databases.txt

.. _ref-databases:

===============================
Notes about supported databases
===============================

Django attempts to support as many features as possible on all database
backends. However, not all database backends are alike, and we've had to make
design decisions on which features to support and which assumptions we can make
safely.

This file describes some of the features that might be relevant to Django
usage. Of course, it is not intended as a replacement for server-specific
documentation or reference manuals.

.. _postgresql-notes:

PostgreSQL notes
================

PostgreSQL 8.2 to 8.2.4
-----------------------

The implementation of the population statistics aggregates ``STDDEV_POP`` and
``VAR_POP`` that shipped with PostgreSQL 8.2 to 8.2.4 are `known to be
faulty`_. Users of these releases of PostgreSQL are advised to upgrade to
`Release 8.2.5`_ or later. Django will raise a ``NotImplementedError`` if you
attempt to use the ``StdDev(sample=False)`` or ``Variance(sample=False)``
aggregate with a database backend that falls within the affected release range.

.. _known to be faulty: http://archives.postgresql.org/pgsql-bugs/2007-07/msg00046.php
.. _Release 8.2.5: http://developer.postgresql.org/pgdocs/postgres/release-8-2-5.html

Transaction handling
---------------------

:ref:`By default <topics-db-transactions>`, Django starts a transaction when a
database connection is first used and commits the result at the end of the
request/response handling. The PostgreSQL backends normally operate the same
as any other Django backend in this respect.

Autocommit mode
~~~~~~~~~~~~~~~

.. versionadded:: 1.1

If your application is particularly read-heavy and doesn't make many database
writes, the overhead of a constantly open transaction can sometimes be
noticeable. For those situations, if you're using the ``postgresql_psycopg2``
backend, you can configure Django to use *"autocommit"* behavior for the
connection, meaning that each database operation will normally be in its own
transaction, rather than having the transaction extend over multiple
operations. In this case, you can still manually start a transaction if you're
doing something that requires consistency across multiple database operations.
The autocommit behavior is enabled by setting the ``autocommit`` key in the
``DATABASE_OPTIONS`` part of your database in :setting:`DATABASES`::

    DATABASE_OPTIONS = {
        "autocommit": True,
    }

In this configuration, Django still ensures that :ref:`delete()
<topics-db-queries-delete>` and :ref:`update() <topics-db-queries-update>`
queries run inside a single transaction, so that either all the affected
objects are changed or none of them are.

.. admonition:: This is database-level autocommit

    This functionality is not the same as the
    :ref:`topics-db-transactions-autocommit` decorator. That decorator is a
    Django-level implementation that commits automatically after data changing
    operations. The feature enabled using the ``DATABASE_OPTIONS`` option
    provides autocommit behavior at the database adapter level. It commits
    after *every* operation.

If you are using this feature and performing an operation akin to delete or
updating that requires multiple operations, you are strongly recommended to
wrap you operations in manual transaction handling to ensure data consistency.
You should also audit your existing code for any instances of this behavior
before enabling this feature. It's faster, but it provides less automatic
protection for multi-call operations.

.. _mysql-notes:

MySQL notes
===========

Django expects the database to support transactions, referential integrity, and
Unicode (UTF-8 encoding). Fortunately, MySQL_ has all these features as
available as far back as 3.23. While it may be possible to use 3.23 or 4.0,
you'll probably have less trouble if you use 4.1 or 5.0.

MySQL 4.1
---------

`MySQL 4.1`_ has greatly improved support for character sets. It is possible to
set different default character sets on the database, table, and column.
Previous versions have only a server-wide character set setting. It's also the
first version where the character set can be changed on the fly. 4.1 also has
support for views, but Django currently doesn't use views.

MySQL 5.0
---------

`MySQL 5.0`_ adds the ``information_schema`` database, which contains detailed
data on all database schema. Django's ``inspectdb`` feature uses this
``information_schema`` if it's available. 5.0 also has support for stored
procedures, but Django currently doesn't use stored procedures.

.. _MySQL: http://www.mysql.com/
.. _MySQL 4.1: http://dev.mysql.com/doc/refman/4.1/en/index.html
.. _MySQL 5.0: http://dev.mysql.com/doc/refman/5.0/en/index.html

Storage engines
---------------

MySQL has several `storage engines`_ (previously called table types). You can
change the default storage engine in the server configuration.

The default engine is MyISAM_ [#]_. The main drawback of MyISAM is that it
doesn't currently support transactions or foreign keys. On the plus side, it's
currently the only engine that supports full-text indexing and searching.

The InnoDB_ engine is fully transactional and supports foreign key references.

The BDB_ engine, like InnoDB, is also fully transactional and supports foreign
key references. However, its use seems to be deprecated.

`Other storage engines`_, including SolidDB_ and Falcon_, are on the horizon.
For now, InnoDB is probably your best choice.

.. _storage engines: http://dev.mysql.com/doc/refman/5.0/en/storage-engines.html
.. _MyISAM: http://dev.mysql.com/doc/refman/5.0/en/myisam-storage-engine.html
.. _BDB: http://dev.mysql.com/doc/refman/5.0/en/bdb-storage-engine.html
.. _InnoDB: http://dev.mysql.com/doc/refman/5.0/en/innodb.html
.. _Other storage engines: http://dev.mysql.com/doc/refman/5.1/en/storage-engines-other.html
.. _SolidDB: http://forge.mysql.com/projects/project.php?id=139
.. _Falcon: http://dev.mysql.com/doc/falcon/en/index.html

.. [#] Unless this was changed by the packager of your MySQL package. We've
   had reports that the Windows Community Server installer sets up InnoDB as
   the default storage engine, for example.

MySQLdb
-------

`MySQLdb`_ is the Python interface to MySQL. Version 1.2.1p2 or later is
required for full MySQL support in Django.

.. note::
    If you see ``ImportError: cannot import name ImmutableSet`` when trying to
    use Django, your MySQLdb installation may contain an outdated ``sets.py``
    file that conflicts with the built-in module of the same name from Python
    2.4 and later. To fix this, verify that you have installed MySQLdb version
    1.2.1p2 or newer, then delete the ``sets.py`` file in the MySQLdb
    directory that was left by an earlier version.

.. _MySQLdb: http://sourceforge.net/projects/mysql-python

Creating your database
----------------------

You can `create your database`_ using the command-line tools and this SQL::

  CREATE DATABASE <dbname> CHARACTER SET utf8;

This ensures all tables and columns will use UTF-8 by default.

.. _create your database: http://dev.mysql.com/doc/refman/5.0/en/create-database.html

.. _mysql-collation:

Collation settings
~~~~~~~~~~~~~~~~~~

