168
pyserve/_path_matcher_py.py Normal file
View File

@ -0,0 +1,168 @@
"""
Pure Python fallback for _path_matcher when Cython is not available.
This module provides the same interface as the Cython _path_matcher module,
allowing the application to run without compilation.
"""
from typing import Any, Dict, List, Optional, Tuple
class FastMountedPath:
__slots__ = ("_path", "_path_with_slash", "_path_len", "_is_root", "name", "strip_path")
def __init__(self, path: str, name: str = "", strip_path: bool = True):
if path.endswith("/") and len(path) > 1:
path = path[:-1]
self._path = path
self._path_len = len(path)
self._is_root = path == "" or path == "/"
self._path_with_slash = path + "/" if not self._is_root else "/"
self.name = name or path
self.strip_path = strip_path
@property
def path(self) -> str:
return self._path
def matches(self, request_path: str) -> bool:
if self._is_root:
return True
req_len = len(request_path)
if req_len < self._path_len:
return False
if req_len == self._path_len:
return request_path == self._path
if request_path[self._path_len] == "/":
return request_path[: self._path_len] == self._path
return False
def get_modified_path(self, original_path: str) -> str:
if not self.strip_path:
return original_path
if self._is_root:
return original_path
new_path = original_path[self._path_len :]
if not new_path:
return "/"
return new_path
def __repr__(self) -> str:
return f"FastMountedPath(path={self._path!r}, name={self.name!r})"
class FastMountManager:
__slots__ = ("_mounts", "_mount_count")
def __init__(self) -> None:
self._mounts: List[FastMountedPath] = []
self._mount_count: int = 0
def add_mount(self, mount: FastMountedPath) -> None:
self._mounts.append(mount)
self._mounts.sort(key=lambda m: len(m.path), reverse=True)
self._mount_count = len(self._mounts)
def get_mount(self, request_path: str) -> Optional[FastMountedPath]:
for mount in self._mounts:
if mount.matches(request_path):
return mount
return None
def remove_mount(self, path: str) -> bool:
if path.endswith("/") and len(path) > 1:
path = path[:-1]
for i, mount in enumerate(self._mounts):
if mount._path == path:
del self._mounts[i]
self._mount_count -= 1
return True
return False
@property
def mounts(self) -> List[FastMountedPath]:
return self._mounts.copy()
@property
def mount_count(self) -> int:
return self._mount_count
def list_mounts(self) -> List[Dict[str, Any]]:
return [
{
"path": mount._path,
"name": mount.name,
"strip_path": mount.strip_path,
}
for mount in self._mounts
]
def path_matches_prefix(request_path: str, mount_path: str) -> bool:
mount_len = len(mount_path)
req_len = len(request_path)
if mount_len == 0 or mount_path == "/":
return True
if req_len < mount_len:
return False
if req_len == mount_len:
return request_path == mount_path
if request_path[mount_len] == "/":
return request_path[:mount_len] == mount_path
return False
def strip_path_prefix(original_path: str, mount_path: str) -> str:
mount_len = len(mount_path)
if mount_len == 0 or mount_path == "/":
return original_path
result = original_path[mount_len:]
if not result:
return "/"
return result
def match_and_modify_path(request_path: str, mount_path: str, strip_path: bool = True) -> Tuple[bool, Optional[str]]:
mount_len = len(mount_path)
req_len = len(request_path)
is_root = mount_len == 0 or mount_path == "/"
if is_root:
return (True, request_path)
if req_len < mount_len:
return (False, None)
if req_len == mount_len:
if request_path == mount_path:
return (True, "/" if strip_path else request_path)
return (False, None)
if request_path[mount_len] == "/" and request_path[:mount_len] == mount_path:
if strip_path:
modified = request_path[mount_len:]
return (True, modified if modified else "/")
return (True, request_path)
return (False, None)

View File

@ -8,7 +8,8 @@ This module provides functionality to mount external ASGI/WSGI applications
import importlib
import sys
from pathlib import Path
from typing import Dict, Any, Optional, Callable, cast
from typing import Any, Callable, Dict, Optional, cast
from starlette.types import ASGIApp, Receive, Scope, Send
from .logging_utils import get_logger
@ -74,20 +75,17 @@ class ASGIAppLoader:
def _wrap_wsgi(self, wsgi_app: Callable) -> ASGIApp:
try:
from a2wsgi import WSGIMiddleware
return cast(ASGIApp, WSGIMiddleware(wsgi_app))
except ImportError:
logger.warning("a2wsgi not installed, trying asgiref")
try:
from asgiref.wsgi import WsgiToAsgi
return cast(ASGIApp, WsgiToAsgi(wsgi_app))
except ImportError:
logger.error(
"Neither a2wsgi nor asgiref installed. "
"Install with: pip install a2wsgi or pip install asgiref"
)
raise ImportError(
"WSGI adapter not available. Install a2wsgi or asgiref."
)
logger.error("Neither a2wsgi nor asgiref installed. " "Install with: pip install a2wsgi or pip install asgiref")
raise ImportError("WSGI adapter not available. Install a2wsgi or asgiref.")
def get_app(self, app_path: str) -> Optional[ASGIApp]:
return self._apps.get(app_path)
@ -132,7 +130,7 @@ class MountedApp:
if self.path == "":
return original_path
new_path = original_path[len(self.path):]
new_path = original_path[len(self.path) :]
return new_path if new_path else "/"
@ -215,10 +213,7 @@ class ASGIMountManager:
modified_scope["path"] = mount.get_modified_path(path)
modified_scope["root_path"] = scope.get("root_path", "") + mount.path
logger.debug(
f"Routing request to mounted app '{mount.name}': "
f"{path} -> {modified_scope['path']}"
)
logger.debug(f"Routing request to mounted app '{mount.name}': " f"{path} -> {modified_scope['path']}")
try:
await mount.app(modified_scope, receive, send)
@ -288,7 +283,8 @@ def create_django_app(
os.environ.setdefault("DJANGO_SETTINGS_MODULE", settings_module)
try:
from django.core.asgi import get_asgi_application # type: ignore[import-untyped]
from django.core.asgi import get_asgi_application
return cast(ASGIApp, get_asgi_application())
except ImportError as e:
logger.error(f"Failed to load Django application: {e}")

View File

@ -1,8 +1,8 @@
import sys
import argparse
import sys
from pathlib import Path
from . import PyServeServer, Config, __version__
from . import Config, PyServeServer, __version__
def main() -> None:
@ -10,30 +10,11 @@ def main() -> None:
description="PyServe - HTTP web server",
prog="pyserve",
)
parser.add_argument(
"-c", "--config",
default="config.yaml",
help="Path to configuration file (default: config.yaml)"
)
parser.add_argument(
"--host",
help="Host to bind the server to"
)
parser.add_argument(
"--port",
type=int,
help="Port to bind the server to"
)
parser.add_argument(
"--debug",
action="store_true",
help="Enable debug mode"
)
parser.add_argument(
"--version",
action="version",
version=f"%(prog)s {__version__}"
)
parser.add_argument("-c", "--config", default="config.yaml", help="Path to configuration file (default: config.yaml)")
parser.add_argument("--host", help="Host to bind the server to")
parser.add_argument("--port", type=int, help="Port to bind the server to")
parser.add_argument("--debug", action="store_true", help="Enable debug mode")
parser.add_argument("--version", action="version", version=f"%(prog)s {__version__}")
args = parser.parse_args()

