test_ok1/py/doc/test.txt

================================
The ``py.test`` tool and library 
================================

.. contents::
.. sectnum::

starting point: ``py.test`` command line tool 
=============================================

First, see `getting started`_ for how to install the 'py.test' tool 
on your system.

``py.test`` is the command line tool to run tests. You can supply it
with any Python module by passing it as an argument::

  py.test test_sample.py

``py.test`` looks for any functions and methods in the module that
start with with ``test_`` and will then run those methods.  Assertions
about test outcomes are done via the standard ``assert`` statement.

This means you can write tests without any boilerplate::

    # content of test_sample.py 
    def test_answer():
        assert 42 == 43 

As you can see, you can have test functions as well as test
methods. This in contrast to the Python standard library's
``unittest.py``.

You can use ``py.test`` to run all tests in a directory structure by
invoking it without any arguments::

  py.test

This will automatically collect and run any Python module whose filenames 
start with ``test_`` or ends with ``_test`` from the directory and any
subdirectories, starting with the current directory, and run them. Each 
Python test module is inspected for test methods starting with ``test_``. 

.. _features: 

Basic Features of ``py.test`` 
=============================

assert with the ``assert`` statement
------------------------------------

Writing assertions is very simple and this is one of py.test's
most noticeable features, as you can use the ``assert``
statement with arbitrary expressions.  For example you can 
write the following in your tests:: 

     assert hasattr(x, 'attribute') 

to state that your object has a certain ``attribute``. In case this
assertion fails the test ``reporter`` will provide you with a very
helpful analysis and a clean traceback.

Note that in order to display helpful analysis of a failing
``assert`` statement some magic takes place behind the
scenes.  For now, you only need to know that if something
looks strange or you suspect a bug in that
*behind-the-scenes-magic* you may turn off the magic by 
providing the ``--nomagic`` option. 

how to write assertions about exceptions 
----------------------------------------

In order to write assertions about exceptions, you use
one of two forms::

    py.test.raises(Exception, func, *args, **kwargs) 
    py.test.raises(Exception, "func(*args, **kwargs)")

both of which execute the given function with args and kwargs and
asserts that the given ``Exception`` is raised.  The reporter will
provide you with helpful output in case of failures such as *no
exception* or *wrong exception*.


automatic collection of tests on all levels
-------------------------------------------

The automated test collection process walks the current
directory (or the directory given as a command line argument)
and all its subdirectories and collects python modules with a
leading ``test_`` or trailing ``_test`` filename.  From each 
test module every function with a leading ``test_`` or class with 
a leading ``Test`` name is collected.  The collecting process can 
be customized at directory, module or class level.  (see 
`collection process`_ for some implementation details). 

.. _`generative tests`: 

generative tests: yielding more tests
------------------------------------- 

*Generative tests* are test methods that are *generator functions* which
``yield`` callables and their arguments.  This is most useful for running a
test function multiple times against different parameters.
Example::

    def test_generative(): 
        for x in (42,17,49): 
            yield check, x 
    
    def check(arg): 
        assert arg % 7 == 0   # second generated tests fails!

Note that ``test_generative()`` will cause three tests 
to get run, notably ``check(42)``, ``check(17)`` and ``check(49)``
of which the middle one will obviously fail. 

.. _`selection by keyword`: 

selecting tests by keyword 
--------------------------

You can selectively run tests by specifiying a keyword 
on the command line. Example:: 

    py.test -k test_simple 

will run all tests that are found from the current directory 
and where the word "test_simple" equals the start of one part of the 
path  leading up to the test item.  Directory and file basenames as well 
as function, class and function/method names each form a possibly
matching name. 

    Note that the exact semantics are still experimental but 
    should always remain intuitive.   

testing with multiple python versions / executables 
---------------------------------------------------

With ``--exec=EXECUTABLE`` you can specify a python
executable (e.g. ``python2.2``) with which the tests 
will be executed. 


testing starts immediately 
--------------------------

Testing starts as soon as the first ``test item`` 
is collected.  The collection process is iterative 
and does not need to complete before your first 
test items are executed. 

no interference with cmdline utilities 
--------------------------------------

As ``py.test`` mainly operates as a separate cmdline 
tool you can easily have a command line utility and
some tests in the same file.  

debug with the ``print`` statement
----------------------------------

By default, ``py.test`` catches text written to stdout/stderr during
the execution of each individual test. This output will only be
displayed however if the test fails; you will not see it
otherwise. This allows you to put debugging print statements in your
code without being overwhelmed by all the output that might be
generated by tests that do not fail.