The collation setting for a column controls the order in which data is sorted
as well as what strings compare as equal. It can be set on a database-wide
level and also per-table and per-column. This is `documented thoroughly`_ in
the MySQL documentation. In all cases, you set the collation by directly
manipulating the database tables; Django doesn't provide a way to set this on
the model definition.

.. _documented thoroughly: http://dev.mysql.com/doc/refman/5.0/en/charset.html

By default, with a UTF-8 database, MySQL will use the
``utf8_general_ci_swedish`` collation. This results in all string equality
comparisons being done in a *case-insensitive* manner. That is, ``"Fred"`` and
``"freD"`` are considered equal at the database level. If you have a unique
constraint on a field, it would be illegal to try to insert both ``"aa"`` and
``"AA"`` into the same column, since they compare as equal (and, hence,
non-unique) with the default collation.

In many cases, this default will not be a problem. However, if you really want
case-sensitive comparisons on a particular column or table, you would change
the column or table to use the ``utf8_bin`` collation. The main thing to be
aware of in this case is that if you are using MySQLdb 1.2.2, the database backend in Django will then return
bytestrings (instead of unicode strings) for any character fields it returns
receive from the database. This is a strong variation from Django's normal
practice of *always* returning unicode strings. It is up to you, the
developer, to handle the fact that you will receive bytestrings if you
configure your table(s) to use ``utf8_bin`` collation. Django itself should work
smoothly with such columns, but if your code must be prepared to call
``django.utils.encoding.smart_unicode()`` at times if it really wants to work
with consistent data -- Django will not do this for you (the database backend
layer and the model population layer are separated internally so the database
layer doesn't know it needs to make this conversion in this one particular
case).

