pychd-pyfuzz 0.1.0__tar.gz

pychd_pyfuzz-0.1.0/.gitignore

@@ -0,0 +1,172 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+**/logging.conf
+.envrc
+
+# Local Claude Code session state
+.claude/
+
+# Codex review scratch (transient, do not commit)
+.codex
+
+# Local PNG previews (SVGs are the committed source of truth)
+assets/*.png

pychd_pyfuzz-0.1.0/PKG-INFO

@@ -0,0 +1,45 @@
+Metadata-Version: 2.4
+Name: pychd-pyfuzz
+Version: 0.1.0
+Summary: Random syntactically-valid Python source generator — produces never-published code for contamination-free decompiler benchmarking
+Author-email: 卍diohabara卍 <diohabara@users.noreply.github.com>
+Requires-Python: >=3.14
+Description-Content-Type: text/markdown
+
+# pychd-pyfuzz
+
+Random, syntactically-valid Python source generator. Built to feed
+contamination-free benchmarks for `pychd` (the Python `.pyc`
+decompiler) and any other decompiler that wants the same.
+
+Every sample is constructed directly from `ast.AST` nodes (not text
+templates), so it is — by construction — a code path that has never
+appeared anywhere on the internet and cannot be in any LLM's training
+data. Pair with `pychd-pyobf` (which strips identifiers + constants
+from compiled `.pyc`) for the strongest available contamination
+guarantee.
+
+See the main [pychd README](https://github.com/diohabara/pychd) for
+the broader story.
+
+## Install
+
+```bash
+pip install pychd-pyfuzz
+```
+
+## Use
+
+```bash
+# 100 samples targeting Python 3.14:
+pychd-pyfuzz generate --target 3.14 --count 100 --seed 0 --out /tmp/fuzz/
+```
+
+Each emitted `.py` file is accompanied by a sidecar `.tags.json` with
+the syntactic-feature tags the sample actually exercised (`match`,
+`try_star`, `walrus`, `fstring`, `type_params`, …) so downstream
+benchmarks can break recovery rates out by feature.
+
+## Status
+
+Pre-release. API and CLI are still evolving with the parent project.

pychd_pyfuzz-0.1.0/README.md

@@ -0,0 +1,37 @@
+# pychd-pyfuzz
+
+Random, syntactically-valid Python source generator. Built to feed
+contamination-free benchmarks for `pychd` (the Python `.pyc`
+decompiler) and any other decompiler that wants the same.
+
+Every sample is constructed directly from `ast.AST` nodes (not text
+templates), so it is — by construction — a code path that has never
+appeared anywhere on the internet and cannot be in any LLM's training
+data. Pair with `pychd-pyobf` (which strips identifiers + constants
+from compiled `.pyc`) for the strongest available contamination
+guarantee.
+
+See the main [pychd README](https://github.com/diohabara/pychd) for
+the broader story.
+
+## Install
+
+```bash
+pip install pychd-pyfuzz
+```
+
+## Use
+
+```bash
+# 100 samples targeting Python 3.14:
+pychd-pyfuzz generate --target 3.14 --count 100 --seed 0 --out /tmp/fuzz/
+```
+
+Each emitted `.py` file is accompanied by a sidecar `.tags.json` with
+the syntactic-feature tags the sample actually exercised (`match`,
+`try_star`, `walrus`, `fstring`, `type_params`, …) so downstream
+benchmarks can break recovery rates out by feature.
+
+## Status
+
+Pre-release. API and CLI are still evolving with the parent project.

pychd_pyfuzz-0.1.0/pychd_pyfuzz/__init__.py

@@ -0,0 +1,24 @@
+"""pychd_pyfuzz — random syntactically-valid Python source generator.
+
+Public API:
+
+* :class:`Fuzzer` — construct with a target version + seed, call
+  :meth:`Fuzzer.generate` or :meth:`Fuzzer.generate_batch`.
+* :class:`Sample` — dataclass returned by ``generate`` carrying the
+  source text, the syntactic-feature tags it exercised, the target
+  version, the seed, and the batch index.
+* :class:`TagSet` — lower-level: the mutable tag accumulator
+  builders write into. End users rarely touch this directly.
+
+The CLI entry point is :func:`pychd_pyfuzz.cli.main` (registered as
+the ``pychd-pyfuzz`` console script).
+"""
+
+from __future__ import annotations
+
+from .generator import Fuzzer, Sample
+from .tags import TagSet
+
+__version__ = "0.1.0"
+
+__all__ = ["Fuzzer", "Sample", "TagSet", "__version__"]