Each failing test that produced output during the running of the test
will have its output displayed in the ``recorded stdout`` section.

The catching of stdout/stderr output can be disabled using the 
``--nocapture`` option to the ``py.test`` tool.  Any output will 
in this case be displayed as soon as it is generated.

order of execution is guaranteed
--------------------------------

Tests will run in the order in which they appear in the files. 
If you invoke ``py.test`` multiple times you should find that tests 
execute in exactly the same order within each file. 

Besides making it easier to compare test output this allows 
multi-stage tests where you can rely on your test to iteratively 
build up a test structure.

`Note: This is not true for distributed tests`

useful tracebacks, recursion detection 
--------------------------------------

A lot of care is taken to present nice tracebacks in case of test
failure. Try::

    py.test py/documentation/example/pytest/failure_demo.py

to see a variety of 17 tracebacks, each tailored to a different
failure situation.

``py.test`` uses the same order for presenting tracebacks as Python
itself: the outer function is shown first, and the most recent call is
shown last. Similarly, a ``py.test`` reported traceback starts with your
failing test function and then works its way downwards. If the maximum
recursion depth has been exceeded during the running of a test, for
instance because of infinite recursion, ``py.test`` will indicate
where in the code the recursion was taking place.  You can  
inhibit traceback "cutting" magic by supplying ``--fulltrace``. 

no inheritance requirement 
--------------------------

Test classes are recognized by their leading ``Test`` name.  Unlike
``unitest.py``, you don't need to inherit from some base class to make
them be found by the test runner. Besides being easier, it also allows
you to write test classes that subclass from application level
classes.

disabling a test class
---------------------- 

If you want to disable a complete test class you
can set the class-level attribute ``disabled``. 
For example, in order to avoid running some tests on Win32:: 

    class TestEgSomePosixStuff: 
        disabled = sys.platform == 'win32'
    
        def test_xxx(self):
            ... 

testing for deprecated APIs
------------------------------

In your tests you can use ``py.test.deprecated_call(func, *args, **kwargs)``
to test that a particular function call triggers a DeprecationWarning. 
This is useful for testing phasing out of old APIs in your projects. 

Managing test state across test modules, classes and methods 
------------------------------------------------------------

Often you want to create some files, database connections or other
state in order to run tests in a certain environment.  With
``py.test`` there are three scopes for which you can provide hooks to
manage such state.  Again, ``py.test`` will detect these hooks in
modules on a name basis. The following module-level hooks will
automatically be called by the session::

    def setup_module(module):
        """ setup up any state specific to the execution
            of the given module. 
        """

    def teardown_module(module):
        """ teardown any state that was previously setup 
            with a setup_module method. 
        """

The following hooks are available for test classes::

    def setup_class(cls): 
        """ setup up any state specific to the execution
            of the given class (which usually contains tests). 
        """

    def teardown_class(cls): 
        """ teardown any state that was previously setup 
            with a call to setup_class.
        """

    def setup_method(self, method):
        """ setup up any state tied to the execution of the given 
            method in a class.  setup_method is invoked for every 
            test method of a class. 
        """

    def teardown_method(self, method): 
        """ teardown any state that was previously setup 
            with a setup_method call. 
        """

The last two hooks, ``setup_method`` and ``teardown_method``, are
equivalent to ``setUp`` and ``tearDown`` in the Python standard
library's ``unitest`` module.

All setup/teardown methods are optional.  You could have a
``setup_module`` but no ``teardown_module`` and the other way round.

Note that while the test session guarantees that for every ``setup`` a
corresponding ``teardown`` will be invoked (if it exists) it does
*not* guarantee that any ``setup`` is called only happens once. For
example, the session might decide to call the ``setup_module`` /
``teardown_module`` pair more than once during the execution of a test
module.

Experimental doctest support 
------------------------------------------------------------

If you want to integrate doctests, ``py.test`` now by default
picks up files matching the ``test_*.txt`` or ``*_test.txt`` 
patterns and processes them as text files containing doctests. 
This is an experimental feature and likely to change
its implementation. 

