Skip to main content

parsanol (Python)

Python runtime for PARG artifacts — checksummed parsanol-tree/v2 envelopes that Ruby, Rust, WebAssembly, TypeScript, and now Python all parse identically.

The engine is the real one: this package drives libparsanol (the pure-portable C ABI cdylib, the same binary the Ruby ffi tier ships) via ctypes. No second parser, no reimplementation.

pip install parsanol

Each platform wheel vendors the cdylib: Linux (manylinux_2_35, x86_64), macOS (arm64), Windows (x64). There is no sdist — the Rust sources are a submodule of the repository, not of the archive, so a source install could never load the engine. Point a locally built library at the runtime with PARSANOL_CDYLIB=/path/to/libparsanol.

from parsanol import PargArtifact

artifact = PargArtifact.from_path("demo.artifact.json")  # verifies sha256
tree = artifact.parse("iso 12345", "identifier")         # shaped tree
run = artifact.run_tests()                               # embedded tests replayed

On rejection, parse raises ParseFailed carrying the engine's failure wire (offset, ranked positions, expected set).

Releases

parsanol is one product on four runtimes: rs (crates.io + @parsanol/wasm + release binaries), ruby (rubygems), python (PyPI, this package), and typescript (npm, parsanol — the wasm-backed runtime). Every engine update releases all four together; a release that ships some runtimes and not the rest is incomplete.

Status and honest boundaries (v0.1)

Capability State
Artifact load + sha256 verification works — the canonical-form port is byte-exact against the reference compiler (all 39 corpus checksums verify)
Native parse (shaped tree, slice wire) works
Failure wire works
Embedded tests: accept / reject / capture works — all 155 corpus cases replay with zero failures
Embedded tests: example / shape works — the bindings/flattening port landed (ticket 0003)
parse_recovering not yet — needs a C-ABI export (ticket 0002)
Dynamic / runtime-state artifacts rejected upstream by contract (Ruby-tier)
Wheels on PyPI published — platform wheels with the vendored cdylib; a post-publish audit replays the corpus through the wheel installed from PyPI (ticket 0001)

Development

cargo build -p parsanol-ffi --release          # in parsanol-rs
export PARSANOL_CDYLIB=$RS/target/release/libparsanol.dylib
ln -s ../parsanol-rs/corpus corpus             # for the corpus gate
python3 -m pip install -e . pytest
python3 -m pytest tests -q

The corpus gate (tests/test_corpus.py) pins the size: 155 cases, zero skips — when it moves, this repo moves with it deliberately.

Explicit native libraries

PargArtifact.from_path(path, library_path="/absolute/path/to/library") uses an independent native context. It never changes the process environment or the default runtime, and an invalid explicit path never falls back to another library. The library loads lazily on the first native operation.

To share one engine between artifacts, create NativeContext(library_path) and pass native_context=context to their constructors or from_path. Each context binds its own parse, error, recovery and free functions; every owned result is freed through the context that created it, including read failures. context.engine_version() identifies that engine. Pass either library_path or native_context. Existing constructors and module-level functions keep their default lookup and singleton behavior.

Metadata

Release files for parsanol 0.1.28

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distributions (wheels)

Table of built distributions (wheels) for parsanol 0.1.28
File Interpreter ABI Platform Actions
parsanol-0.1.28-py3-none-win_amd64.whl Python 3 none Windows x86-64 Details for parsanol-0.1.28-py3-none-win_amd64.whl
parsanol-0.1.28-py3-none-manylinux_2_35_x86_64.whl Python 3 none Linux glibc 2.35+ x86-64 Details for parsanol-0.1.28-py3-none-manylinux_2_35_x86_64.whl
parsanol-0.1.28-py3-none-macosx_11_0_arm64.whl Python 3 none macOS 11.0+ ARM64 Details for parsanol-0.1.28-py3-none-macosx_11_0_arm64.whl

Total release size: 3.0 MB

Release files / parsanol-0.1.28-py3-none-win_amd64.whl

Download URL parsanol-0.1.28-py3-none-win_amd64.whl
Size 934.8 kB
Tags Python 3 Windows x86-64
SHA-256 checksum
How to use checksums
94190ca3ee0fa3a0eb2d4054240dfb9d1f9d0da238ee024790e66a85bc862a84
BLAKE2b-256 checksum
How to use checksums
c8d0f6bebb8343275ad4e436029a019a99e19fcfed44a9a750c26ffd2c95ba9a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 12, 2026.

Transparency log

Release files / parsanol-0.1.28-py3-none-manylinux_2_35_x86_64.whl

Download URL parsanol-0.1.28-py3-none-manylinux_2_35_x86_64.whl
Size 1.1 MB
Tags Linux glibc 2.35+ x86-64 Python 3
SHA-256 checksum
How to use checksums
e70c513f2f47107c45e0dd43e0eced34d7eed09dd20d86a94194c02c5396f28c
BLAKE2b-256 checksum
How to use checksums
1e9ad9f1f5fdf1921f0a63f3263b870b0fb6e759ee4dc2dac3838a952d7a0f78
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 12, 2026.

Transparency log

Release files / parsanol-0.1.28-py3-none-macosx_11_0_arm64.whl

Download URL parsanol-0.1.28-py3-none-macosx_11_0_arm64.whl
Size 973.5 kB
Tags Python 3 macOS 11.0+ ARM64
SHA-256 checksum
How to use checksums
6f658e4b3791a194808b23bf658cd9e6457b74d1cf5a49cc965a8ac6ffb51488
BLAKE2b-256 checksum
How to use checksums
178b6dac8c944b40fb4f63c5061ba0af667db74ec247e5dc8a0093004219be1b
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 12, 2026.

Transparency log

Security for parsanol 0.1.28

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 3 files in this release have been uploaded with provenance metadata.

Select from 3 files

Attestation for parsanol-0.1.28-py3-none-win_amd64.whl

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 12, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.28 This release

3 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page