If you're using MySQLdb 1.2.1p2, Django's standard
:class:`~django.db.models.CharField` class will return unicode strings even
with ``utf8_bin`` collation. However, :class:`~django.db.models.TextField`
fields will be returned as an ``array.array`` instance (from Python's standard
``array`` module). There isn't a lot Django can do about that, since, again,
the information needed to make the necessary conversions isn't available when
the data is read in from the database. This problem was `fixed in MySQLdb
1.2.2`_, so if you want to use :class:`~django.db.models.TextField` with
``utf8_bin`` collation, upgrading to version 1.2.2 and then dealing with the
bytestrings (which shouldn't be too difficult) is the recommended solution.

Should you decide to use ``utf8_bin`` collation for some of your tables with
MySQLdb 1.2.1p2, you should still use ``utf8_collation_ci_swedish`` (the
default) collation for the :class:`django.contrib.sessions.models.Session`
table (usually called ``django_session`` and the table
:class:`django.contrib.admin.models.LogEntry` table (usually called
``django_admin_log``). Those are the two standard tables that use
:class:`~django.db.model.TextField` internally.

.. _fixed in MySQLdb 1.2.2: http://sourceforge.net/tracker/index.php?func=detail&aid=1495765&group_id=22307&atid=374932

Connecting to the database
--------------------------

Refer to the :ref:`settings documentation <ref-settings>`.

Connection settings are used in this order:

    1. ``DATABASE_OPTIONS``.
    2. ``DATABASE_NAME``, ``DATABASE_USER``,
       ``DATABASE_PASSWORD``, ``DATABASE_HOST``,
       ``DATABASE_PORT``
    3. MySQL option files.

In other words, if you set the name of the database in ``DATABASE_OPTIONS``,
this will take precedence over ``DATABASE_NAME``, which would override
anything in a `MySQL option file`_.

Here's a sample configuration which uses a MySQL option file::

    # settings.py
    DATABASES = {
        'default': {
            'DATABASE_ENGINE': "mysql",
            'DATABASE_OPTIONS': {
                'read_default_file': '/path/to/my.cnf',
            },
        }
    }


    # my.cnf
    [client]
    database = DATABASE_NAME
    user = DATABASE_USER
    password = DATABASE_PASSWORD
    default-character-set = utf8

Several other MySQLdb connection options may be useful, such as ``ssl``,
``use_unicode``, ``init_command``, and ``sql_mode``. Consult the
`MySQLdb documentation`_ for more details.

.. _MySQL option file: http://dev.mysql.com/doc/refman/5.0/en/option-files.html
.. _MySQLdb documentation: http://mysql-python.sourceforge.net/

Creating your tables
--------------------

When Django generates the schema, it doesn't specify a storage engine, so
tables will be created with whatever default storage engine your database
server is configured for. The easiest solution is to set your database server's
default storage engine to the desired engine.

If you're using a hosting service and can't change your server's default
storage engine, you have a couple of options.

    * After the tables are created, execute an ``ALTER TABLE`` statement to
      convert a table to a new storage engine (such as InnoDB)::

          ALTER TABLE <tablename> ENGINE=INNODB;

      This can be tedious if you have a lot of tables.

    * Another option is to use the ``init_command`` option for MySQLdb prior to
      creating your tables::

          DATABASE_OPTIONS = {
             "init_command": "SET storage_engine=INNODB",
          }

      This sets the default storage engine upon connecting to the database.
      After your tables have been created, you should remove this option.

    * Another method for changing the storage engine is described in
      AlterModelOnSyncDB_.