View File

@ -1,8 +1,10 @@
import yaml
import os
from typing import Dict, Any, List, cast
from dataclasses import dataclass, field
import logging
import os
from dataclasses import dataclass, field
from typing import Any, Dict, List, cast
import yaml
from .logging_utils import setup_logging
@ -84,7 +86,7 @@ class Config:
@classmethod
def from_yaml(cls, file_path: str) -> "Config":
try:
with open(file_path, 'r', encoding='utf-8') as f:
with open(file_path, "r", encoding="utf-8") as f:
data = yaml.safe_load(f)
return cls._from_dict(data)
@ -99,133 +101,117 @@ class Config:
def _from_dict(cls, data: Dict[str, Any]) -> "Config":
config = cls()
if 'http' in data:
http_data = data['http']
if "http" in data:
http_data = data["http"]
config.http = HttpConfig(
static_dir=http_data.get('static_dir', config.http.static_dir),
templates_dir=http_data.get('templates_dir', config.http.templates_dir)
static_dir=http_data.get("static_dir", config.http.static_dir),
templates_dir=http_data.get("templates_dir", config.http.templates_dir),
)
if 'server' in data:
server_data = data['server']
if "server" in data:
server_data = data["server"]
config.server = ServerConfig(
host=server_data.get('host', config.server.host),
port=server_data.get('port', config.server.port),
backlog=server_data.get('backlog', config.server.backlog),
default_root=server_data.get('default_root', config.server.default_root),
proxy_timeout=server_data.get('proxy_timeout', config.server.proxy_timeout),
redirect_instructions=server_data.get('redirect_instructions', {})
host=server_data.get("host", config.server.host),
port=server_data.get("port", config.server.port),
backlog=server_data.get("backlog", config.server.backlog),
default_root=server_data.get("default_root", config.server.default_root),
proxy_timeout=server_data.get("proxy_timeout", config.server.proxy_timeout),
redirect_instructions=server_data.get("redirect_instructions", {}),
)
if 'ssl' in data:
ssl_data = data['ssl']
if "ssl" in data:
ssl_data = data["ssl"]
config.ssl = SSLConfig(
enabled=ssl_data.get('enabled', config.ssl.enabled),
cert_file=ssl_data.get('cert_file', config.ssl.cert_file),
key_file=ssl_data.get('key_file', config.ssl.key_file)
enabled=ssl_data.get("enabled", config.ssl.enabled),
cert_file=ssl_data.get("cert_file", config.ssl.cert_file),
key_file=ssl_data.get("key_file", config.ssl.key_file),
)
if 'logging' in data:
log_data = data['logging']
format_data = log_data.get('format', {})
if "logging" in data:
log_data = data["logging"]
format_data = log_data.get("format", {})
global_format = LogFormatConfig(
type=format_data.get('type', 'standard'),
use_colors=format_data.get('use_colors', True),
show_module=format_data.get('show_module', True),
timestamp_format=format_data.get('timestamp_format', '%Y-%m-%d %H:%M:%S')
type=format_data.get("type", "standard"),
use_colors=format_data.get("use_colors", True),
show_module=format_data.get("show_module", True),
timestamp_format=format_data.get("timestamp_format", "%Y-%m-%d %H:%M:%S"),
)
console_data = log_data.get('console', {})
console_format_data = console_data.get('format', {})
console_data = log_data.get("console", {})
console_format_data = console_data.get("format", {})
console_format = LogFormatConfig(
type=console_format_data.get('type', global_format.type),
use_colors=console_format_data.get('use_colors', global_format.use_colors),
show_module=console_format_data.get('show_module', global_format.show_module),
timestamp_format=console_format_data.get('timestamp_format', global_format.timestamp_format)
)
console_config = LogHandlerConfig(
level=console_data.get('level', log_data.get('level', 'INFO')),
format=console_format
type=console_format_data.get("type", global_format.type),
use_colors=console_format_data.get("use_colors", global_format.use_colors),
show_module=console_format_data.get("show_module", global_format.show_module),
timestamp_format=console_format_data.get("timestamp_format", global_format.timestamp_format),
)
console_config = LogHandlerConfig(level=console_data.get("level", log_data.get("level", "INFO")), format=console_format)
files_config = []
if 'log_file' in log_data:
if "log_file" in log_data:
default_file_format = LogFormatConfig(
type=global_format.type,
use_colors=False,
show_module=global_format.show_module,
timestamp_format=global_format.timestamp_format
type=global_format.type, use_colors=False, show_module=global_format.show_module, timestamp_format=global_format.timestamp_format
)
default_file = LogFileConfig(
path=log_data['log_file'],
level=log_data.get('level', 'INFO'),
path=log_data["log_file"],
level=log_data.get("level", "INFO"),
format=default_file_format,
loggers=[], # Empty list means including all loggers
max_bytes=10 * 1024 * 1024,
backup_count=5
backup_count=5,
)
files_config.append(default_file)
if 'files' in log_data:
for file_data in log_data['files']:
file_format_data = file_data.get('format', {})
if "files" in log_data:
for file_data in log_data["files"]:
file_format_data = file_data.get("format", {})
file_format = LogFormatConfig(
type=file_format_data.get('type', global_format.type),
use_colors=file_format_data.get('use_colors', False),
show_module=file_format_data.get('show_module', global_format.show_module),
timestamp_format=file_format_data.get('timestamp_format', global_format.timestamp_format)
type=file_format_data.get("type", global_format.type),
use_colors=file_format_data.get("use_colors", False),
show_module=file_format_data.get("show_module", global_format.show_module),
timestamp_format=file_format_data.get("timestamp_format", global_format.timestamp_format),
)
file_config = LogFileConfig(
path=file_data.get('path', './logs/pyserve.log'),
level=file_data.get('level', log_data.get('level', 'INFO')),
path=file_data.get("path", "./logs/pyserve.log"),
level=file_data.get("level", log_data.get("level", "INFO")),
format=file_format,
loggers=file_data.get('loggers', []),
max_bytes=file_data.get('max_bytes', 10 * 1024 * 1024),
backup_count=file_data.get('backup_count', 5)
loggers=file_data.get("loggers", []),
max_bytes=file_data.get("max_bytes", 10 * 1024 * 1024),
backup_count=file_data.get("backup_count", 5),
)
files_config.append(file_config)
if 'show_module' in console_format_data:
print(
"\033[33mWARNING: Parameter 'show_module' in console.format in development and may work incorrectly\033[0m"
)
console_config.format.show_module = console_format_data.get('show_module')
if "show_module" in console_format_data:
print("\033[33mWARNING: Parameter 'show_module' in console.format in development and may work incorrectly\033[0m")
console_config.format.show_module = console_format_data.get("show_module")
for i, file_data in enumerate(log_data.get('files', [])):
if 'format' in file_data and 'show_module' in file_data['format']:
print(
f"\033[33mWARNING: Parameter 'show_module' in files[{i}].format in development and may work incorrectly\033[0m"
)
for i, file_data in enumerate(log_data.get("files", [])):
if "format" in file_data and "show_module" in file_data["format"]:
print(f"\033[33mWARNING: Parameter 'show_module' in files[{i}].format in development and may work incorrectly\033[0m")
if not files_config:
default_file_format = LogFormatConfig(
type=global_format.type,
use_colors=False,
show_module=global_format.show_module,
timestamp_format=global_format.timestamp_format
type=global_format.type, use_colors=False, show_module=global_format.show_module, timestamp_format=global_format.timestamp_format
)
default_file = LogFileConfig(
path='./logs/pyserve.log',
level=log_data.get('level', 'INFO'),
path="./logs/pyserve.log",
level=log_data.get("level", "INFO"),
format=default_file_format,
loggers=[],
max_bytes=10 * 1024 * 1024,
backup_count=5
backup_count=5,
)
files_config.append(default_file)
config.logging = LoggingConfig(
level=log_data.get('level', 'INFO'),
console_output=log_data.get('console_output', True),
level=log_data.get("level", "INFO"),
console_output=log_data.get("console_output", True),
format=global_format,
console=console_config,
files=files_config
files=files_config,
)
if 'extensions' in data:
for ext_data in data['extensions']:
extension = ExtensionConfig(
type=ext_data.get('type', ''),
config=ext_data.get('config', {})
)
if "extensions" in data:
for ext_data in data["extensions"]:
extension = ExtensionConfig(type=ext_data.get("type", ""), config=ext_data.get("config", {}))
config.extensions.append(extension)
return config
@ -245,14 +231,14 @@ class Config:
if not (1 <= self.server.port <= 65535):
errors.append(f"Invalid port: {self.server.port}")
valid_log_levels = ['DEBUG', 'INFO', 'WARNING', 'ERROR', 'CRITICAL']
valid_log_levels = ["DEBUG", "INFO", "WARNING", "ERROR", "CRITICAL"]
if self.logging.level.upper() not in valid_log_levels:
errors.append(f"Invalid logging level: {self.logging.level}")
if self.logging.console.level.upper() not in valid_log_levels:
errors.append(f"Invalid console logging level: {self.logging.console.level}")
valid_format_types = ['standard', 'json']
valid_format_types = ["standard", "json"]
if self.logging.format.type not in valid_format_types:
errors.append(f"Invalid logging format type: {self.logging.format.type}")
@ -283,40 +269,40 @@ class Config:
def setup_logging(self) -> None:
config_dict = {
'level': self.logging.level,
'console_output': self.logging.console_output,
'format': {
'type': self.logging.format.type,
'use_colors': self.logging.format.use_colors,
'show_module': self.logging.format.show_module,
'timestamp_format': self.logging.format.timestamp_format
"level": self.logging.level,
"console_output": self.logging.console_output,
"format": {
"type": self.logging.format.type,
"use_colors": self.logging.format.use_colors,
"show_module": self.logging.format.show_module,
"timestamp_format": self.logging.format.timestamp_format,
},
'console': {
'level': self.logging.console.level,
'format': {
'type': self.logging.console.format.type,
'use_colors': self.logging.console.format.use_colors,
'show_module': self.logging.console.format.show_module,
'timestamp_format': self.logging.console.format.timestamp_format
}
"console": {
"level": self.logging.console.level,
"format": {
"type": self.logging.console.format.type,
"use_colors": self.logging.console.format.use_colors,
"show_module": self.logging.console.format.show_module,
"timestamp_format": self.logging.console.format.timestamp_format,
},
},
'files': []
"files": [],
}
for file_config in self.logging.files:
file_dict = {
'path': file_config.path,
'level': file_config.level,
'loggers': file_config.loggers,
'max_bytes': file_config.max_bytes,
'backup_count': file_config.backup_count,
'format': {
'type': file_config.format.type,
'use_colors': file_config.format.use_colors,
'show_module': file_config.format.show_module,
'timestamp_format': file_config.format.timestamp_format
}
"path": file_config.path,
"level": file_config.level,
"loggers": file_config.loggers,
"max_bytes": file_config.max_bytes,
"backup_count": file_config.backup_count,
"format": {
"type": file_config.format.type,
"use_colors": file_config.format.use_colors,
"show_module": file_config.format.show_module,
"timestamp_format": file_config.format.timestamp_format,
},
}
cast(List[Dict[str, Any]], config_dict['files']).append(file_dict)
cast(List[Dict[str, Any]], config_dict["files"]).append(file_dict)
setup_logging(config_dict)

