implement preds.py as a Rust wheel with PyO3 & Maturin

.github/workflows/ci.yml

@@ -35,12 +35,25 @@ jobs:
       uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
+    - name: Set up Rust
+      uses: dtolnay/rust-toolchain@stable
     - name: Install dependencies
       run: |
         pip install --upgrade pip wheel
         pip install --editable .
-    - name: Test with bikeshed
+    - name: Build Rust extension
+      run: |
+        pip install maturin
+        maturin build --release
+        pip install --force-reinstall --find-links rust/target/wheels bikeshed_rust
+    - name: Test with bikeshed (Python mode)
+      run: bikeshed --no-update test
+      env:
+        BIKESHED_USE_RUST: '0'
+    - name: Test with bikeshed (Rust mode)
       run: bikeshed --no-update test
+      env:
+        BIKESHED_USE_RUST: '1'
 
   lint:
 

.gitignore

@@ -18,4 +18,5 @@ node_modules/
 /playwright/.cache/
 /env
 /docs/*.html
-/html-perf-test
+/rust/target/
+/html-perf-test

README.md

@@ -67,6 +67,13 @@ though most such specs have switched their source file extensions to `.bs` now.
 Using `.src.html` in most text editors will display the file with HTML source formatting,
 which isn't generally what you want.)
 
+Rust
+-----------
+
+Bikeshed includes optional Rust extensions, in an effort to port some or all of the code into Rust.
+
+To enable: `export BIKESHED_USE_RUST=1`
+
 License
 -------
 

bikeshed/h/parser/parser.py

@@ -6,7 +6,7 @@
 
 from ... import config, constants, t
 from ... import messages as m
-from . import preds
+from . import preds_wrapper as preds
 from .nodes import (
     Comment,
     Doctype,

bikeshed/h/parser/preds_wrapper.py

@@ -0,0 +1,82 @@
+from __future__ import annotations
+
+import os
+
+_USE_RUST = os.environ.get("BIKESHED_USE_RUST", "").lower() in ("1", "true")
+
+if _USE_RUST:
+    try:
+        import bikeshed_rust
+        from . import preds as _preds
+
+        isASCII = bikeshed_rust.is_ascii
+        isASCIIAlpha = bikeshed_rust.is_ascii_alpha
+        isASCIIAlphanum = bikeshed_rust.is_ascii_alphanum
+        isASCIILowerAlpha = bikeshed_rust.is_ascii_lower_alpha
+        isASCIIUpperAlpha = bikeshed_rust.is_ascii_upper_alpha
+        isAttrNameChar = bikeshed_rust.is_attr_name_char
+        isControl = bikeshed_rust.is_control
+        isDigit = bikeshed_rust.is_digit
+        isHexDigit = bikeshed_rust.is_hex_digit
+        isNoncharacter = bikeshed_rust.is_noncharacter
+        isTagnameChar = bikeshed_rust.is_tagname_char
+        isWhitespace = bikeshed_rust.is_whitespace
+
+        charRefs = _preds.charRefs
+        xmlishTagnames = _preds.xmlishTagnames
+        isXMLishTagname = _preds.isXMLishTagname
+    except ImportError:
+        from . import preds as _preds
+
+        charRefs = _preds.charRefs
+        xmlishTagnames = _preds.xmlishTagnames
+        isASCII = _preds.isASCII
+        isASCIIAlpha = _preds.isASCIIAlpha
+        isASCIIAlphanum = _preds.isASCIIAlphanum
+        isASCIILowerAlpha = _preds.isASCIILowerAlpha
+        isASCIIUpperAlpha = _preds.isASCIIUpperAlpha
+        isAttrNameChar = _preds.isAttrNameChar
+        isControl = _preds.isControl
+        isDigit = _preds.isDigit
+        isHexDigit = _preds.isHexDigit
+        isNoncharacter = _preds.isNoncharacter
+        isTagnameChar = _preds.isTagnameChar
+        isWhitespace = _preds.isWhitespace
+        isXMLishTagname = _preds.isXMLishTagname
+
+else:
+    from . import preds as _preds
+
+    charRefs = _preds.charRefs
+    xmlishTagnames = _preds.xmlishTagnames
+    isASCII = _preds.isASCII
+    isASCIIAlpha = _preds.isASCIIAlpha
+    isASCIIAlphanum = _preds.isASCIIAlphanum
+    isASCIILowerAlpha = _preds.isASCIILowerAlpha
+    isASCIIUpperAlpha = _preds.isASCIIUpperAlpha
+    isAttrNameChar = _preds.isAttrNameChar
+    isControl = _preds.isControl
+    isDigit = _preds.isDigit
+    isHexDigit = _preds.isHexDigit
+    isNoncharacter = _preds.isNoncharacter
+    isTagnameChar = _preds.isTagnameChar
+    isWhitespace = _preds.isWhitespace
+    isXMLishTagname = _preds.isXMLishTagname
+
+__all__ = [
+    "charRefs",
+    "xmlishTagnames",
+    "isASCII",
+    "isASCIIAlpha",
+    "isASCIIAlphanum",
+    "isASCIILowerAlpha",
+    "isASCIIUpperAlpha",
+    "isAttrNameChar",
+    "isControl",
+    "isDigit",
+    "isHexDigit",
+    "isNoncharacter",
+    "isTagnameChar",
+    "isWhitespace",
+    "isXMLishTagname",
+]

bikeshed/test.py

@@ -12,6 +12,19 @@
 from . import messages as m
 from .Spec import Spec
 
+
+def _getTestTitle() -> str:
+    try:
+        from .h.parser import preds_wrapper
+        # Check if we're using the Rust implementation
+        if hasattr(preds_wrapper, 'isASCII'):
+            module = getattr(preds_wrapper.isASCII, '__module__', '')
+            if 'bikeshed_rust' in module:
+                return "Running tests [R]"
+    except Exception:
+        pass
+    return "Running tests"
+
 if t.TYPE_CHECKING:
     import argparse
 
@@ -101,7 +114,7 @@ def run(
     numPassed = 0
     total = 0
     fails = []
-    pathProgress = alive_it(paths, dual_line=True, length=20)
+    pathProgress = alive_it(paths, dual_line=True, length=20, title=_getTestTitle())
     try:
         for path in pathProgress:
             testName = testNameForPath(path)
@@ -149,7 +162,7 @@ def rebase(
     if len(paths) == 0:
         m.p("No tests were found.")
         return True
-    pathProgress = alive_it(paths, dual_line=True, length=20)
+    pathProgress = alive_it(paths, dual_line=True, length=20, title=_getTestTitle())
     try:
         for path in pathProgress:
             testName = testNameForPath(path)

pyproject.toml

@@ -1,3 +1,17 @@
+[build-system]
+requires = ["maturin>=1.0,<2.0"]
+build-backend = "maturin"
+
+[project]
+name = "bikeshed"
+version = "0.1.0"
+requires-python = ">=3.9"
+
+[tool.maturin]
+manifest-path = "rust/Cargo.toml"
+module-name = "bikeshed_rust"
+python-source = "bikeshed"
+
 [tool.black]
 line-length = 120
 

rust/Cargo.toml

@@ -0,0 +1,16 @@
+[package]
+name = "bikeshed_rust"
+version = "0.1.0"
+edition = "2024"
+
+[lib]
+name = "bikeshed_rust"
+crate-type = ["cdylib"]
+
+[dependencies]
+pyo3 = { version = "0.26", features = ["extension-module"] }
+
+[profile.release]
+lto = true
+codegen-units = 1
+strip = true