Python Bindings for BerkeleyDB 3.3 thru 4.5

2008-03: This project has moved! See The new pybsddb Python BerkeleyDB library website.

The website, mailing list and subversion repository all reside at the new location linked to above. The links below may be out of date. -greg

Introduction
This handcrafted package contains Python wrappers for Berkeley DB, the Open Source embedded database system. Berkeley DB is a programmatic toolkit that provides high-performance built-in database support for desktop and server applications.

The Berkeley DB access methods include B+tree, Extended Linear Hashing, Fixed and Variable-length records, and Queues. Berkeley DB provides full transactional support, database recovery, online backups, multi-threaded and multi-process access, etc.

The Python wrappers allow you to store Python string objects of any length, keyed either by strings or integers depending on the database access method. With the use of another module in the package standard shelve-like functionality is provided allowing you to store any picklable Python object!

History
This module was started by Andrew Kuchling (amk) to remove the dependency on SWIG in a package by Gregory P. Smith who based his work on a similar package by Robin Dunn which wrapped Berkeley DB 2.7.x.

Development then returned full circle back to Robin Dunn working in behalf of Digital Creations to complete the SWIG-less wrapping of the DB 3.x API and to build a solid unit test suite. Having completed that, Robin is now busy with another project (wxPython) and Greg has returned as maintainer.

This module is included in the standard python >= 2.3 distribution as the bsddb module. The only reason you should look here is for documentation or to get a more up to date version. The bsddb.db module aims to mirror much of the BerkeleyDB C/C++ API.

The Important Stuff
And now, finally, here is the stuff you came to this site for:

SourceForge Logo