View File

@ -1,7 +1,9 @@
from abc import ABC, abstractmethod
from typing import Dict, Any, List, Optional, Type
from typing import Any, Dict, List, Optional, Type
from starlette.requests import Request
from starlette.responses import Response
from .logging_utils import get_logger
logger = get_logger(__name__)
@ -36,6 +38,7 @@ class RoutingExtension(Extension):
default_proxy_timeout = config.get("default_proxy_timeout", 30.0)
self.router = create_router_from_config(regex_locations)
from .routing import RequestHandler
self.handler = RequestHandler(self.router, default_proxy_timeout=default_proxy_timeout)
async def process_request(self, request: Request) -> Optional[Response]:
@ -54,11 +57,9 @@ class SecurityExtension(Extension):
super().__init__(config)
self.allowed_ips = config.get("allowed_ips", [])
self.blocked_ips = config.get("blocked_ips", [])
self.security_headers = config.get("security_headers", {
"X-Content-Type-Options": "nosniff",
"X-Frame-Options": "DENY",
"X-XSS-Protection": "1; mode=block"
})
self.security_headers = config.get(
"security_headers", {"X-Content-Type-Options": "nosniff", "X-Frame-Options": "DENY", "X-XSS-Protection": "1; mode=block"}
)
async def process_request(self, request: Request) -> Optional[Response]:
client_ip = request.client.host if request.client else "unknown"
@ -66,11 +67,13 @@ class SecurityExtension(Extension):
if self.blocked_ips and client_ip in self.blocked_ips:
logger.warning(f"Blocked request from IP: {client_ip}")
from starlette.responses import PlainTextResponse
return PlainTextResponse("403 Forbidden", status_code=403)
if self.allowed_ips and client_ip not in self.allowed_ips:
logger.warning(f"Access denied for IP: {client_ip}")
from starlette.responses import PlainTextResponse
return PlainTextResponse("403 Forbidden", status_code=403)
return None
@ -108,33 +111,30 @@ class MonitoringExtension(Extension):
async def process_request(self, request: Request) -> Optional[Response]:
if self.enable_metrics:
self.request_count += 1
request.state.start_time = __import__('time').time()
request.state.start_time = __import__("time").time()
return None
async def process_response(self, request: Request, response: Response) -> Response:
if self.enable_metrics and hasattr(request.state, 'start_time'):
response_time = __import__('time').time() - request.state.start_time
if self.enable_metrics and hasattr(request.state, "start_time"):
response_time = __import__("time").time() - request.state.start_time
self.response_times.append(response_time)
if response.status_code >= 400:
self.error_count += 1
logger.info(f"Request: {request.method} {request.url.path} - "
f"Status: {response.status_code} - "
f"Time: {response_time:.3f}s")
logger.info(f"Request: {request.method} {request.url.path} - " f"Status: {response.status_code} - " f"Time: {response_time:.3f}s")
return response
def get_metrics(self) -> Dict[str, Any]:
avg_response_time = (sum(self.response_times) / len(self.response_times)
if self.response_times else 0)
avg_response_time = sum(self.response_times) / len(self.response_times) if self.response_times else 0
return {
"request_count": self.request_count,
"error_count": self.error_count,
"error_rate": self.error_count / max(self.request_count, 1),
"avg_response_time": avg_response_time,
"total_response_times": len(self.response_times)
"total_response_times": len(self.response_times),
}

