NextERP/documentation
5
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[ADD] developer/reference: add a reference page on performance

Browse Source 
closes odoo/documentation#2794

X-original-commit: 4d8e3f9785
Signed-off-by: Antoine Vandevenne (anv) <anv@odoo.com>
Signed-off-by: Xavier Dollé (xdo) <xdo@odoo.com>
Co-authored-by: Antoine Vandevenne (anv) <anv@odoo.com>
This commit is contained in:
Xavier-Do 2021-09-30 14:53:09 +00:00 committed by Antoine Vandevenne (anv)
parent aab2d6615a
commit 46a80f60c5
17 changed files with 554 additions and 561 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
content/applications/general/developer_mode.rst
View File

@ -59,6 +59,8 @@ debug mode, add `?debug=0` instead.
:ref:`assets mode <frontend/framework/assets_debug_mode>`, and `?debug=tests` enables
the :ref:`tests mode <frontend/framework/tests_debug_mode>`.
.. _developer-mode/mode-tools:
Locate the mode tools
=====================

26
content/contributing/development/coding_guidelines.rst
View File

@ -589,32 +589,6 @@ Programming in Odoo
- As in python, use ``filtered``, ``mapped``, ``sorted``, ... methods to
ease code reading and performance.
Make your method work in batch
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
When adding a function, make sure it can process multiple records by iterating
on self to treat each record.
.. code-block:: python
def my_method(self)
for record in self:
record.do_cool_stuff()
For performance issue, when developing a 'stat button' (for instance), do not
perform a ``search`` or a ``search_count`` in a loop. It
is recommended to use ``read_group`` method, to compute all value in only one request.
.. code-block:: python
def _compute_equipment_count(self):
""" Count the number of equipment per category """
equipment_data = self.env['hr.equipment'].read_group([('category_id', 'in', self.ids)], ['category_id'], ['category_id'])
mapped_data = dict([(m['category_id'][0], m['category_id_count']) for m in equipment_data])
for category in self:
category.equipment_count = mapped_data.get(category.id, 0)
Propagate the context
~~~~~~~~~~~~~~~~~~~~~
The context is a ``frozendict`` that cannot be modified. To call a method with

8
content/developer/cli.rst
View File

@ -709,8 +709,9 @@ Database Population
.. program:: odoo-bin populate
Odoo CLI supports database population features. If the feature is
:ref:`implemented on a given model <reference/testing/populate/methods>`, it allows automatic data
generation of the model's records to test your modules in databases containing non-trivial amounts of records.
:ref:`implemented on a given model <reference/performance/populate/methods>`, it allows automatic
data generation of the model's records to test your modules in databases containing non-trivial
amounts of records.
.. code-block:: console
@ -727,8 +728,7 @@ generation of the model's records to test your modules in databases containing n
of a given model (cf. the :file:`populate` folder of modules for further details).
.. seealso::
:ref:`reference/testing/populate`
:ref:`reference/performance/populate`
.. _reference/cmdline/cloc:

1
content/developer/howtos.rst
View File

@ -12,7 +12,6 @@ Tutorials
howtos/website
howtos/backend
howtos/web_services
howtos/profilecode
howtos/company
howtos/accounting_localization
howtos/translations

131
content/developer/howtos/profilecode.rst
View File

@ -1,131 +0,0 @@
===================
Profiling Odoo code
===================
.. warning::
This tutorial requires :ref:`having installed Odoo <setup/install>`
and :doc:`writing Odoo code <backend>`
Graph a method
==============
Odoo embeds a profiler of code. This embedded profiler output can be used to
generate a graph of calls triggered by the method, number of queries, percentage
of time taken in the method itself as well as the time that the method took and
its sub-called methods.
.. code:: python
from odoo.tools.misc import profile
[...]
@profile('/temp/prof.profile')
def mymethod(...)
This produces a file called /temp/prof.profile
A tool called *gprof2dot* will produce a graph with this result:
.. code:: bash
gprof2dot -f pstats -o /temp/prof.xdot /temp/prof.profile
A tool called *xdot* will display the resulting graph:
.. code:: bash
xdot /temp/prof.xdot
Log a method
============
Another profiler can be used to log statistics on a method:
.. code:: python
from odoo.tools.profiler import profile
[...]
@profile
@api.model
def mymethod(...):
The statistics will be displayed into the logs once the method to be analysed is
completely reviewed.
