pandas_access

What it is and what it does

pandas_access is a thin wrapper around MDBTools that lets you load MS Access database tables directly into Pandas DataFrames. It exposes two main operations: listing tables in an .mdb file and reading a table (with optional chunking for large datasets). The package has no Python runtime dependencies beyond Pandas itself, but requires MDBTools to be installed as a system library before use.

The tool is designed for one-off data extraction tasks—reading a legacy .mdb file into a DataFrame for analysis or migration. It is not a full database driver; the description itself recommends alternatives for more advanced use cases. The subprocess-based design means it spawns external MDBTools processes for each operation, which may be slower than native drivers but avoids complex C bindings.

Use it for:

• Extract a single MS Access table into a Pandas DataFrame for data analysis or cleaning.
• Batch-read large Access tables in chunks to avoid memory exhaustion.
• List all tables in a legacy .mdb file to understand its schema before migration.
• One-time conversion of Access data to CSV or other formats via Pandas export methods.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Reads MS Access database files (.mdb) into Pandas DataFrames by wrapping the MDBTools command-line utility via subprocess.

No. The package is abandoned (last updated 2016-06-07, no commits since 2023-07-26), requires a system dependency (MDBTools) that adds significant install friction, and has not been maintained for years. If you must read an .mdb file, consider maintained alternatives.

Install

pandas-access on PyPI

pip install pandas-access

uv add pandas-access

poetry add pandas-access

Installing pandas_access

Before you install

High install friction: requires MDBTools as a system dependency (installed separately via Homebrew on macOS or equivalent on other platforms) before the Python package can function. Package is abandoned—last release was 2016-06-07 with no commits since 2023-07-26.

License in practice

MIT License (permissive) imposes no significant restrictions on use, modification, or distribution.

Quickstart

import pandas_access as mdb

# List tables
for tbl in mdb.list_tables("my.mdb"):
    print(tbl)

# Read a table
df = pandas_access.read_table("my.mdb", "MyTable")

# Read in chunks
for chunk in pandas_access.read_table("my.mdb", "MyTable", chunksize=10000):
    process(chunk)

MDBTools must be installed as a system library (e.g., via Homebrew on macOS) before pandas_access can function.

Verify before relying

• Whether MDBTools installation works reliably on Windows and Linux, or if macOS is the primary supported platform.
• Current compatibility with modern Pandas versions and Python 3.x, given no updates since 2016-06-07.
• Whether the subprocess-based approach handles concurrent access or large files without stability issues.

Package facts

Evidence: pandas_access-0.0.1.tar.gz

Tags

More Front-Ends packages