Working Examples
================

Example for managing state at module, class and method level 
------------------------------------------------------------

Here is a working example for what goes on when you setup modules, 
classes and methods:: 

    # [[from py/documentation/example/pytest/test_setup_flow_example.py]]

    def setup_module(module):
        module.TestStateFullThing.classcount = 0

    class TestStateFullThing:
        def setup_class(cls):
            cls.classcount += 1

        def teardown_class(cls):
            cls.classcount -= 1

        def setup_method(self, method):
            self.id = eval(method.func_name[5:])

        def test_42(self):
            assert self.classcount == 1
            assert self.id == 42

        def test_23(self):
            assert self.classcount == 1
            assert self.id == 23

    def teardown_module(module):
        assert module.TestStateFullThing.classcount == 0

For this example the control flow happens as follows::

    import test_setup_flow_example
    setup_module(test_setup_flow_example)
       setup_class(TestStateFullThing) 
           instance = TestStateFullThing()
           setup_method(instance, instance.test_42) 
              instance.test_42()
           setup_method(instance, instance.test_23) 
              instance.test_23()
       teardown_class(TestStateFullThing) 
    teardown_module(test_setup_flow_example)


Note that ``setup_class(TestStateFullThing)`` is called and not 
``TestStateFullThing.setup_class()`` which would require you
to insert ``setup_class = classmethod(setup_class)`` to make
your setup function callable. Did we mention that lazyness
is a virtue? 

.. _`basicpicture`:


Collecting and running tests / implementation remarks 
====================================================== 

In order to customize ``py.test`` it's good to understand 
its basic architure (WARNING: these are not guaranteed 
yet to stay the way they are now!)::

     ___________________
    |                   |
    |    Collector      |
    |___________________|
           / \                
            |                Item.run()
            |               ^
     receive test Items    /
            |             /execute test Item 
            |            /
     ___________________/
    |                   |      
    |     Session       |
    |___________________|
                                          
                        .............................
                        . conftest.py configuration .
                        . cmdline options           .
                        .............................


The *Session* basically receives test *Items* from a *Collector*, 
and executes them via the ``Item.run()`` method.  It monitors 
the outcome of the test and reports about failures and successes. 

.. _`collection process`: 

Collectors and the test collection process 
------------------------------------------

The collecting process is iterative, i.e. the session 
traverses and generates a *collector tree*.  Here is an example of such
a tree, generated with the command ``py.test --collectonly py/xmlobj``:: 

    <Directory 'xmlobj'>
        <Directory 'testing'>
            <Module 'test_html.py' (py.__.xmlobj.testing.test_html)>
                <Function 'test_html_name_stickyness'>
                <Function 'test_stylenames'>
                <Function 'test_class_None'>
                <Function 'test_alternating_style'>
            <Module 'test_xml.py' (py.__.xmlobj.testing.test_xml)>
                <Function 'test_tag_with_text'>
                <Function 'test_class_identity'>
                <Function 'test_tag_with_text_and_attributes'>
                <Function 'test_tag_with_subclassed_attr_simple'>
                <Function 'test_tag_nested'>
                <Function 'test_tag_xmlname'>


By default all directories not starting with a dot are traversed, 
looking for ``test_*.py`` and ``*_test.py`` files.  Those files 
are imported under their `package name`_. 

.. _`collector API`: 

test items are collectors as well
--------------------------------- 

To make the reporting life simple for the session object 
items offer a ``run()`` method as well.  In fact the session
distinguishes "collectors" from "items" solely by interpreting 
their return value.  If it is a list, then we recurse into 
it, otherwise we consider the "test" as passed.  

.. _`package name`: 

constructing the package name for modules
----------------------------------------- 

Test modules are imported under their fully qualified 
name.  Given a module ``path`` the fully qualified package 
name is constructed as follows:

* determine the last "upward" directory from ``path`` that 
  contains an ``__init__.py`` file.  Going upwards 
  means repeatedly calling the ``dirpath()`` method 
  on a path object (which returns the parent directory
  as a path object). 

* insert this base directory into the sys.path list 
  as its first element 

* import the root package 

* determine the fully qualified name for the module located 
  at ``path`` ... 

  * if the imported root package has a __package__ object  
    then call ``__package__.getimportname(path)`` 

  * otherwise use the relative path of the module path to
    the base dir and turn slashes into dots and strike
    the trailing ``.py``. 

The Module collector will eventually trigger 
``__import__(mod_fqdnname, ...)`` to finally get to 
the live module object.  

Side note: this whole logic is performed by local path 
object's ``pyimport()`` method. 

Module Collector 
----------------- 

The default Module collector looks for test functions 
and test classes and methods. Test functions and methods
are prefixed ``test`` by default.  Test classes must 
start with a capitalized ``Test`` prefix. 