.. code:: bash
2018-03-28 06:18:23,196 22878 INFO openerp odoo.tools.profiler:
calls queries ms
project.task ------------------------ /home/odoo/src/odoo/addons/project/models/project.py, 638
1 0 0.02 @profile
@api.model
def create(self, vals):
# context: no_log, because subtype already handle this
1 0 0.01 context = dict(self.env.context, mail_create_nolog=True)
# for default stage
1 0 0.01 if vals.get('project_id') and not context.get('default_project_id'):
context['default_project_id'] = vals.get('project_id')
# user_id change: update date_assign
1 0 0.01 if vals.get('user_id'):
vals['date_assign'] = fields.Datetime.now()
# Stage change: Update date_end if folded stage
1 0 0.0 if vals.get('stage_id'):
vals.update(self.update_date_end(vals['stage_id']))
1 108 631.8 task = super(Task, self.with_context(context)).create(vals)
1 0 0.01 return task
Total:
1 108 631.85
Dump stack
==========
Sending the SIGQUIT signal to an Odoo process (only available on POSIX) makes
this process output the current stack trace to log, with info level. When an
odoo process seems stuck, sending this signal to the process permit to know
what the process is doing, and letting the process continue his job.
Tracing code execution
======================
Instead of sending the SIGQUIT signal to an Odoo process often enough, to check
where the processes are performing worse than expected, we can use the `py-spy`_ tool to
do it for us.
Install py-spy
--------------
.. code:: bash
python3 -m pip install py-spy
Record executed code
--------------------
As py-spy is installed, we now record the executed code lines.
This tool will record, multiple times a second, the stacktrace of the process.
.. code:: bash
# record to raw file
py-spy record -o profile.json -f speedscope --pid <PID>
# OR record directly to svg
py-spy record -o profile.svg --pid <PID>
where <PID> is the process ID of the odoo process you want to graph.
To open profile.json you can use online tool `speedscope.app`_.
To open profile.svg you should use browser, because other viewer may not
support interactive part.
.. image:: profilecode/flamegraph.svg
.. _py-spy: https://github.com/benfred/py-spy
.. _speedscope.app: https://www.speedscope.app/

315
content/developer/howtos/profilecode/flamegraph.svg
View File

File diff suppressed because one or more lines are too long
Side by Side

Before

Width:  |  Height:  |  Size: 110 KiB

1
content/developer/reference/backend.rst
View File

@ -14,6 +14,7 @@ Python framework
backend/module
backend/reports
backend/security
backend/performance
backend/testing
backend/http
backend/mixins

540
content/developer/reference/backend/performance.rst Normal file
View File

@ -0,0 +1,540 @@
:custom-css: performance.css
===========
Performance
===========
.. _performance/profiling:
Profiling
=========
.. currentmodule:: odoo.tools.profiler
Profiling is about analysing the execution of a program and measure aggregated data. These data can
be the elapsed time for each function, the executed SQL queries...
While profiling does not improve the performance of a program by itself, it can prove very helpful
in finding performance issues and identifying which part of the program is responsible for them.
Odoo provides an integrated profiling tool that allows recording all executed queries and stack
traces during execution. It can be used to profile either a set of requests of a user session, or a
specific portion of code. Profiling results can be either inspected with the integrated `speedscope
<https://github.com/jlfwong/speedscope>`_ :dfn:`open source app allowing to visualize a flamegraph`
view or analyzed with custom tools by first saving them in a JSON file or in the database.
.. _performance/profiling/enable:
Enable the profiler
-------------------
The profiler can either be enabled from the user interface, which is the easiest way to do so but
allows profiling only web requests, or from Python code, which allows profiling any piece of code
including tests.
.. tabs::
.. tab:: Enable from the user interface
#. :ref:`Enable the developer mode <developer-mode>`.
#. Before starting a profiling session, the profiler must be enabled globally on the database.
This can be done in two ways:
- Open the :ref:`developer mode tools <developer-mode/mode-tools>`, then toggle the
:guilabel:`Enable profiling` button. A wizard suggests a set of expiry times for the
profiling. Click on :guilabel:`ENABLE PROFILING` to enable the profiler globally.
.. image:: performance/enable_profiling_wizard.png
- Go to :guilabel:`Settings --> General Settings --> Performance` and set the desired time to
the field :guilabel:`Enable profiling until`.
#. After the profiler is enabled on the database, users can enable it on their session. To do
so, toggle the :guilabel:`Enable profiling` button in the :ref:`developer mode tools
<developer-mode/mode-tools>` again. By default, the recommended options :guilabel:`Record
sql` and :guilabel:`Record traces` are enabled. To learn more about the different options,
head over to :ref:`performance/profiling/collectors`.