View File

@ -3,9 +3,10 @@ import logging.handlers
import sys
import time
from pathlib import Path
from typing import Dict, Any, List, cast, Callable
from typing import Any, Callable, Dict, List, cast
import structlog
from structlog.types import FilteringBoundLogger, EventDict
from structlog.types import EventDict, FilteringBoundLogger
from . import __version__
@ -21,15 +22,15 @@ class StructlogFilter(logging.Filter):
return True
for logger_name in self.logger_names:
if record.name == logger_name or record.name.startswith(logger_name + '.'):
if record.name == logger_name or record.name.startswith(logger_name + "."):
return True
return False
class UvicornStructlogFilter(logging.Filter):
def filter(self, record: logging.LogRecord) -> bool:
if hasattr(record, 'name') and 'uvicorn.access' in record.name:
if hasattr(record, 'getMessage'):
if hasattr(record, "name") and "uvicorn.access" in record.name:
if hasattr(record, "getMessage"):
msg = record.getMessage()
if ' - "' in msg and '" ' in msg:
parts = msg.split(' - "')
@ -56,14 +57,14 @@ def add_log_level(logger: FilteringBoundLogger, method_name: str, event_dict: Ev
def add_module_info(logger: FilteringBoundLogger, method_name: str, event_dict: EventDict) -> EventDict:
if hasattr(logger, '_context') and 'logger_name' in logger._context:
logger_name = logger._context['logger_name']
if logger_name.startswith('pyserve'):
if hasattr(logger, "_context") and "logger_name" in logger._context:
logger_name = logger._context["logger_name"]
if logger_name.startswith("pyserve"):
event_dict["module"] = logger_name
elif logger_name.startswith('uvicorn'):
event_dict["module"] = 'uvicorn'
elif logger_name.startswith('starlette'):
event_dict["module"] = 'starlette'
elif logger_name.startswith("uvicorn"):
event_dict["module"] = "uvicorn"
elif logger_name.startswith("starlette"):
event_dict["module"] = "starlette"
else:
event_dict["module"] = logger_name
return event_dict
@ -74,18 +75,19 @@ def filter_module_info(show_module: bool) -> Callable[[FilteringBoundLogger, str
if not show_module and "module" in event_dict:
del event_dict["module"]
return event_dict
return processor
def colored_console_renderer(use_colors: bool = True, show_module: bool = True) -> structlog.dev.ConsoleRenderer:
return structlog.dev.ConsoleRenderer(
colors=use_colors and hasattr(sys.stderr, 'isatty') and sys.stderr.isatty(),
colors=use_colors and hasattr(sys.stderr, "isatty") and sys.stderr.isatty(),
level_styles={
"critical": "\033[35m", # Magenta
"error": "\033[31m", # Red
"warning": "\033[33m", # Yellow
"info": "\033[32m", # Green
"debug": "\033[36m", # Cyan
"error": "\033[31m", # Red
"warning": "\033[33m", # Yellow
"info": "\033[32m", # Green
"debug": "\033[36m", # Cyan
},
pad_event=25,
)
@ -113,43 +115,35 @@ class PyServeLogManager:
if self.configured:
return
if 'format' not in config and 'console' not in config and 'files' not in config:
level = config.get('level', 'INFO').upper()
console_output = config.get('console_output', True)
log_file = config.get('log_file', './logs/pyserve.log')
if "format" not in config and "console" not in config and "files" not in config:
level = config.get("level", "INFO").upper()
console_output = config.get("console_output", True)
log_file = config.get("log_file", "./logs/pyserve.log")
config = {
'level': level,
'console_output': console_output,
'format': {
'type': 'standard',
'use_colors': True,
'show_module': True,
'timestamp_format': '%Y-%m-%d %H:%M:%S'
},
'files': [{
'path': log_file,
'level': level,
'loggers': [],
'max_bytes': 10 * 1024 * 1024,
'backup_count': 5,
'format': {
'type': 'standard',
'use_colors': False,
'show_module': True,
'timestamp_format': '%Y-%m-%d %H:%M:%S'
"level": level,
"console_output": console_output,
"format": {"type": "standard", "use_colors": True, "show_module": True, "timestamp_format": "%Y-%m-%d %H:%M:%S"},
"files": [
{
"path": log_file,
"level": level,
"loggers": [],
"max_bytes": 10 * 1024 * 1024,
"backup_count": 5,
"format": {"type": "standard", "use_colors": False, "show_module": True, "timestamp_format": "%Y-%m-%d %H:%M:%S"},
}
}]
],
}
main_level = config.get('level', 'INFO').upper()
console_output = config.get('console_output', True)
main_level = config.get("level", "INFO").upper()
console_output = config.get("console_output", True)
global_format = config.get('format', {})
console_config = config.get('console', {})
files_config = config.get('files', [])
global_format = config.get("format", {})
console_config = config.get("console", {})
files_config = config.get("files", [])
console_format = {**global_format, **console_config.get('format', {})}
console_level = console_config.get('level', main_level)
console_format = {**global_format, **console_config.get("format", {})}
console_level = console_config.get("level", main_level)
self._save_original_handlers()
self._clear_all_handlers()
@ -159,38 +153,33 @@ class PyServeLogManager:
console_output=console_output,
console_format=console_format,
console_level=console_level,
files_config=files_config
files_config=files_config,
)
self._configure_stdlib_loggers(main_level)
logger = self.get_logger('pyserve')
logger = self.get_logger("pyserve")
logger.info(
"PyServe logger initialized",
version=__version__,
level=main_level,
console_output=console_output,
console_format=console_format.get('type', 'standard')
console_format=console_format.get("type", "standard"),
)
for i, file_config in enumerate(files_config):
logger.info(
"File logging configured",
file_index=i,
path=file_config.get('path'),
level=file_config.get('level', main_level),
format_type=file_config.get('format', {}).get('type', 'standard')
path=file_config.get("path"),
level=file_config.get("level", main_level),
format_type=file_config.get("format", {}).get("type", "standard"),
)
self.configured = True
def _configure_structlog(
self,
main_level: str,
console_output: bool,
console_format: Dict[str, Any],
console_level: str,
files_config: List[Dict[str, Any]]
self, main_level: str, console_output: bool, console_format: Dict[str, Any], console_level: str, files_config: List[Dict[str, Any]]
) -> None:
shared_processors = [
structlog.stdlib.filter_by_level,
@ -202,57 +191,46 @@ class PyServeLogManager:
]
if console_output:
console_show_module = console_format.get('show_module', True)
console_show_module = console_format.get("show_module", True)
console_processors = shared_processors.copy()
console_processors.append(filter_module_info(console_show_module))
if console_format.get('type') == 'json':
