Metadata-Version: 2.4
Name: largecardinals
Version: 0.1.0
Summary: Symbolic large-cardinal specifications, typed implications and exact finite models
Author-email: "Dr. Dimitrios Geromichalos" <riskdatascience@web.de>
Maintainer-email: "Dr. Dimitrios Geromichalos" <riskdatascience@web.de>
License: Apache-2.0
Project-URL: Repository, https://github.com/rds0001/largecardinals-python
Keywords: set theory,large cardinals,mathematical logic,Mahlo,determinacy
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Mathematics
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: Apache Software License
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
License-File: NOTICE
Provides-Extra: graphs
Requires-Dist: networkx>=3.0; extra == "graphs"
Provides-Extra: plot
Requires-Dist: networkx>=3.0; extra == "plot"
Requires-Dist: matplotlib>=3.5; extra == "plot"
Provides-Extra: test
Requires-Dist: pytest>=7; extra == "test"
Requires-Dist: hypothesis>=6; extra == "test"
Dynamic: license-file

# largecardinals

A native Python toolkit for symbolic set theory: typed formulas, explicit theory
contexts, large-cardinal specifications, curated mathematical implications and
bounded exact finite models. Optional NetworkX support exports typed relations.

See the [API reference](https://github.com/rds0001/largecardinals-python/blob/v0.1.0/docs/API.md) and the
[guide to mathematical scope and evidence](https://github.com/rds0001/largecardinals-python/blob/v0.1.0/docs/semantics.rst).

Version 0.1.0 does not decide existence of large cardinals, prove consistency
of set theories, or implement a full proof assistant.
A catalogue premise is not a machine-checked proof of its theorem. The general
isolated-measurable measure-count statement has an unverified sufficient
hypothesis; its record remains `DRAFT`, returns `UNKNOWN` when applicable, and
cannot be used as a relation step. See the scope guide for details.

Python 3.10 or later is required. The core uses only the standard library.
Install from PyPI:

```sh
python -m pip install largecardinals==0.1.0
```

Graph export additionally needs NetworkX; the SVG example also uses Matplotlib.
For development, `python -m pip install '.[test,plot]'` installs the package and
its testing and plotting dependencies. Run `python -m pytest` from the repository.
The [CI matrix](https://github.com/rds0001/largecardinals-python/actions/workflows/ci.yml)
covers Python 3.10–3.14 on Linux, with additional Python 3.14 jobs on Windows and macOS.

```python
import largecardinals as lc

mahlo = lc.mahlo_property("kappa")
result = lc.query_relation("measurable", "mahlo")
assert result["status"] == "ESTABLISHED"
lc.check_derivation(result["witness"])

# Optional: pip install 'largecardinals[graphs]'
graph = lc.to_networkx(nodes=["measurable", "ramsey", "weakly_compact", "mahlo"])
```

Use `definition`, `lookup_statement` and `catalogue_manifest` to inspect the
precise conventions, assumptions and scope. Missing evidence returns `UNKNOWN`,
not a refutation or an independence assertion. Different relation kinds must not
be composed merely because a graph algorithm finds a path.

Copyright RiskDataScience GmbH. Author and maintainer: Dr. Dimitrios Geromichalos,
riskdatascience@web.de. Licensed under Apache License 2.0.