.. image:: performance/profiling_debug_menu.png
When the profiler is enabled, all the requests made to the server are profiled and saved into
an `ir.profile` record. Such records are grouped into the current profiling session which
spans from when the profiler was enabled until it is disabled.
.. note::
Odoo Online (SaaS) databases cannot be profiled.
.. tab:: Enable from Python code
Starting the profiler manually can be convenient to profile a specific method or a part of the
code. This code can be a test, a compute method, the entire loading, etc.
To start the profiler from Python code, call it as a context manager. You may specify *what*
you want to record through the parameters. A shortcut is available for profiling test classes:
:code:`self.profile()`. See :ref:`performance/profiling/collectors` for more information on
the `collectors` parameter.
.. example::
.. code-block:: python
with Profiler():
do_stuff()
.. example::
.. code-block:: python
with Profiler(collectors=['sql', PeriodicCollector(interval=0.1)]):
do_stuff()
.. example::
.. code-block:: python
with self.profile():
with self.assertQueryCount(__system__=1211):
do_stuff()
.. note::
The profiler is called outside of the `assertQueryCount` in order to catch queries made
when exiting the context manager (e.g., flush).
.. autoclass:: Profiler()
:special-members: __init__
When the profiler is enabled, all executions of a test method are profiled and saved into an
`ir.profile` record. Such records are grouped into a single profiling session. This is
especially useful when using the :code:`@warmup` and :code:`@users` decorators.
.. tip::
It can be complicated to analyze profiling results of a method that is called several times
because all the calls are grouped together in the stack trace. Add an **execution context**
as a context manager to break down the results into multiple frames.
.. example::
.. code-block:: python
for index in range(max_index):
with ExecutionContext(current_index=index): # Identify each call in speedscope results.
do_stuff()
.. _performance/profiling/analyse:
Analyse the results
-------------------
To browse the profiling results, make sure that the :ref:`profiler is enabled globally on the
database <performance/profiling/enable>`, then open the :ref:`developer mode tools
<developer-mode/mode-tools>` and click on the button in the top-right corner of the profiling
section. A list view of the `ir.profile` records grouped by profiling session opens.
.. image:: performance/profiling_web.png
:align: center
Each record has a clickable link that opens the speedscope results in a new tab.
.. image:: performance/flamegraph_example.png
:align: center
Speedscope falls out of the scope of this documentation but there are a lot of tools to try: search,
highlight of similar frames, zoom on frame, timeline, left heavy, sandwich view...
Depending on the profiling options that were activated, Odoo generates different view modes that you
can access from the top menu.
.. image:: performance/speedscope_modes.png
:align: center
- The :guilabel:`Combined` view shows all the SQL queries and traces merged togethers.
- The :guilabel:`Combined no context` view shows the same result but ignores the saved execution
context <performance/profiling/enable>`.
- The :guilabel:`sql (no gap)` view shows all the SQL queries as if they were executed one after
another, without any Python logic. This is useful for optimizing SQL only.
- The :guilabel:`sql (density)` view shows only all the SQL queries, leaving gap between them. This
can be useful to spot if eiter SQL or Python code is the problem, and to identify zones in where
many small queries could be batched.
- The :guilabel:`frames` view shows the results of only the :ref:`periodic collector
<performance/profiling/collectors/periodic>`.
.. important::
Even though the profiler has been designed to be as light as possible, it can still impact
performance, especially when using the :ref:`Sync collector
<performance/profiling/collectors/sync>`. Keep that in mind when analyzing speedscope results.
.. _performance/profiling/collectors:
Collectors
----------
Whereas the profiler is about the *when* of profiling, the collectors take care of the *what*.
Each collector specializes in collecting profiling data in its own format and manner. They can be
individually enabled from the user interface through their dedicated toggle button in the
:ref:`developer mode tools <developer-mode/mode-tools>`, or from Python code through their key or
class.
There are currently four collectors available in Odoo:
.. list-table::
:header-rows: 1
* - Name
- Toggle button
- Python key
- Python class
* - :ref:`SQL collector <performance/profiling/collectors/sql>`
- :guilabel:`Record sql`
- `sql`
- `SqlCollector`
* - :ref:`Periodic collector <performance/profiling/collectors/periodic>`
- :guilabel:`Record traces`
- `traces_async`
- `PeriodicCollector`
* - :ref:`QWeb collector <performance/profiling/collectors/qweb>`
- :guilabel:`Record qweb`
- `qweb`
- `QwebCollector`
* - :ref:`Sync collector <performance/profiling/collectors/sync>`
- No
- `traces_sync`
- `SyncCollector`
By default, the profiler enables the SQL and the Periodic collectors. Both when it is enabled from
the user interface or Python code.
