Add benchmarking

.gitignore

@@ -1,5 +1,7 @@
 # Local cache
 .cache
+resources/test
+!resources/test/src
 
 ###
 # Rust.gitignore

Cargo.lock

@@ -1529,6 +1529,7 @@ dependencies = [
  "clearscreen",
  "colored",
  "fern",
+ "lazy_static",
  "log",
  "notify",
  "pyo3",

Cargo.toml

@@ -16,6 +16,7 @@ clap = { version = "3.2.16", features = ["derive"] }
 clearscreen = { version = "1.0.10" }
 colored = { version = "2.0.0" }
 fern = { version = "0.6.1" }
+lazy_static = { version = "1.4.0" }
 log = { version = "0.4.17" }
 notify = { version = "4.0.17" }
 pyo3 = { version = "0.16.5", features = ["extension-module", "abi3-py37"] }
@@ -25,3 +26,7 @@ rustpython-parser = { git = "https://github.com/RustPython/RustPython.git", rev
 serde = { version = "1.0.143", features = ["derive"] }
 serde_json = { version = "1.0.83" }
 walkdir = { version = "2.3.2" }
+
+[profile.release]
+lto = true
+panic = "abort"

README.md

@@ -55,3 +55,31 @@ cargo run resources/test/src
 maturin publish --skip-existing --target x86_64-apple-darwin
 maturin publish --skip-existing --target aarch64-apple-darwin
 ```
+
+## Benchmarking
+
+First, clone [CPython](https://github.com/python/cpython). It's a large and diverse Python codebase,
+which makes it a good target for benchmarking. Note that we clone v3.9, as `RustPython` doesn't yet
+support pattern matching, which was introduced in v3.10.
+
+```shell
+git clone --branch 3.9 https://github.com/python/cpython.git resources/test/cpython
+```
+
+Next, to benchmark the release build:
+
+```shell
+cargo build --release
+
+hyperfine --warmup 5 \
+  "./target/release/rust_python_linter ./resources/test/cpython/ --no-cache" \
+  "./target/release/rust_python_linter ./resources/test/cpython/"
+
+Benchmark 1: ./target/release/rust_python_linter ./resources/test/cpython/ --no-cache
+  Time (mean ± σ):     353.6 ms ±   7.6 ms    [User: 2868.8 ms, System: 171.5 ms]
+  Range (min … max):   344.4 ms … 367.3 ms    10 runs
+
+Benchmark 2: ./target/release/rust_python_linter ./resources/test/cpython/
+  Time (mean ± σ):      59.6 ms ±   2.5 ms    [User: 36.4 ms, System: 345.6 ms]
+  Range (min … max):    55.9 ms …  67.0 ms    48 runs
+```

src/fs.rs

@@ -1,10 +1,63 @@
+use anyhow::Result;
+use lazy_static::lazy_static;
+use std::collections::HashSet;
 use std::fs::File;
 use std::io::{BufRead, BufReader, Read};
+use std::ops::Deref;
 use std::path::{Path, PathBuf};
-
-use anyhow::Result;
 use walkdir::{DirEntry, WalkDir};
 
+lazy_static! {
+    // TODO(charlie): Make these configurable.
+    static ref EXCLUDES: HashSet<&'static str> = vec![
+        "resources/test/cpython/Lib/ctypes/test/test_numbers.py",
+        "resources/test/cpython/Lib/dataclasses.py",
+        "resources/test/cpython/Lib/lib2to3/tests/data/bom.py",
+        "resources/test/cpython/Lib/lib2to3/tests/data/crlf.py",
+        "resources/test/cpython/Lib/lib2to3/tests/data/different_encoding.py",
+        "resources/test/cpython/Lib/lib2to3/tests/data/false_encoding.py",
+        "resources/test/cpython/Lib/lib2to3/tests/data/py2_test_grammar.py",
+        "resources/test/cpython/Lib/sqlite3/test/factory.py",
+        "resources/test/cpython/Lib/sqlite3/test/hooks.py",
+        "resources/test/cpython/Lib/sqlite3/test/regression.py",
+        "resources/test/cpython/Lib/sqlite3/test/transactions.py",
+        "resources/test/cpython/Lib/sqlite3/test/types.py",
+        "resources/test/cpython/Lib/test/bad_coding2.py",
+        "resources/test/cpython/Lib/test/badsyntax_3131.py",
+        "resources/test/cpython/Lib/test/badsyntax_pep3120.py",
+        "resources/test/cpython/Lib/test/encoded_modules/module_iso_8859_1.py",
+        "resources/test/cpython/Lib/test/encoded_modules/module_koi8_r.py",
+        "resources/test/cpython/Lib/test/sortperf.py",
+        "resources/test/cpython/Lib/test/test_email/torture_test.py",
+        "resources/test/cpython/Lib/test/test_fstring.py",
+        "resources/test/cpython/Lib/test/test_genericpath.py",
+        "resources/test/cpython/Lib/test/test_getopt.py",
+        "resources/test/cpython/Lib/test/test_htmlparser.py",
+        "resources/test/cpython/Lib/test/test_importlib/stubs.py",
+        "resources/test/cpython/Lib/test/test_importlib/test_files.py",
+        "resources/test/cpython/Lib/test/test_importlib/test_metadata_api.py",
+        "resources/test/cpython/Lib/test/test_importlib/test_open.py",
+        "resources/test/cpython/Lib/test/test_importlib/test_util.py",
+        "resources/test/cpython/Lib/test/test_named_expressions.py",
+        "resources/test/cpython/Lib/test/test_peg_generator/__main__.py",
+        "resources/test/cpython/Lib/test/test_pipes.py",
+        "resources/test/cpython/Lib/test/test_source_encoding.py",
+        "resources/test/cpython/Lib/test/test_weakref.py",
+        "resources/test/cpython/Lib/test/test_webbrowser.py",
+        "resources/test/cpython/Lib/tkinter/__main__.py",
+        "resources/test/cpython/Lib/tkinter/test/test_tkinter/test_variables.py",
+        "resources/test/cpython/Modules/_decimal/libmpdec/literature/fnt.py",
+        "resources/test/cpython/Modules/_decimal/tests/deccheck.py",
+        "resources/test/cpython/Tools/i18n/pygettext.py",
+        "resources/test/cpython/Tools/test2to3/maintest.py",
+        "resources/test/cpython/Tools/test2to3/setup.py",
+        "resources/test/cpython/Tools/test2to3/test/test_foo.py",
+        "resources/test/cpython/Tools/test2to3/test2to3/hello.py",
+    ]
+    .into_iter()
+    .collect();
+}
+
 fn is_not_hidden(entry: &DirEntry) -> bool {
     entry
         .file_name()
@@ -20,6 +73,7 @@ pub fn iter_python_files(path: &PathBuf) -> impl Iterator<Item = DirEntry> {
         .filter_entry(is_not_hidden)
         .filter_map(|entry| entry.ok())
         .filter(|entry| entry.path().to_string_lossy().ends_with(".py"))
+        .filter(|entry| !EXCLUDES.contains(entry.path().to_string_lossy().deref()))
 }
 
 pub fn read_line(path: &Path, row: &usize) -> Result<String> {