django/docs/howto/initial-data.txt

141 lines
5.3 KiB
Plaintext

.. _howto-initial-data:
=================================
Providing initial data for models
=================================
It's sometimes useful to pre-populate your database with hard-coded data when
you're first setting up an app. There's a couple of ways you can have Django
automatically create this data: you can provide `initial data via fixtures`_, or
you can provide `initial data as SQL`_.
In general, using a fixture is a cleaner method since it's database-agnostic,
but initial SQL is also quite a bit more flexible.
.. _initial data as sql: `providing initial sql data`_
.. _initial data via fixtures: `providing initial data with fixtures`_
Providing initial data with fixtures
====================================
A fixture is a collection of data that Django knows how to import into a
database. The most straightforward way of creating a fixture if you've already
got some data is to use the :djadmin:`manage.py dumpdata` command. Or, you can
write fixtures by hand; fixtures can be written as XML, YAML, or JSON documents.
The :ref:`serialization documentation <topics-serialization>` has more details
about each of these supported :ref:`serialization formats
<serialization-formats>`.
As an example, though, here's what a fixture for a simple ``Person`` model might
look like in JSON:
.. code-block:: js
[
{
"model": "myapp.person",
"pk": 1,
"fields": {
"first_name": "John",
"last_name": "Lennon",
}
},
{
"model": "myapp.person",
"pk": 2,
"fields": {
"first_name": "Paul",
"last_name": "McCartney",
}
},
]
And here's that same fixture as YAML:
.. code-block:: none
- model: myapp.person
pk: 1
fields:
first_name: John
last_name: Lennon
- model: myapp.person
pk: 2
fields:
first_name: Paul
last_name: McCartney
You'll store this data in a ``fixtures`` directory inside you app.
Loading data is easy: just call :djadmin:`manage.py loaddata fixturename
<loaddata>`, where *fixturename* is the name of the fixture file you've created.
Every time you run :djadmin:`loaddata` the data will be read from the fixture
and re-loaded into the database. Note that this means that if you change one of
the rows created by a fixture and the run :djadmin:`loaddata` again you'll wipe
out any changes you've made.
Automatically loading initial data fixtures
-------------------------------------------
If you create a fixture named ``initial_data.[xml/yml/json]``, that fixture will
be loaded every time you run :djadmin:`syncdb`. This is extremely convenient,
but be careful: remember that the data will be refreshed *every time* you run
:djadmin:`syncdb`. So don't use ``initial_data`` for data you'll want to edit.
.. seealso::
Fixtures are also used by the :ref:`testing framework
<topics-testing-fixtures>` to help set up a consistent test environment.
.. _initial-sql:
Providing initial SQL data
==========================
Django provides a hook for passing the database arbitrary SQL that's executed
just after the CREATE TABLE statements when you run :djadmin:`syncdb`. You can
use this hook to populate default records, or you could also create SQL
functions, views, triggers, etc.
The hook is simple: Django just looks for a file called ``sql/<modelname>.sql``,
in your app directory, where ``<modelname>`` is the model's name in lowercase.
So, if you had a ``Person`` model in an app called ``myapp``, you could add
arbitrary SQL to the file ``sql/person.sql`` inside your ``myapp`` directory.
Here's an example of what the file might contain:
.. code-block:: sql
INSERT INTO myapp_person (first_name, last_name) VALUES ('John', 'Lennon');
INSERT INTO myapp_person (first_name, last_name) VALUES ('Paul', 'McCartney');
Each SQL file, if given, is expected to contain valid SQL statements
which will insert the desired data (e.g., properly-formatted
``INSERT`` statements separated by semicolons).
The SQL files are read by the :djadmin:`sqlcustom`, :djadmin:`sqlreset`,
:djadmin:`sqlall` and :djadmin:`reset` commands in :ref:`manage.py
<ref-django-admin>`. Refer to the :ref:`manage.py documentation
<ref-django-admin>` for more information.
Note that if you have multiple SQL data files, there's no guarantee of the order
in which they're executed. The only thing you can assume is that, by the time
your custom data files are executed, all the database tables already will have
been created.
Database-backend-specific SQL data
----------------------------------
There's also a hook for backend-specific SQL data. For example, you can have
separate initial-data files for PostgreSQL and MySQL. For each app, Django
looks for a file called ``<appname>/sql/<modelname>.<backend>.sql``, where
``<appname>`` is your app directory, ``<modelname>`` is the model's name in
lowercase and ``<backend>`` is the value of :setting:`DATABASE_ENGINE` in your
settings file (e.g., ``postgresql``, ``mysql``).
Backend-specific SQL data is executed before non-backend-specific SQL data. For
example, if your app contains the files ``sql/person.sql`` and
``sql/person.postgresql.sql`` and you're installing the app on PostgreSQL,
Django will execute the contents of ``sql/person.postgresql.sql`` first, then
``sql/person.sql``.