.. _performance/profiling/collectors/sql:
SQL collector
~~~~~~~~~~~~~
The SQL collector saves all the SQL queries made to the database in the current thread (for all
cursors), as well as the stack trace. The overhead of the collector is added to the analysed thread
for each query, which means that using it on a lot of small queries may impact execution time and
other profilers.
It is especially useful to debug query counts, or to add information to the :ref:`Periodic collector
<performance/profiling/collectors/periodic>` in the combined speedscope view.
.. autoclass:: SQLCollector
.. _performance/profiling/collectors/periodic:
Periodic collector
~~~~~~~~~~~~~~~~~~
This collector runs in a separate thread and saves the stack trace of the analysed thread at every
interval. The interval (by default 10 ms) can be defined through the :guilabel:`Interval` option in
the user interface, or the `interval` parameter in Python code.
.. warning::
If the interval is set at a very low value, profiling long requests will generate memory issues.
If the interval is set at a very high value, information on short function executions will be
lost.
It is one of the best way to analyse performance as it should have a very low impact on the
execution time thanks to its separate thread.
.. autoclass:: PeriodicCollector
.. _performance/profiling/collectors/qweb:
QWeb collector
~~~~~~~~~~~~~~
This collector saves the Python execution time and queries of all directives. As for the :ref:`SQL
collector <performance/profiling/collectors/sql>`, the overhead can be important when executing a
lot of small directives. The results are different from other collectors in terms of collected data,
and can be analysed from the `ir.profile` form view using a custom widget.
It is mainly useful for optimizing views.
.. autoclass:: QwebCollector
.. _performance/profiling/collectors/sync:
Sync collector
~~~~~~~~~~~~~~
This collector saves the stack for every function's call and return and runs on the same thread,
which greatly impacts performance.
It can be useful to debug and understand complex flows, and follow their execution in the code. It
is however not recommended for performance analysis because the overhead is high.
.. autoclass:: SyncCollector
.. _performance/profiling/pitfalls:
Performance pitfalls
--------------------
- Be careful with randomness. Multiple executions may lead to different results. E.g., a garbage
collector being triggered during execution.
- Be careful with blocking calls. In some cases, external `c_call` may take some time before
releasing the GIL, thus leading to unexpected long frames with the :ref:`Periodic collector
<performance/profiling/collectors/periodic>`. This should be detected by the profiler and give a
warning. It is possible to trigger the profiler manually before such calls if needed.
- Pay attention to the cache. Profiling before that the `view`/`assets`/... are in cache can lead to
different results.
- Be aware of the profiler's overhead. The :ref:`SQL collector
<performance/profiling/collectors/sql>`'s overhead can be important when a lot of small queries
are executed. Profiling is practical to spot a problem but you may want to disable the profiler in
order to measure the real impact of a code change.
- Profiling results can be memory intensive. In some cases (e.g., profiling an install or a long
request), it is possible that you reach memory limit, especially when rendering the speedscope
results, which can lead to an HTTP 500 error. In this case, you may need to start the server with
a higher memory limit: `--limit-memory-hard $((8*1024**3))`.
.. _reference/performance/populate:
Database population
===================
Odoo CLI offers a :ref:`database population <reference/cmdline/populate>` feature through the CLI
command :command:`odoo-bin populate`.
Instead of the tedious manual, or programmatic, specification of test data, one can use this feature
to fill a database on demand with the desired number of test data. This can be used to detect
diverse bugs or performance issues in tested flows.
.. _reference/performance/populate/methods:
To populate a given model, the following methods and attributes can be defined.
.. currentmodule:: odoo.models
.. autoattribute:: Model._populate_sizes
.. autoattribute:: Model._populate_dependencies
.. automethod:: Model._populate
.. automethod:: Model._populate_factories
.. note::
You have to define at least :meth:`~odoo.models.Model._populate` or
:meth:`~odoo.models.Model._populate_factories` on the model to enable database population.
.. example::
.. code-block:: python
from odoo.tools import populate
class CustomModel(models.Model)
_inherit = "custom.some_model"
_populate_sizes = {"small": 100, "medium": 2000, "large": 10000}
_populate_dependencies = ["custom.some_other_model"]
def _populate_factories(self):
# Record ids of previously populated models are accessible in the registry
some_other_ids = self.env.registry.populated_models["custom.some_other_model"]
def get_some_field(values=None, random=None, **kwargs):
""" Choose a value for some_field depending on other fields values.