.. _AlterModelOnSyncDB: http://code.djangoproject.com/wiki/AlterModelOnSyncDB

Notes on specific fields
------------------------

Boolean fields
~~~~~~~~~~~~~~

Since MySQL doesn't have a direct ``BOOLEAN`` column type, Django uses a
``TINYINT`` column with values of ``1`` and ``0`` to store values for the
:class:`~django.db.models.BooleanField` model field. Refer to the documentation
of that field for more details, but usually this won't be something that will
matter unless you're printing out the field values and are expecting to see
``True`` and ``False.``.

Character fields
~~~~~~~~~~~~~~~~

Any fields that are stored with ``VARCHAR`` column types have their
``max_length`` restricted to 255 characters if you are using ``unique=True``
for the field. This affects :class:`~django.db.models.CharField`,
:class:`~django.db.models.SlugField` and
:class:`~django.db.models.CommaSeparatedIntegerField`.

Furthermore, if you are using a version of MySQL prior to 5.0.3, all of those
column types have a maximum length restriction of 255 characters, regardless
of whether ``unique=True`` is specified or not.

.. _sqlite-notes:

SQLite notes
============

SQLite_ provides an excellent development alternative for applications that
are predominantly read-only or require a smaller installation footprint. As
with all database servers, though, there are some differences that are
specific to SQLite that you should be aware of.

.. _SQLite: http://www.sqlite.org/

.. _sqlite-string-matching:

String matching for non-ASCII strings
--------------------------------------

SQLite doesn't support case-insensitive matching for non-ASCII strings. Some
possible workarounds for this are `documented at sqlite.org`_, but they are
not utilised by the default SQLite backend in Django. Therefore, if you are
using the ``iexact`` lookup type in your queryset filters, be aware that it
will not work as expected for non-ASCII strings.

.. _documented at sqlite.org: http://www.sqlite.org/faq.html#q18

SQLite 3.3.6 or newer strongly recommended
------------------------------------------

Versions of SQLite 3.3.5 and older contains the following bugs:

 * A bug when `handling`_ ``ORDER BY`` parameters. This can cause problems when
   you use the ``select`` parameter for the ``extra()`` QuerySet method. The bug
   can be identified by the error message ``OperationalError: ORDER BY terms
   must not be non-integer constants``.

 * A bug when handling `aggregation`_ together with DateFields and
   DecimalFields.

.. _handling: http://www.sqlite.org/cvstrac/tktview?tn=1768
.. _aggregation: http://code.djangoproject.com/ticket/10031

SQLite 3.3.6 was released in April 2006, so most current binary distributions
for different platforms include newer version of SQLite usable from Python
through either the ``pysqlite2`` or the ``sqlite3`` modules.

However, some platform/Python version combinations include older versions of
SQLite (e.g. the official binary distribution of Python 2.5 for Windows, 2.5.4
as of this writing, includes SQLite 3.3.4). There are (as of Django 1.1) even
some tests in the Django test suite that will fail when run under this setup.

As described :ref:`below<using-newer-versions-of-pysqlite>`, this can be solved
by downloading and installing a newer version of ``pysqlite2``
(``pysqlite-2.x.x.win32-py2.5.exe`` in the described case) that includes and
uses a newer version of SQLite. Python 2.6 for Windows ships with a version of
SQLite that is not affected by these issues.

Version 3.5.9
-------------

The Ubuntu "Intrepid Ibex" (8.10) SQLite 3.5.9-3 package contains a bug that
causes problems with the evaluation of query expressions. If you are using
Ubuntu "Intrepid Ibex", you will need to update the package to version
3.5.9-3ubuntu1 or newer (recommended) or find an alternate source for SQLite
packages, or install SQLite from source.

At one time, Debian Lenny shipped with the same malfunctioning SQLite 3.5.9-3
package. However the Debian project has subsequently issued updated versions
of the SQLite package that correct these bugs. If you find you are getting
unexpected results under Debian, ensure you have updated your SQLite package
to 3.5.9-5 or later.

The problem does not appear to exist with other versions of SQLite packaged
with other operating systems.

Version 3.6.2
--------------

SQLite version 3.6.2 (released August 30, 2008) introduced a bug into ``SELECT
DISTINCT`` handling that is triggered by, amongst other things, Django's
``DateQuerySet`` (returned by the ``dates()`` method on a queryset).

