Metadata-Version: 2.4
Name: tablevault
Version: 0.2.1
Summary: Centralized data repository for cross-process data filtering in Python.
Author-email: Jinjin Zhao <j2zhao@uchicago.edu>
License-Expression: MIT
Keywords: example,package,python
Requires-Python: >=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: python-arango
Requires-Dist: psutil
Provides-Extra: dev
Requires-Dist: ruff; extra == "dev"
Requires-Dist: pytest; extra == "dev"
Requires-Dist: mkdocs-material; extra == "dev"
Requires-Dist: mike; extra == "dev"
Dynamic: license-file

# TableVault

Version 2.0 of TableVault. Retains basic functionality from v1.0 but with more robust backend storage and simpler API.

You can view the previous website at: www.tablevault.org.

# Changes from v1.0

- Data is stored by data type rather than by instance.

- Tables are stored in ArangoDB rather than as DataFrames.

- First level support for arrays, embeddings, and text files.

- API changes to match industry standards.

# New Core Feature: Data Queries

The main reason for the redesign is that we want to design the metadata layer to support robust queries from the ground up.

The API supports:

- Vector search over data descriptions

- Reproducible search (with timestamps)

- Direct and indirect data lineage (based on code traces)

I would also like to include:

# Planned Features

- Experimental LLM Layer to support natural language search over context

