sqlparams 3.0.0

Convert between various DB API 2.0 parameter styles.

SQL Params

sqlparams is a utility package for converting between various SQL parameter styles. This can simplify the use of SQL parameters in queries by allowing the use of named parameters where only ordinal are supported. Some Python DB API 2.0 compliant modules only support the ordinal qmark or format style parameters (e.g., pyodbc only supports qmark). This package provides a helper class, SQLParams, that is used to convert from any parameter style (qmark, numeric, named, format, pyformat; and the non-standard numeric_dollar and named_dollar), and have them safely converted to the desired parameter style.

Tutorial

You first create an SQLParams instance specifying the named parameter style you’re converting from, and what ordinal style you’re converting to. Let’s convert from named to qmark style:

>>> import sqlparams
>>> query = sqlparams.SQLParams('named', 'qmark')

Now, lets to convert a simple SQL SELECT query using the SQLParams.format method which accepts an SQL query, and a dict of parameters:

>>> sql, params = query.format('SELECT * FROM users WHERE name = :name;', {'name': "Thorin"})

This returns the new SQL query using ordinal qmark parameters with the corresponding list of ordinal parameters, which can be passed to the .execute() method on a database cursor:

>>> print sql
SELECT * FROM users WHERE name = ?;
>>> print params
['Thorin']

tuples are also supported which allows for safe use of the SQL IN operator:

>>> sql, params = query.format("SELECT * FROM users WHERE name IN :names;", {'names': ("Dori", "Nori", "Ori")})
>>> print sql
SELECT * FROM users WHERE name in (?,?,?);
>>> print params
['Dori', 'Nori', 'Ori']

You can also format multiple parameters for a single, shared query useful with the .executemany() method of a database cursor:

>>> sql, manyparams = query.formatmany("UPDATE users SET age = :age WHERE name = :name;", [{'name': "Dwalin", 'age': 169}, {'name': "Balin", 'age': 178}])
>>> print sql
UPDATE users SET age = ? WHERE name = ?;
>>> print manyparams
[[169, 'Dwalin'], [178, 'Balin']]

Please note that if an expanded tuple is used in SQLParams.formatmany, the tuple must be the same size in each of the parameter lists. Otherwise, you might well use SQLParams.format in a for-loop.

Source

The source code for sqlparams is available from the GitHub repo cpburnz/python-sql-parameters.

Installation

sqlparams can be installed from source with:

python setup.py install

sqlparams is also available for install through PyPI:

pip install sqlparams

Documentation

Documentation for sqlparams is available on Read the Docs.

Change History

3.0.0 (2020-04-04)

• Major changes to internal implementation.

• Support converting any parameter style to any parameter style (all named, numeric, and ordinal styles).

• Renamed attribute named to in_style on sqlparams.SQLParams.

• Renamed attribute ordinal to out_style on sqlparams.SQLParams.

• Removed attributes match and replace from sqlparams.SQLParams which should have been private.

• Named parameters must now be valid identifiers (can no longer start with a digit to help prevent incorrectly matching common strings such as datetimes). Fixes Issue #4.

• Issue #7: Support dollar sign style for numeric and named parameters: ``

2.0.0 (2020-02-26)

• Drop support for EOL Python 2.7, 3.2, 3.3, 3.4.

1.2.0 (2020-02-26)

• Require setuptools.

• Support up to Python 3.8.

1.1.2 (2018-05-04)

• Improved support for byte strings.

1.1.1 (2017-09-07)

• Fixed support for byte strings.

1.1.0 (2017-08-30)

• Support Python 3.2+.

1.0.3 (2012-12-28)

• Fixed documentation for issue 1.

1.0.2 (2012-12-22)

• Added sphinx documentation.

1.0.1 (2012-12-20)

• Fixed running test as a script.

1.0.0 (2012-12-20)

• Initial release.