You should avoid using this version of SQLite with Django. Either upgrade to
3.6.3 (released September 22, 2008) or later, or downgrade to an earlier
version of SQLite.

.. _using-newer-versions-of-pysqlite:

Using newer versions of the SQLite DB-API 2.0 driver
----------------------------------------------------

.. versionadded:: 1.1

For versions of Python 2.5 or newer that include ``sqlite3`` in the standard
library Django will now use a ``pysqlite2`` interface in preference to
``sqlite3`` if it finds one is available.

This provides the ability to upgrade both the DB-API 2.0 interface or SQLite 3
itself to versions newer than the ones included with your particular Python
binary distribution, if needed.

"Database is locked" errors
-----------------------------------------------

SQLite is meant to be a lightweight database, and thus can't support a high
level of concurrency. ``OperationalError: database is locked`` errors indicate
that your application is experiencing more concurrency than ``sqlite`` can
handle in default configuration. This error means that one thread or process has
an exclusive lock on the database connection and another thread timed out
waiting for the lock the be released.

Python's SQLite wrapper has
a default timeout value that determines how long the second thread is allowed to
wait on the lock before it times out and raises the ``OperationalError: database
is locked`` error.

If you're getting this error, you can solve it by:

    * Switching to another database backend. At a certain point SQLite becomes
      too "lite" for real-world applications, and these sorts of concurrency
      errors indicate you've reached that point.

    * Rewriting your code to reduce concurrency and ensure that database
      transactions are short-lived.

    * Increase the default timeout value by setting the ``timeout`` database
      option option::

          DATABASE_OPTIONS = {
              # ...
             "timeout": 20,
              # ...
          }

      This will simply make SQLite wait a bit longer before throwing "database
      is locked" errors; it won't really do anything to solve them.

.. _oracle-notes:

Oracle notes
============

Django supports `Oracle Database Server`_ versions 9i and
higher. Oracle version 10g or later is required to use Django's
``regex`` and ``iregex`` query operators. You will also need at least
version 4.3.1 of the `cx_Oracle`_ Python driver.

Note that due to a Unicode-corruption bug in ``cx_Oracle`` 5.0, that
version of the driver should **not** be used with Django;
``cx_Oracle`` 5.0.1 resolved this issue, so if you'd like to use a
more recent ``cx_Oracle``, use version 5.0.1.

.. _`Oracle Database Server`: http://www.oracle.com/
.. _`cx_Oracle`: http://cx-oracle.sourceforge.net/

In order for the ``python manage.py syncdb`` command to work, your Oracle
database user must have privileges to run the following commands:

    * CREATE TABLE
    * CREATE SEQUENCE
    * CREATE PROCEDURE
    * CREATE TRIGGER

To run Django's test suite, the user needs these *additional* privileges:

    * CREATE USER
    * DROP USER
    * CREATE TABLESPACE
    * DROP TABLESPACE
    * CONNECT WITH ADMIN OPTION
    * RESOURCE WITH ADMIN OPTION

Connecting to the database
--------------------------

Your Django settings.py file should look something like this for Oracle::

    DATABASES = {
        'default': {
            'DATABASE_ENGINE': 'oracle',
            'DATABASE_NAME': 'xe',
            'DATABASE_USER': 'a_user',
            'DATABASE_PASSWORD': 'a_password',
            'DATABASE_HOST': '',
            'DATABASE_PORT': ''     ,
        }
    }


If you don't use a ``tnsnames.ora`` file or a similar naming method that
recognizes the SID ("xe" in this example), then fill in both
``DATABASE_HOST`` and ``DATABASE_PORT`` like so::

    DATABASES = {
        'default': {
            'DATABASE_ENGINE': 'oracle',
            'DATABASE_NAME': 'xe',
            'DATABASE_USER': 'a_user',
            'DATABASE_PASSWORD': 'a_password',
            'DATABASE_HOST': 'dbprod01ned.mycompany.com',
            'DATABASE_PORT': '1540',
        }
    }

You should supply both ``DATABASE_HOST`` and ``DATABASE_PORT``, or leave both
as empty strings.