if console_format.get("type") == "json":
console_processors.append(json_renderer())
else:
console_processors.append(
colored_console_renderer(
console_format.get('use_colors', True),
console_show_module
)
)
console_processors.append(colored_console_renderer(console_format.get("use_colors", True), console_show_module))
console_handler = logging.StreamHandler(sys.stdout)
console_handler.setLevel(getattr(logging, console_level))
console_handler.addFilter(UvicornStructlogFilter())
console_formatter = structlog.stdlib.ProcessorFormatter(
processor=colored_console_renderer(
console_format.get('use_colors', True),
console_show_module
)
if console_format.get('type') != 'json'
else json_renderer(),
processor=(
colored_console_renderer(console_format.get("use_colors", True), console_show_module)
if console_format.get("type") != "json"
else json_renderer()
),
)
console_handler.setFormatter(console_formatter)
root_logger = logging.getLogger()
root_logger.setLevel(logging.DEBUG)
root_logger.addHandler(console_handler)
self.handlers['console'] = console_handler
self.handlers["console"] = console_handler
for i, file_config in enumerate(files_config):
file_path = file_config.get('path', './logs/pyserve.log')
file_level = file_config.get('level', main_level)
file_loggers = file_config.get('loggers', [])
max_bytes = file_config.get('max_bytes', 10 * 1024 * 1024)
backup_count = file_config.get('backup_count', 5)
file_format = file_config.get('format', {})
file_show_module = file_format.get('show_module', True)
file_path = file_config.get("path", "./logs/pyserve.log")
file_level = file_config.get("level", main_level)
file_loggers = file_config.get("loggers", [])
max_bytes = file_config.get("max_bytes", 10 * 1024 * 1024)
backup_count = file_config.get("backup_count", 5)
file_format = file_config.get("format", {})
file_show_module = file_format.get("show_module", True)
self._ensure_log_directory(file_path)
file_handler = logging.handlers.RotatingFileHandler(
file_path,
maxBytes=max_bytes,
backupCount=backup_count,
encoding='utf-8'
)
file_handler = logging.handlers.RotatingFileHandler(file_path, maxBytes=max_bytes, backupCount=backup_count, encoding="utf-8")
file_handler.setLevel(getattr(logging, file_level))
if file_loggers:
@ -262,15 +240,13 @@ class PyServeLogManager:
file_processors.append(filter_module_info(file_show_module))
file_formatter = structlog.stdlib.ProcessorFormatter(
processor=json_renderer()
if file_format.get('type') == 'json'
else plain_console_renderer(file_show_module),
processor=json_renderer() if file_format.get("type") == "json" else plain_console_renderer(file_show_module),
)
file_handler.setFormatter(file_formatter)
root_logger = logging.getLogger()
root_logger.addHandler(file_handler)
self.handlers[f'file_{i}'] = file_handler
self.handlers[f"file_{i}"] = file_handler
base_processors = [
structlog.stdlib.filter_by_level,
@ -293,14 +269,14 @@ class PyServeLogManager:
def _configure_stdlib_loggers(self, main_level: str) -> None:
library_configs = {
'uvicorn': 'DEBUG' if main_level == 'DEBUG' else 'WARNING',
'uvicorn.access': 'DEBUG' if main_level == 'DEBUG' else 'WARNING',
'uvicorn.error': 'DEBUG' if main_level == 'DEBUG' else 'ERROR',
'uvicorn.asgi': 'DEBUG' if main_level == 'DEBUG' else 'WARNING',
'starlette': 'DEBUG' if main_level == 'DEBUG' else 'WARNING',
'asyncio': 'WARNING',
'concurrent.futures': 'WARNING',
'multiprocessing': 'WARNING',
"uvicorn": "DEBUG" if main_level == "DEBUG" else "WARNING",
"uvicorn.access": "DEBUG" if main_level == "DEBUG" else "WARNING",
"uvicorn.error": "DEBUG" if main_level == "DEBUG" else "ERROR",
"uvicorn.asgi": "DEBUG" if main_level == "DEBUG" else "WARNING",
"starlette": "DEBUG" if main_level == "DEBUG" else "WARNING",
"asyncio": "WARNING",
"concurrent.futures": "WARNING",
"multiprocessing": "WARNING",
}
for logger_name, level in library_configs.items():
@ -309,7 +285,7 @@ class PyServeLogManager:
logger.propagate = True
def _save_original_handlers(self) -> None:
logger_names = ['', 'uvicorn', 'uvicorn.access', 'uvicorn.error', 'starlette']
logger_names = ["", "uvicorn", "uvicorn.access", "uvicorn.error", "starlette"]
for name in logger_names:
logger = logging.getLogger(name)
@ -320,7 +296,7 @@ class PyServeLogManager:
for handler in root_logger.handlers[:]:
root_logger.removeHandler(handler)
logger_names = ['uvicorn', 'uvicorn.access', 'uvicorn.error', 'starlette']
logger_names = ["uvicorn", "uvicorn.access", "uvicorn.error", "starlette"]
for name in logger_names:
logger = logging.getLogger(name)
for handler in logger.handlers[:]:
@ -335,14 +311,17 @@ class PyServeLogManager:
def get_logger(self, name: str) -> structlog.stdlib.BoundLogger:
if not self._structlog_configured:
structlog.configure(
processors=cast(Any, [
structlog.stdlib.filter_by_level,
add_timestamp,
add_log_level,
structlog.processors.StackInfoRenderer(),
structlog.processors.format_exc_info,
structlog.stdlib.ProcessorFormatter.wrap_for_formatter,
]),
processors=cast(
Any,
[
structlog.stdlib.filter_by_level,
add_timestamp,
add_log_level,
structlog.processors.StackInfoRenderer(),
structlog.processors.format_exc_info,
structlog.stdlib.ProcessorFormatter.wrap_for_formatter,
],
),
context_class=dict,
logger_factory=structlog.stdlib.LoggerFactory(),
wrapper_class=structlog.stdlib.BoundLogger,
@ -370,16 +349,8 @@ class PyServeLogManager:
handler.close()
del self.handlers[name]
def create_access_log(
self,
method: str,
path: str,
status_code: int,
response_time: float,
client_ip: str,
user_agent: str = ""
) -> None:
access_logger = self.get_logger('pyserve.access')
def create_access_log(self, method: str, path: str, status_code: int, response_time: float, client_ip: str, user_agent: str = "") -> None:
access_logger = self.get_logger("pyserve.access")
access_logger.info(
"HTTP access",
method=method,
@ -388,7 +359,7 @@ class PyServeLogManager:
response_time_ms=round(response_time * 1000, 2),
client_ip=client_ip,
user_agent=user_agent,
timestamp_format="access"
timestamp_format="access",
)
def shutdown(self) -> None:
@ -416,14 +387,7 @@ def get_logger(name: str) -> structlog.stdlib.BoundLogger:
return log_manager.get_logger(name)
def create_access_log(
method: str,
path: str,
status_code: int,
response_time: float,
client_ip: str,
user_agent: str = ""
) -> None:
def create_access_log(method: str, path: str, status_code: int, response_time: float, client_ip: str, user_agent: str = "") -> None:
log_manager.create_access_log(method, path, status_code, response_time, client_ip, user_agent)

