DBAPI 2.0 Cursor for Datascope

This is a implementation of the DBAPI 2.0 database API standard for the Datascope database system designed by Boulder Real Time Technologies.

Summary

The purpose of this module is to abstract processing programs from any specific database backend. See the PEP 249 -- Python Database API Specification v2.0 for details on the specifications, and the python class, function and method docstrings for details on the implementation.

Implementation of Execute

The execute method of the Cursor currently accepts the string name of any Dbptr method. A subsequent sequence or mapping is then passed as arguments to this method on the current Cursor pointer. Variable substitution is currently not supported. Any execution returning a pointer sets this as the new cursor pointer and returns the number of records. All other returns values (i.e. queries) are returned directly.

As a non-standard convenience, one can run any Dbptr method directly on the 'execute' object as if it were a pointer:

Example

>>> from curds2.dbapi2 import connect
>>> curs = connect('/tmp/spam').cursor()

# Standard DBAPI2 style
>>> nrecs1 = curs.execute('lookup', {'table':'origin'})
>>> nrecs2 = curs.executemany('join',[('assoc', False), ('arrival', True)])

# Full function call via native API:
>>> nrecs3 = curs.execute.lookup(table='origin')
>>> nrecs4 = curs.execute.join('assoc', outer=False)
>>> nrecs1 == nrecs3
True

Customizations

NULL support

Datascope has no NULL type, each field defines its own value which compares equal to NULL. Therefore, NULLs must be explicitly looked up and converted, at a slight performance overhead. This can be enabled for this module by setting the Cursor attribute CONVERT_NULL to True. All fields in any rows returned the the fetch* methods will contain a python None for any NULL value.

DATETIME support

There is support for converting Datascope floats with type dbTIME to a date object. The default uses the TimestampFromTicks function, which defaults to return a python datetime.datetime. This can be turned on by setting the Cursor attribute CONVERT_DATETIME to True. The TimestampFromTicks function can be changed at the module level with any function that accepts an epoch float timestamp (i.e. the ObsPy UTCDateTime constructor)

Factory support

Cursor Factory

Custom Cursors are allowed by passing the class to the cursor_factory attribute of any Connection. Possible uses would be to inherit the current Cursor class and override an internal method for more efficient object-relational mapping of rows...

Row Factory

This module supports row factory classes similar to those of the sqlite3 (among others) implementation of the DBAPI. Instances of a Cursor or Connection have a attribute called row_factory. Setting this attribute to a special class constuctor which has the format: GenericRowFactory(cursor, row) allows for the custom building of rows. The default row returned by the fetch* methods is the standard tuple. Currently this module has several pre-defined row factory classes:

• NamedTupleRow - Rows of python namedtuples with attribute-style access to each item.
• OrderedDictRow - Rows of python OrderedDict instances.

Raw Interface

An implementation of the module using only the 'raw' interface C-library wrapper is available, as curds2.raw.dbapi2. Not recommended for most uses, becuase the syntax of calling functions via 'execute' is basically the same as the C functions, and is pretty inconvenient for interactive use and a bugger to script. Use for more speed, more complete access to the lib wrapper, or if you just want to run 'dbprocess' lines and don't have to make dbjoin and dblookup calls. The syntax for execute then becomes the full function name:

>>> from curds2.raw.dbapi2 import connect
>>> curs = connect('demo').cursor()

# Closer to C usage
>>> nrecs = curs.execute('dblookup', ('','origin','','') )

# Example of useful application
>>> nrecs = curs.execute('dbprocess', (['dbopen origin', 'dbjoin assoc', 'dbjoin arrival'],) )

Contact

Copyright 2013 Mark Williams at Nevada Seismological Laboratory

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.