Tablespace options
------------------

A common paradigm for optimizing performance in Oracle-based systems is the
use of `tablespaces`_ to organize disk layout. The Oracle backend supports
this use case by adding ``db_tablespace`` options to the ``Meta`` and
``Field`` classes.  (When you use a backend that lacks support for tablespaces,
Django ignores these options.)

.. _`tablespaces`: http://en.wikipedia.org/wiki/Tablespace

A tablespace can be specified for the table(s) generated by a model by
supplying the ``db_tablespace`` option inside the model's ``class Meta``.
Additionally, you can pass the ``db_tablespace`` option to a ``Field``
constructor to specify an alternate tablespace for the ``Field``'s column
index. If no index would be created for the column, the ``db_tablespace``
option is ignored::

    class TablespaceExample(models.Model):
        name = models.CharField(max_length=30, db_index=True, db_tablespace="indexes")
        data = models.CharField(max_length=255, db_index=True)
        edges = models.ManyToManyField(to="self", db_tablespace="indexes")

        class Meta:
            db_tablespace = "tables"

In this example, the tables generated by the ``TablespaceExample`` model
(i.e., the model table and the many-to-many table) would be stored in the
``tables`` tablespace. The index for the name field and the indexes on the
many-to-many table would be stored in the ``indexes`` tablespace. The ``data``
field would also generate an index, but no tablespace for it is specified, so
it would be stored in the model tablespace ``tables`` by default.

.. versionadded:: 1.0

Use the :setting:`DEFAULT_TABLESPACE` and :setting:`DEFAULT_INDEX_TABLESPACE`
settings to specify default values for the db_tablespace options.
These are useful for setting a tablespace for the built-in Django apps and
other applications whose code you cannot control.

Django does not create the tablespaces for you. Please refer to `Oracle's
documentation`_ for details on creating and managing tablespaces.

.. _`Oracle's documentation`: http://download.oracle.com/docs/cd/B19306_01/server.102/b14200/statements_7003.htm#SQLRF01403

Naming issues
-------------

Oracle imposes a name length limit of 30 characters. To accommodate this, the
backend truncates database identifiers to fit, replacing the final four
characters of the truncated name with a repeatable MD5 hash value.

NULL and empty strings
----------------------

Django generally prefers to use the empty string ('') rather than
NULL, but Oracle treats both identically. To get around this, the
Oracle backend coerces the ``null=True`` option on fields that have
the empty string as a possible value. When fetching from the database,
it is assumed that a NULL value in one of these fields really means
the empty string, and the data is silently converted to reflect this
assumption.

``TextField`` limitations
-------------------------

The Oracle backend stores ``TextFields`` as ``NCLOB`` columns. Oracle imposes
some limitations on the usage of such LOB columns in general:

  * LOB columns may not be used as primary keys.

  * LOB columns may not be used in indexes.

  * LOB columns may not be used in a ``SELECT DISTINCT`` list. This means that
    attempting to use the ``QuerySet.distinct`` method on a model that
    includes ``TextField`` columns will result in an error when run against
    Oracle. A workaround to this is to keep ``TextField`` columns out of any
    models that you foresee performing ``distinct()`` queries on, and to
    include the ``TextField`` in a related model instead.

.. _third-party-notes:

Using a 3rd-party database backend
==================================

In addition to the officially supported databases, there are backends provided
by 3rd parties that allow you to use other databases with Django:

* `Sybase SQL Anywhere`_
* `IBM DB2`_
* `Microsoft SQL Server 2005`_
* Firebird_
* ODBC_

The Django versions and ORM features supported by these unofficial backends
vary considerably. Queries regarding the specific capabilities of these
unofficial backends, along with any support queries, should be directed to
the support channels provided by each 3rd party project.

.. _Sybase SQL Anywhere: http://code.google.com/p/sqlany-django/
.. _IBM DB2: http://code.google.com/p/ibm-db/
.. _Microsoft SQL Server 2005: http://code.google.com/p/django-mssql/
.. _Firebird: http://code.google.com/p/django-firebird/
.. _ODBC: http://code.google.com/p/django-pyodbc/