pychd_pyfuzz-0.1.0/pychd_pyfuzz/cli.py

@@ -0,0 +1,117 @@
+"""``pychd-pyfuzz`` command-line entry point.
+
+Subcommands:
+
+* ``pychd-pyfuzz generate`` — write N samples to a directory, one
+  ``.py`` per sample, each accompanied by a ``.tags.json`` sidecar.
+* ``pychd-pyfuzz emit`` — print a single sample to stdout (no
+  sidecar). Handy for ad-hoc inspection.
+
+The CLI is deliberately thin; library users should call
+:class:`pychd_pyfuzz.Fuzzer` directly.
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import sys
+from pathlib import Path
+
+from .generator import Fuzzer, Sample
+
+
+def _parse_version(s: str) -> tuple[int, int]:
+    parts = s.split(".")
+    if len(parts) != 2:
+        raise argparse.ArgumentTypeError(f"--target must be MAJOR.MINOR (got {s!r})")
+    try:
+        return (int(parts[0]), int(parts[1]))
+    except ValueError as e:
+        raise argparse.ArgumentTypeError(str(e)) from e
+
+
+def _write_sample(sample: Sample, out_dir: Path, stem: str) -> None:
+    out_dir.mkdir(parents=True, exist_ok=True)
+    py_path = out_dir / f"{stem}.py"
+    tags_path = out_dir / f"{stem}.tags.json"
+    py_path.write_text(sample.source)
+    tags_path.write_text(
+        json.dumps(
+            {
+                "target": f"{sample.target[0]}.{sample.target[1]}",
+                "seed": sample.seed,
+                "index": sample.index,
+                "length": sample.length,
+                "tags": sample.tags,
+            },
+            indent=2,
+        )
+    )
+
+
+def _cmd_generate(args: argparse.Namespace) -> int:
+    fuzzer = Fuzzer(
+        target=args.target,
+        seed=args.seed,
+        max_depth=args.max_depth,
+        max_top_items=args.max_top_items,
+    )
+    out_dir = Path(args.out)
+    out_dir.mkdir(parents=True, exist_ok=True)
+    written = 0
+    for sample in fuzzer.generate_batch(args.count):
+        stem = f"sample_{sample.index:04d}"
+        _write_sample(sample, out_dir, stem)
+        written += 1
+    print(
+        f"pychd-pyfuzz: wrote {written} samples (target={args.target[0]}."
+        f"{args.target[1]}, seed={args.seed}) to {out_dir}",
+        file=sys.stderr,
+    )
+    return 0
+
+
+def _cmd_emit(args: argparse.Namespace) -> int:
+    fuzzer = Fuzzer(
+        target=args.target,
+        seed=args.seed,
+        max_depth=args.max_depth,
+        max_top_items=args.max_top_items,
+    )
+    sample = fuzzer.generate(index=0)
+    sys.stdout.write(sample.source)
+    if not sample.source.endswith("\n"):
+        sys.stdout.write("\n")
+    return 0
+
+
+def main(argv: list[str] | None = None) -> int:
+    parser = argparse.ArgumentParser(prog="pychd-pyfuzz", description=__doc__)
+    sub = parser.add_subparsers(dest="cmd", required=True)
+
+    gen = sub.add_parser(
+        "generate",
+        help="write N samples to OUT_DIR as ``sample_NNNN.py`` + ``.tags.json``",
+    )
+    gen.add_argument("--target", type=_parse_version, default=(3, 14))
+    gen.add_argument("--count", type=int, default=10)
+    gen.add_argument("--seed", type=int, default=0)
+    gen.add_argument("--max-depth", type=int, default=3)
+    gen.add_argument("--max-top-items", type=int, default=6)
+    gen.add_argument("--out", type=Path, required=True)
+    gen.set_defaults(func=_cmd_generate)
+
+    emit = sub.add_parser("emit", help="print a single sample to stdout")
+    emit.add_argument("--target", type=_parse_version, default=(3, 14))
+    emit.add_argument("--seed", type=int, default=0)
+    emit.add_argument("--max-depth", type=int, default=3)
+    emit.add_argument("--max-top-items", type=int, default=6)
+    emit.set_defaults(func=_cmd_emit)
+
+    args = parser.parse_args(argv)
+    return args.func(args)
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())