33
pyserve/path_matcher.py Normal file
View File

@ -0,0 +1,33 @@
"""
Path matcher module - uses Cython implementation if available, falls back to pure Python.
"""
try:
from pyserve._path_matcher import (
FastMountedPath,
FastMountManager,
match_and_modify_path,
path_matches_prefix,
strip_path_prefix,
)
CYTHON_AVAILABLE = True
except ImportError:
from pyserve._path_matcher_py import (
FastMountedPath,
FastMountManager,
match_and_modify_path,
path_matches_prefix,
strip_path_prefix,
)
CYTHON_AVAILABLE = False
__all__ = [
"FastMountedPath",
"FastMountManager",
"path_matches_prefix",
"strip_path_prefix",
"match_and_modify_path",
"CYTHON_AVAILABLE",
]

View File

@ -1,11 +1,13 @@
import re
import mimetypes
import re
from pathlib import Path
from typing import Dict, Any, Optional, Pattern
from typing import Any, Dict, Optional, Pattern
from urllib.parse import urlparse
import httpx
from starlette.requests import Request
from starlette.responses import Response, FileResponse, PlainTextResponse
from starlette.responses import FileResponse, PlainTextResponse, Response
from .logging_utils import get_logger
logger = get_logger(__name__)
@ -100,8 +102,7 @@ class RequestHandler:
text = ""
content_type = config.get("content_type", "text/plain")
return PlainTextResponse(text, status_code=status_code,
media_type=content_type)
return PlainTextResponse(text, status_code=status_code, media_type=content_type)
if "proxy_pass" in config:
return await self._handle_proxy(request, config, route_match.params)
@ -171,8 +172,7 @@ class RequestHandler:
return PlainTextResponse("404 Not Found", status_code=404)
async def _handle_proxy(self, request: Request, config: Dict[str, Any],
params: Dict[str, str]) -> Response:
async def _handle_proxy(self, request: Request, config: Dict[str, Any], params: Dict[str, str]) -> Response:
proxy_url = config["proxy_pass"]
for key, value in params.items():
@ -197,9 +197,15 @@ class RequestHandler:
proxy_headers = dict(request.headers)
hop_by_hop_headers = [
"connection", "keep-alive", "proxy-authenticate",
"proxy-authorization", "te", "trailers", "transfer-encoding",
"upgrade", "host"
"connection",
"keep-alive",
"proxy-authenticate",
"proxy-authorization",
"te",
"trailers",
"transfer-encoding",
"upgrade",
"host",
]
for header in hop_by_hop_headers:
proxy_headers.pop(header, None)

