docparity
Detects drift between a function's docstring parameters and its real signature -- the classic case of renaming or removing a parameter and forgetting to update the docstring, so the documentation quietly lies to the next reader (or to an LLM/IDE that trusts it).
Supports Google, NumPy, and Sphinx (:param:) docstring styles. Zero
runtime dependencies, pure ast-based static analysis.
Why
Docstrings drift silently:
def charge_card(amount, currency):
"""Charge a card.
Args:
amount (float): the amount to charge
card_token (str): the card token to charge # <- renamed to `currency`, doc never updated
"""
Nothing catches this today. Type checkers don't read docstrings. Linters that check docstring presence (pydocstyle, flake8-docstrings) don't check that documented parameter names still match the signature. The result is stale documentation that misleads humans, and misleads LLM coding assistants and IDEs that surface docstrings as ground truth.
Install
pip install docparity
Usage
docparity src/ # scan a directory recursively
docparity mymodule.py # scan a single file
docparity --stale-only src/ # only report renamed/removed params, not missing docs
Exit codes: 0 clean, 1 findings reported, 2 usage/syntax error.
What it checks (v0.1)
- Stale documented parameter -- the docstring documents a parameter name that no longer exists in the real signature (renamed or removed).
- Missing documentation for a real parameter -- the function already has a params section, and a real parameter isn't in it (added to the signature, forgotten in the docs).
docparity only checks functions that already document at least one parameter. It never asks you to add documentation to an undocumented function -- that's a coverage tool's job, not a drift tool's job. This keeps false positives near zero.
Library usage
from docparity import scan_file
findings = scan_file("mymodule.py")
for f in findings:
print(f)
See docs/USAGE.md for more, and DETAILS.md for design rationale and scope boundaries.
License
MIT
Metadata
Release files for docparity 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| docparity-0.1.0.tar.gz | 9.5 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| docparity-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 16.8 kB
Release files / docparity-0.1.0.tar.gz
| Download URL | docparity-0.1.0.tar.gz |
|---|---|
| Size | 9.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
40b8f35c1094a2e5a05088733b62ab5737cceecd483945c26bb5dd83976ec6c4
|
|
BLAKE2b-256 checksum How to use checksums |
21175fe334ce331e132245a6fb6d15eadfab0811f2374bf2874e630d5096545a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.10.12
|
Release files / docparity-0.1.0-py3-none-any.whl
| Download URL | docparity-0.1.0-py3-none-any.whl |
|---|---|
| Size | 7.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
4401e4f0e29e3329f13be347e86aba037549f982bd1c23171b4901d718a73682
|
|
BLAKE2b-256 checksum How to use checksums |
69c5eb978500bffd0e275a21d744b898238393e49d784328e4f0ae1bee40fab5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.10.12
|
Security for docparity 0.1.0
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.
No provenance found
None of the files in this release were uploaded with provenance metadata.
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.