Metadata-Version: 2.4
Name: abstract_utilities
Version: 0.2.2.787
Summary: Dependency-free, stdlib-only build of abstract_utilities. Heavy libs (pandas, geopandas, tiktoken, flask, werkzeug, dotenv, PDF stack) are optional and lazy — the package imports with none installed.
Author-email: putkoff <partners@abstractendeavors.com>
License: MIT
Project-URL: Homepage, https://github.com/AbstractEndeavors/abstract_essentials
Keywords: utilities,stdlib,helpers,dependency-free
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Operating System :: OS Independent
Classifier: Topic :: Software Development :: Libraries
Requires-Python: >=3.8
Description-Content-Type: text/markdown

# abstract-package-test

> **This is `abstract_utilities`.**
> The distribution is named `abstract-package-test` (a test upload), but the
> import package it installs is **`abstract_utilities`** — the exact same API:
>
> ```python
> import abstract_utilities as au
> ```

## What this is

A **dependency-free, stdlib-only** build of `abstract_utilities`. Importing it
pulls in **no third-party packages**. The heavy dependencies it used to import
unconditionally are now optional: each feature uses its library only when that
library is already installed, and otherwise falls back to a standard-library
implementation (or degrades cleanly).

| Area | Installed | Not installed |
|------|-----------|---------------|
| pandas / geopandas (readers: csv/tsv/txt, xlsx, geojson, …) | real DataFrames | stdlib list-of-dicts; `.xlsx` via a stdlib zip/XML reader; `.xls`/`.parquet`/`.shp` raise a clear "install X" error |
| tiktoken (token chunking) | exact BPE counts | calibrated stdlib approximate encoder |
| flask `jsonify` | real `Response` (`application/json`) | vendored stdlib JSON serializer |
| werkzeug `secure_filename` | werkzeug's | vendored stdlib copy (identical behaviour) |
| PDF stack (pdfplumber/pytesseract/PyPDF2/pdf2image) | full PDF reading/OCR | PDFs are **voided with a single warning** |
| python-dotenv `load_dotenv` | python-dotenv | vendored stdlib `.env` parser |

`import abstract_utilities` succeeds with none of the above installed.

## Notes

- Import name is `abstract_utilities`, so it can shadow a separately-installed
  `abstract_utilities`. This is a **test** distribution.
- A previously-vendored copy of yt_dlp's internal `utils` (an unused artifact
  that made the package require `yt_dlp`) has been removed.