:param dict values:
:param random: seeded :class:`random.Random` object
"""
field_1 = values['field_1']
if field_1 in [value2, value3]:
return random.choice(some_field_values)
return False
return [
("field_1", populate.randomize([value1, value2, value3])),
("field_2", populate.randomize([value_a, value_b], [0.5, 0.5])),
("some_other_id", populate.randomize(some_other_ids)),
("some_field", populate.compute(get_some_field, seed="some_field")),
('active', populate.cartesian([True, False])),
]
def _populate(self, size):
records = super()._populate(size)
# If you want to update the generated records
# E.g setting the parent-child relationships
records.do_something()
return records
Population tools
----------------
Multiple population tools are available to easily create the needed data generators.
.. automodule:: odoo.tools.populate
:members: cartesian, compute, constant, iterate, randint, randomize
.. _performance/good_practices:
Good practices
==============
.. _performance/good_practices/batch:
Batch operations
----------------
When working with recordsets, it is almost always better to batch operations.
.. example::
Don't call a method that runs SQL queries while looping over a recordset because it will do so
for each record of the set.
.. rst-class:: bad-example
.. code-block:: python
def _compute_count(self):
for record in self:
domain = [('related_id', '=', record.id)]
record.count = other_model.search_count(domain)
Instead, replace the `search_count` with a `read_group` to execute one SQL query for the entire
batch of records.
.. rst-class:: good-example
.. code-block:: python
def _compute_count(self):
if self.ids:
domain = [('related_id', 'in', self.ids)]
counts_data = other_model.read_group(domain, ['related_id'], ['related_id'])
mapped_data = {
count['related_id'][0]: count['related_id_count'] for count in counts_data
}
else:
mapped_data = {}
for record in self:
record.count = mapped_data.get(record.id, 0)
.. note::
This example is not optimal nor correct in all cases. It is only a substitute for a
`search_count`. Another solution could be to prefetch and count the inverse `One2many` field.
.. example::
Don't create records one after another.
.. rst-class:: bad-example
.. code-block:: python
for name in ['foo', 'bar']:
model.create({'name': name})
Instead, accumulate the create values and call the `create` method on the batch. Doing so has
mostly no impact and helps the framework optimize fields computation.
.. rst-class:: good-example
.. code-block:: python
create_values = []
for name in ['foo', 'bar']:
create_values.append({'name': name})
records = model.create(create_values)
.. example::
Fail to prefetch the fields of a recordset while browsing a single record inside a loop.
.. rst-class:: bad-example
.. code-block:: python
for record_id in record_ids:
model.browse(record_id)
record.foo # One query is executed per record.
Instead, browse the entire recordset first.
.. rst-class:: good-example
.. code-block:: python
records = model.browse(record_ids)
for record in records:
record.foo # One query is executed for the entire recordset.
We can verify that the records are prefetched in batch by reading the field `prefetch_ids` which
includes each of the record ids.browsing all records together is unpractical,
If needed, the `with_prefetch` method can be used to disable batch prefetching:
.. code-block:: python
for values in values_list:
message = self.browse(values['id']).with_prefetch(self.ids)
.. _performance/good_practices/algorithmic_complexity:
Reduce the algorithmic complexity
---------------------------------
Algorithmic complexity is a measure of how long an algorithm would take to complete in regard to the
size `n` of the input. When the complexity is high, the execution time can grow quickly as the input
becomes larger. In some cases, the algorithmic complexity can be reduced by preparing the input's
data correctly.
.. example::
For a given problem, let's consider a naive algorithm crafted with two nested loops for which the
complexity in in O(n²).
.. rst-class:: bad-example
.. code-block:: python
for record in self:
for result in results:
if results['id'] == record.id:
record.foo = results['foo']
break
Assuming that all results have a different id, we can prepare the data to reduce the complexity.
.. rst-class:: good-example
.. code-block:: python
mapped_result = {result['id']: result['foo'] for result in results}
for record in self:
record.foo = mapped_result.get(record.id)
.. example::
Choosing the bad data structure to hold the input can lead to quadratic complexity.
.. rst-class:: bad-example
.. code-block:: python
invalid_ids = self.search(domain).ids
for record in self:
if record.id in invalid_ids:
...
If `invalid_ids` is a list-like data structure, the complexity of the algorithm may be quadratic.