View File

@ -1,18 +1,19 @@
import ssl
import uvicorn
import time
from pathlib import Path
from typing import Any, Dict, Optional
import uvicorn
from starlette.applications import Starlette
from starlette.requests import Request
from starlette.responses import Response, PlainTextResponse
from starlette.responses import PlainTextResponse, Response
from starlette.routing import Route
from starlette.types import ASGIApp, Receive, Scope, Send
from pathlib import Path
from typing import Optional, Dict, Any
from .config import Config
from .extensions import ExtensionManager, ASGIExtension
from .logging_utils import get_logger
from . import __version__
from .config import Config
from .extensions import ASGIExtension, ExtensionManager
from .logging_utils import get_logger
logger = get_logger(__name__)
@ -21,7 +22,7 @@ class PyServeMiddleware:
def __init__(self, app: ASGIApp, extension_manager: ExtensionManager):
self.app = app
self.extension_manager = extension_manager
self.access_logger = get_logger('pyserve.access')
self.access_logger = get_logger("pyserve.access")
async def __call__(self, scope: Scope, receive: Receive, send: Send) -> None:
if scope["type"] != "http":
@ -48,14 +49,7 @@ class PyServeMiddleware:
await response(scope, receive, send)
async def _try_asgi_mount(
self,
scope: Scope,
receive: Receive,
send: Send,
request: Request,
start_time: float
) -> bool:
async def _try_asgi_mount(self, scope: Scope, receive: Receive, send: Send, request: Request, start_time: float) -> bool:
for extension in self.extension_manager.extensions:
if isinstance(extension, ASGIExtension):
mount = extension.get_asgi_handler(request)
@ -65,10 +59,7 @@ class PyServeMiddleware:
modified_scope["path"] = mount.get_modified_path(request.url.path)
modified_scope["root_path"] = scope.get("root_path", "") + mount.path
logger.debug(
f"Routing to ASGI mount '{mount.name}': "
f"{request.url.path} -> {modified_scope['path']}"
)
logger.debug(f"Routing to ASGI mount '{mount.name}': " f"{request.url.path} -> {modified_scope['path']}")
try:
response_started = False
@ -92,15 +83,12 @@ class PyServeMiddleware:
mount=mount.name,
status_code=status_code,
process_time_ms=process_time,
user_agent=request.headers.get("user-agent", "")
user_agent=request.headers.get("user-agent", ""),
)
return True
except Exception as e:
logger.error(f"Error in ASGI mount '{mount.name}': {e}")
error_response = PlainTextResponse(
"500 Internal Server Error",
status_code=500
)
error_response = PlainTextResponse("500 Internal Server Error", status_code=500)
await error_response(scope, receive, send)
return True
return False
@ -122,7 +110,7 @@ class PyServeMiddleware:
path=path,
status_code=status_code,
process_time_ms=process_time,
user_agent=request.headers.get("user-agent", "")
user_agent=request.headers.get("user-agent", ""),
)
@ -145,27 +133,13 @@ class PyServeServer:
if ext_config.type == "routing":
config.setdefault("default_proxy_timeout", self.config.server.proxy_timeout)
self.extension_manager.load_extension(
ext_config.type,
config
)
self.extension_manager.load_extension(ext_config.type, config)
def _create_app(self) -> None:
routes = [
Route("/health", self._health_check, methods=["GET"]),
Route("/metrics", self._metrics, methods=["GET"]),
Route(
"/{path:path}",
self._catch_all,
methods=[
"GET",
"POST",
"PUT",
"DELETE",
"PATCH",
"OPTIONS"
]
),
Route("/{path:path}", self._catch_all, methods=["GET", "POST", "PUT", "DELETE", "PATCH", "OPTIONS"]),
]
self.app = Starlette(routes=routes)
@ -178,19 +152,16 @@ class PyServeServer:
metrics = {}
for extension in self.extension_manager.extensions:
if hasattr(extension, 'get_metrics'):
if hasattr(extension, "get_metrics"):
try:
ext_metrics = getattr(extension, 'get_metrics')()
ext_metrics = getattr(extension, "get_metrics")()
metrics.update(ext_metrics)
except Exception as e:
logger.error("Error getting metrics from extension",
extension=type(extension).__name__, error=str(e))
logger.error("Error getting metrics from extension", extension=type(extension).__name__, error=str(e))
import json
return Response(
json.dumps(metrics, ensure_ascii=False, indent=2),
media_type="application/json"
)
return Response(json.dumps(metrics, ensure_ascii=False, indent=2), media_type="application/json")
async def _catch_all(self, request: Request) -> Response:
return PlainTextResponse("404 Not Found", status_code=404)
@ -209,10 +180,7 @@ class PyServeServer:
try:
context = ssl.create_default_context(ssl.Purpose.CLIENT_AUTH)
context.load_cert_chain(
self.config.ssl.cert_file,
self.config.ssl.key_file
)
context.load_cert_chain(self.config.ssl.cert_file, self.config.ssl.key_file)
logger.info("SSL context created successfully")
return context
except Exception as e:
@ -237,20 +205,17 @@ class PyServeServer:
}
if ssl_context:
uvicorn_config.update({
"ssl_keyfile": self.config.ssl.key_file,
"ssl_certfile": self.config.ssl.cert_file,
})
uvicorn_config.update(
{
"ssl_keyfile": self.config.ssl.key_file,
"ssl_certfile": self.config.ssl.cert_file,
}
)
protocol = "https"
else:
protocol = "http"
logger.info(
"Starting PyServe server",
protocol=protocol,
host=self.config.server.host,
port=self.config.server.port
)
logger.info("Starting PyServe server", protocol=protocol, host=self.config.server.host, port=self.config.server.port)
try:
assert self.app is not None, "App not initialized"
@ -306,6 +271,7 @@ class PyServeServer:
self.extension_manager.cleanup()
from .logging_utils import shutdown_logging
shutdown_logging()
logger.info("Server stopped")
@ -317,13 +283,12 @@ class PyServeServer:
metrics = {"server_status": "running"}
for extension in self.extension_manager.extensions:
if hasattr(extension, 'get_metrics'):
if hasattr(extension, "get_metrics"):
try:
ext_metrics = getattr(extension, 'get_metrics')()
ext_metrics = getattr(extension, "get_metrics")()
metrics.update(ext_metrics)
except Exception as e:
logger.error("Error getting metrics from extension",
extension=type(extension).__name__, error=str(e))
logger.error("Error getting metrics from extension", extension=type(extension).__name__, error=str(e))
return metrics

72
scripts/build_cython.py Normal file
View File

@ -0,0 +1,72 @@
"""
Build script for Cython extensions.
Usage:
python scripts/build_cython.py build_ext --inplace
Or via make:
make build-cython
"""
import os
import sys
from pathlib import Path
def build_extensions():
try:
from Cython.Build import cythonize
except ImportError:
print("Cython not installed. Skipping Cython build.")
print("Install with: pip install cython")
return False
try:
from setuptools import Extension
from setuptools.dist import Distribution
from setuptools.command.build_ext import build_ext
except ImportError:
print("setuptools not installed. Skipping Cython build.")
print("Install with: pip install setuptools")
return False
extensions = [
Extension(
"pyserve._path_matcher",
sources=["pyserve/_path_matcher.pyx"],
extra_compile_args=["-O3", "-ffast-math"],
define_macros=[("NPY_NO_DEPRECATED_API", "NPY_1_7_API_VERSION")],
),
]
ext_modules = cythonize(
extensions,
compiler_directives={
"language_level": "3",
"boundscheck": False,
"wraparound": False,
"cdivision": True,
"embedsignature": True,
},
annotate=True,
)
dist = Distribution({"ext_modules": ext_modules})
dist.package_dir = {"": "."}
cmd = build_ext(dist)
cmd.ensure_finalized()
cmd.inplace = True
cmd.run()
print("\nCython extensions built successfully!")
print(" - pyserve/_path_matcher" + (".pyd" if sys.platform == "win32" else ".so"))
return True
if __name__ == "__main__":
project_root = Path(__file__).parent.parent
os.chdir(project_root)
success = build_extensions()
sys.exit(0 if success else 1)

