Differences between revisions 37 and 40 (spanning 3 versions)
Revision 37 as of 2009-08-07 22:19:00
Size: 3940
Editor: jkakar
Comment: Improved introductory text and added a table of contents to improve navigation.
Revision 40 as of 2011-03-15 07:59:09
Size: 3962
Editor: therve
Deletions are marked like this. Additions are marked like this.
Line 61: Line 61:
 * [[http://twistedmatrix.com/users/radix/storm-api/|API docs]]  * [[http://people.canonical.com/~therve/storm/|API docs]]
Line 88: Line 88:
  {{{bzr branch http://bazaar.launchpad.net/~storm/storm/trunk}}}   {{{bzr branch lp:storm}}}
Line 102: Line 102:

Please follow the ReleaseProcedure when making a new release.

What is Storm?

Storm is an object-relational mapper (ORM) for Python developed at Canonical. The project was in development for more than a year for use in Canonical projects such as Launchpad and Landscape before being released as free software on July 9th, 2007.



  • Clean and lightweight API offers a short learning curve and long-term maintainability.
  • Storm is developed in a test-driven manner. An untested line of code is considered a bug.
  • Storm needs no special class constructors, nor imperative base classes.
  • Storm is well designed (different classes have very clear boundaries, with small and clean public APIs).
  • Designed from day one to work both with thin relational databases, such as SQLite, and big iron systems like PostgreSQL and MySQL.
  • Storm is easy to debug, since its code is written with a KISS principle, and thus is easy to understand.
  • Designed from day one to work both at the low end, with trivial small databases, and the high end, with applications accessing billion row tables and committing to multiple database backends.
  • It's very easy to write and support backends for Storm (current backends have around 100 lines of code).


  • Storm is fast.
  • Storm lets you efficiently access and update large datasets by allowing you to formulate complex queries spanning multiple tables using Python.
  • Storm allows you to fallback to SQL if needed (or if you just prefer), allowing you to mix "old school" code and ORM code
  • Storm handles composed primary keys with ease (no need for surrogate keys).
  • Storm doesn't do schema management, and as a result you're free to manage the schema as wanted, and creating classes that work with Storm is clean and simple.
  • Storm works very well connecting to several databases and using the same Python types (or different ones) with all of them.
  • Storm can handle obj.attr = <A SQL expression> assignments, when that's really needed (the expression is executed at INSERT/UPDATE time).

  • Storm handles relationships between objects even before they were added to a database.
  • Storm works well with existing database schemas.
  • Storm will flush changes to the database automatically when needed, so that queries made affect recently modified objects.


There's a tutorial available. We are always working on improving the documentation. Questions are welcome in the mailing list.


Storm is licensed under the LGPL 2.1. Contributions must have copyright assigned to Canonical.


The Storm mailing list is publicly available at:

There is also a #storm IRC channel on irc.freenode.net; stop by and chat! We have a PublishBot up and running.


Development of Storm may be tracked in Launchpad:

The source code may be obtained using Bazaar:

  • bzr branch lp:storm

Code may be browsed at:

If you want to contribute, please see DevelopmentProcedure.


You can find released files at:

Please follow the ReleaseProcedure when making a new release.

FrontPage (last edited 2020-05-28 12:17:27 by cjwatson)