Instead, prefer using set operations like casting `invalid_ids` to a set.
.. rst-class:: good-example
.. code-block:: python
invalid_ids = set(invalid_ids)
for record in self:
if record.id in invalid_ids:
...
Depending on the input, recordset operations can also be used.
.. rst-class:: good-example
.. code-block:: python
invalid_ids = self.search(domain)
for record in self - invalid_ids:
...
.. _performance/good_practices/index:
Use indexes
-----------
Database indexes can help fasten search operations, be it from a search in the or through the user
interface.
.. code-block:: python
name = fields.Char(string="Name", index=True)
.. warning::
Be careful not to index every field as indexes consume space and impact on performance when
executing one of `INSERT`, `UPDATE`, and `DELETE`.

BIN
content/developer/reference/backend/performance/enable_profiling_wizard.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 10 KiB

BIN
content/developer/reference/backend/performance/flamegraph_example.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 22 KiB

BIN
content/developer/reference/backend/performance/profiling_debug_menu.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 3.8 KiB

BIN
content/developer/reference/backend/performance/profiling_example_after.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 3.5 KiB

BIN
content/developer/reference/backend/performance/profiling_example_before.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 8.3 KiB

BIN
content/developer/reference/backend/performance/profiling_web.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 54 KiB

BIN
content/developer/reference/backend/performance/speedscope_modes.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 4.3 KiB

84
content/developer/reference/backend/testing.rst
View File

@ -864,90 +864,6 @@ you can use the :meth:`~odoo.tests.common.BaseCase.assertQueryCount` method, int
with self.assertQueryCount(11):
do_something()
.. _reference/testing/populate:
Database population
-------------------
Odoo CLI offers a :ref:`database population<reference/cmdline/populate>` feature.
.. code-block:: console
odoo-bin populate
Instead of the tedious manual, or programmatic, specification of test data,
one can use this feature to fill a database on demand with the desired number of test data.
This can be used to detect diverse bugs or performance issues in tested flows.
.. _reference/testing/populate/methods:
To specify this feature for a given model, the following methods and attributes can be defined.
.. currentmodule:: odoo.models
.. autoattribute:: Model._populate_sizes
.. autoattribute:: Model._populate_dependencies
.. automethod:: Model._populate
.. automethod:: Model._populate_factories
.. note::
You have to define at least :meth:`~odoo.models.Model._populate` or :meth:`~odoo.models.Model._populate_factories`
on the model to enable database population.
Example model
~~~~~~~~~~~~~
.. code-block:: python
from odoo.tools import populate
class CustomModel(models.Model)
_inherit = "custom.some_model"
_populate_sizes = {"small": 100, "medium": 2000, "large": 10000}
_populate_dependencies = ["custom.some_other_model"]
def _populate_factories(self):
# Record ids of previously populated models are accessible in the registry
some_other_ids = self.env.registry.populated_models["custom.some_other_model"]
def get_some_field(values=None, random=None, **kwargs):
""" Choose a value for some_field depending on other fields values.
:param dict values:
:param random: seeded :class:`random.Random` object
"""
field_1 = values['field_1']
if field_1 in [value2, value3]:
return random.choice(some_field_values)
return False
return [
("field_1", populate.randomize([value1, value2, value3])),
("field_2", populate.randomize([value_a, value_b], [0.5, 0.5])),
("some_other_id", populate.randomize(some_other_ids)),
("some_field", populate.compute(get_some_field, seed="some_field")),
('active', populate.cartesian([True, False])),
]
def _populate(self, size):
records = super()._populate(size)
# If you want to update the generated records
# E.g setting the parent-child relationships
records.do_something()
return records
Population tools
~~~~~~~~~~~~~~~~
Multiple population tools are available to easily create
the needed data generators.
.. automodule:: odoo.tools.populate
:members: cartesian, compute, constant, iterate, randint, randomize
.. _qunit: https://qunitjs.com/
.. _qunit_config.js: https://github.com/odoo/odoo/blob/51ee0c3cb59810449a60dae0b086b49b1ed6f946/addons/web/static/tests/helpers/qunit_config.js#L49
.. _web.tests_assets: https://github.com/odoo/odoo/blob/51ee0c3cb59810449a60dae0b086b49b1ed6f946/addons/web/views/webclient_templates.xml#L594

7
static/css/performance.css Normal file
View File

@ -0,0 +1,7 @@
.bad-example {
border-left: 3px solid red;
}
.good-example {
border-left: 3px solid green;
}