273
tests/test_path_matcher.py Normal file
View File

@ -0,0 +1,273 @@
"""
Tests for path_matcher module.
Run with: pytest tests/test_path_matcher.py -v
"""
import pytest
from pyserve.path_matcher import (
FastMountedPath,
FastMountManager,
path_matches_prefix,
strip_path_prefix,
match_and_modify_path,
CYTHON_AVAILABLE,
)
class TestFastMountedPath:
def test_root_mount_matches_everything(self):
"""Root mount should match all paths."""
mount = FastMountedPath("")
assert mount.matches("/") is True
assert mount.matches("/api") is True
assert mount.matches("/api/users") is True
assert mount.matches("/anything/at/all") is True
def test_slash_root_mount_matches_everything(self):
"""'/' mount should match all paths."""
mount = FastMountedPath("/")
assert mount.matches("/") is True
assert mount.matches("/api") is True
assert mount.matches("/api/users") is True
def test_exact_path_match(self):
"""Exact path should match."""
mount = FastMountedPath("/api")
assert mount.matches("/api") is True
assert mount.matches("/api/") is True
assert mount.matches("/api/users") is True
def test_no_false_prefix_match(self):
"""/api should not match /api-v2."""
mount = FastMountedPath("/api")
assert mount.matches("/api-v2") is False
assert mount.matches("/api2") is False
assert mount.matches("/apiv2") is False
def test_shorter_path_no_match(self):
"""Request path shorter than mount path should not match."""
mount = FastMountedPath("/api/v1")
assert mount.matches("/api") is False
assert mount.matches("/ap") is False
assert mount.matches("/") is False
def test_trailing_slash_normalized(self):
"""Trailing slashes should be normalized."""
mount1 = FastMountedPath("/api/")
mount2 = FastMountedPath("/api")
assert mount1.path == "/api"
assert mount2.path == "/api"
assert mount1.matches("/api/users") is True
assert mount2.matches("/api/users") is True
def test_get_modified_path_strips_prefix(self):
"""Modified path should have prefix stripped."""
mount = FastMountedPath("/api")
assert mount.get_modified_path("/api") == "/"
assert mount.get_modified_path("/api/") == "/"
assert mount.get_modified_path("/api/users") == "/users"
assert mount.get_modified_path("/api/users/123") == "/users/123"
def test_get_modified_path_no_strip(self):
"""With strip_path=False, path should not be modified."""
mount = FastMountedPath("/api", strip_path=False)
assert mount.get_modified_path("/api/users") == "/api/users"
assert mount.get_modified_path("/api") == "/api"
def test_root_mount_modified_path(self):
"""Root mount should return original path."""
mount = FastMountedPath("")
assert mount.get_modified_path("/api/users") == "/api/users"
assert mount.get_modified_path("/") == "/"
def test_name_property(self):
"""Name should be set correctly."""
mount1 = FastMountedPath("/api")
mount2 = FastMountedPath("/api", name="API Mount")
assert mount1.name == "/api"
assert mount2.name == "API Mount"
def test_repr(self):
"""Repr should be informative."""
mount = FastMountedPath("/api", name="API")
assert "FastMountedPath" in repr(mount)
assert "/api" in repr(mount)
class TestFastMountManager:
def test_empty_manager(self):
"""Empty manager should return None."""
manager = FastMountManager()
assert manager.get_mount("/api") is None
assert manager.mount_count == 0
def test_add_mount(self):
"""Adding mounts should work."""
manager = FastMountManager()
mount = FastMountedPath("/api")
manager.add_mount(mount)
assert manager.mount_count == 1
assert manager.get_mount("/api/users") is mount
def test_longest_prefix_matching(self):
"""Longer prefixes should match first."""
manager = FastMountManager()
api_mount = FastMountedPath("/api", name="api")
api_v1_mount = FastMountedPath("/api/v1", name="api_v1")
api_v2_mount = FastMountedPath("/api/v2", name="api_v2")
manager.add_mount(api_mount)
manager.add_mount(api_v2_mount)
manager.add_mount(api_v1_mount)
assert manager.get_mount("/api/v1/users").name == "api_v1"
assert manager.get_mount("/api/v2/items").name == "api_v2"
assert manager.get_mount("/api/v3/other").name == "api"
assert manager.get_mount("/api").name == "api"
def test_remove_mount(self):
"""Removing mounts should work."""
manager = FastMountManager()
manager.add_mount(FastMountedPath("/api"))
manager.add_mount(FastMountedPath("/admin"))
assert manager.mount_count == 2
result = manager.remove_mount("/api")
assert result is True
assert manager.mount_count == 1
assert manager.get_mount("/api/users") is None
assert manager.get_mount("/admin/users") is not None
def test_remove_nonexistent_mount(self):
"""Removing nonexistent mount should return False."""
manager = FastMountManager()
result = manager.remove_mount("/api")
assert result is False
def test_list_mounts(self):
"""list_mounts should return mount info."""
manager = FastMountManager()
manager.add_mount(FastMountedPath("/api", name="API"))
manager.add_mount(FastMountedPath("/admin", name="Admin"))
mounts = manager.list_mounts()
assert len(mounts) == 2
assert all("path" in m and "name" in m and "strip_path" in m for m in mounts)
def test_mounts_property_returns_copy(self):
"""mounts property should return a copy."""
manager = FastMountManager()
manager.add_mount(FastMountedPath("/api"))
mounts1 = manager.mounts
mounts2 = manager.mounts
assert mounts1 is not mounts2
assert mounts1 == mounts2
class TestUtilityFunctions:
"""Tests for standalone utility functions."""
def test_path_matches_prefix_basic(self):
"""Basic prefix matching."""
assert path_matches_prefix("/api/users", "/api") is True
assert path_matches_prefix("/api", "/api") is True
assert path_matches_prefix("/api-v2", "/api") is False
assert path_matches_prefix("/ap", "/api") is False
def test_path_matches_prefix_root(self):
"""Root prefix matches everything."""
assert path_matches_prefix("/anything", "") is True
assert path_matches_prefix("/anything", "/") is True
def test_strip_path_prefix_basic(self):
"""Basic path stripping."""
assert strip_path_prefix("/api/users", "/api") == "/users"
assert strip_path_prefix("/api", "/api") == "/"
assert strip_path_prefix("/api/", "/api") == "/"
def test_strip_path_prefix_root(self):
"""Root prefix doesn't strip anything."""
assert strip_path_prefix("/api/users", "") == "/api/users"
assert strip_path_prefix("/api/users", "/") == "/api/users"
def test_match_and_modify_combined(self):
"""Combined match and modify operation."""
matches, path = match_and_modify_path("/api/users", "/api")
assert matches is True
assert path == "/users"
matches, path = match_and_modify_path("/api", "/api")
assert matches is True
assert path == "/"
matches, path = match_and_modify_path("/other", "/api")
assert matches is False
assert path is None
def test_match_and_modify_no_strip(self):
"""Combined operation with strip_path=False."""
matches, path = match_and_modify_path("/api/users", "/api", strip_path=False)
assert matches is True
assert path == "/api/users"
class TestCythonAvailability:
def test_cython_available_is_bool(self):
"""CYTHON_AVAILABLE should be a boolean."""
assert isinstance(CYTHON_AVAILABLE, bool)
def test_module_works_regardless(self):
"""Module should work whether Cython is available or not."""
mount = FastMountedPath("/test")
assert mount.matches("/test/path") is True
class TestPerformance:
def test_many_matches(self):
"""Should handle many match operations."""
mount = FastMountedPath("/api/v1/users")
for _ in range(10000):
assert mount.matches("/api/v1/users/123/posts") is True
assert mount.matches("/other/path") is False
def test_many_mounts(self):
"""Should handle many mounts."""
manager = FastMountManager()
for i in range(100):
manager.add_mount(FastMountedPath(f"/api/v{i}"))
assert manager.mount_count == 100
mount = manager.get_mount("/api/v50/users")
assert mount is not None
assert mount.path == "/api/v50"
if __name__ == "__main__":
pytest.main([__file__, "-v"])