Framework for creating cmd.Cmd powered interactive command line database interfaces, ala psql and mysql
Switch branches/tags
Nothing to show
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


pyDBCLI is a package for creating interactive command line database query tools, using Python's cmd.Cmd, similar to tools like psql and mysql.


pyDBCLI itself just needs cmd to be present, so as long as you've got Python >=1.4, you're good.

The odbc extra requires PyODBC >=2.0.


Install pyDBCLI using easy_install:

easy_install pyDBCLI

Or from the setup script:

python install


For the most part you can drop pyDBCLI.utils.Utility in place of cmd.Cmd, with a few extra properties, e.g.

What you want the prompt to be replaced with when in the middle of a multi-line SQL statement.
DBAPI cursor to use to run queries, this should be set on the instance of the Utility rather than in the class definition.
Same as cursor but used for non-schema specific queries, e.g. display lists of schemas and other metadata.

List/tuple of "escaped" commands, e.g. those beginning with \, such as \d. This can be overridden in the class definition, but as you probably don't want to clear the existing set of commands, you can instead add to this list in your constructor:

def __init__(self, *args, **kwargs):
    Utility.__init__(self, *args, **kwargs)
Used for caching metadata, such as lists of schemas, and tables. If you redefine this, use something that implements UserDict or uses the DictMixin, as pyDBCLI.helpers.memoized expects a dict like instance.
Dict of Python type classes (as returned by DBAPI cursors) to appropriate SQL type strings.

The only methods not fully implemented by Utility are:

  • get_schemas
  • get_tables
  • get_columns
  • connect

As these can all be done in different ways depeneding on the DBAPI compliant library used to connect to the database being queried.

You can also, as with cmd.Cmd add your own commands handles (e.g. def do_mycommand), or override any existing commands if you need to.

Some helper methods are provided by pyDBCLI.helpers, specifically a method to pretty print tabular data, a memoize decorator that targets the data_cache property of Utility, and some CLI helpers for printing usage and handling errors.