Customizing the testing process 
===============================

writing conftest.py files
-----------------------------------

XXX


customizing the collecting and running process 
-----------------------------------------------

To introduce different test items you can create 
one or more ``conftest.py`` files in your project. 
When the collection process traverses directories 
and modules the default collectors will produce 
custom Collectors and Items if they are found 
in a local ``conftest.py`` file.  

example: perform additional ReST checs 
++++++++++++++++++++++++++++++++++++++

With your custom collectors or items you can completely 
derive from the standard way of collecting and running
tests in a localized manner.  Let's look at an example. 
If you invoke ``py.test --collectonly py/documentation`` 
then you get:: 

    <DocDirectory 'documentation'>
        <DocDirectory 'example'>
            <DocDirectory 'pytest'>
                <Module 'test_setup_flow_example.py' (test_setup_flow_example)>
                    <Class 'TestStateFullThing'>
                        <Instance '()'>
                            <Function 'test_42'>
                            <Function 'test_23'>
        <ReSTChecker 'TODO.txt'>
            <ReSTSyntaxTest 'TODO.txt'>
            <LinkCheckerMaker 'checklinks'>
        <ReSTChecker 'api.txt'>
            <ReSTSyntaxTest 'api.txt'>
            <LinkCheckerMaker 'checklinks'>
                <CheckLink 'getting-started.html'>
        ... 

In ``py/documentation/conftest.py`` you find the following 
customization:: 

    class DocDirectory(py.test.collect.Directory):

        def run(self):
            results = super(DocDirectory, self).run()
            for x in self.fspath.listdir('*.txt', sort=True):
                    results.append(x.basename)
            return results

        def join(self, name):
            if not name.endswith('.txt'):
                return super(DocDirectory, self).join(name)
            p = self.fspath.join(name)
            if p.check(file=1):
                return ReSTChecker(p, parent=self)

    Directory = DocDirectory

The existence of the 'Directory' name in the 
``pypy/documentation/conftest.py`` module makes the collection 
process defer to our custom "DocDirectory" collector.  We extend 
the set of collected test items by ``ReSTChecker`` instances 
which themselves create ``ReSTSyntaxTest`` and ``LinkCheckerMaker`` 
items.  All of this instances (need to) follow the `collector API`_. 


Customizing the collection process in a module
---------------------------------------------- 

    REPEATED WARNING: details of the collection and running process are 
    still subject to refactorings and thus details will change.  
    If you are customizing py.test at "Item" level then you 
    definitely want to be subscribed to the `py-dev mailing list`_ 
    to follow ongoing development.

If you have a module where you want to take responsibility for
collecting your own test Items and possibly even for executing
a test then you can provide `generative tests`_ that yield 
callables and possibly arguments as a tuple.   This should 
serve some immediate purposes like paramtrized tests. 

The other extension possibility goes deeper into the machinery 
and allows you to specify a custom test ``Item`` class which 
is responsible for setting up and executing an underlying 
test.  [XXX not working: You can integrate your custom ``py.test.Item`` subclass 
by binding an ``Item`` name to a test class.]  Or you can 
extend the collection process for a whole directory tree 
by putting Items in a ``conftest.py`` configuration file. 
The collection process constantly looks at according names 
in the *chain of conftest.py* modules to determine collectors
and items at ``Directory``, ``Module``, ``Class``, ``Function`` 
or ``Generator`` level.  Note that, right now, except for ``Function`` 
items all classes are pure collectors, i.e. will return a list 
of names (possibly empty). 

XXX implement doctests as alternatives to ``Function`` items. 

Customizing execution of Functions 
---------------------------------- 

- Function test items allow total control of executing their 
  contained test method.  ``function.run()`` will get called by the
  session in order to actually run a test.  The method is responsible
  for performing proper setup/teardown ("Test Fixtures") for a 
  Function test. 

- ``Function.execute(target, *args)`` methods are invoked by
  the default ``Function.run()`` to actually execute a python 
  function with the given (usually empty set of) arguments. 

.. _`getting started`: getting-started.html
.. _`py-dev mailing list`: http://codespeak.net/mailman/listinfo/py-dev 


Automated Distributed Testing
==================================

If you have a project with a large number of tests, and you have 
machines accessible through SSH, ``py.test`` can distribute
tests across the machines.  It does not require any particular
installation on the remote machine sides as it uses `py.execnet`_ 
mechanisms to distribute execution.  Using distributed testing 
can speed up your development process considerably and it
may also be useful where you need to use a remote server
that has more resources (e.g. RAM/diskspace) than your
local machine. 

*WARNING*: support for distributed testing is experimental, 
its mechanics and configuration options may change without 
prior notice.  Particularly, not all reporting features 
of the in-process py.test have been integrated into
the distributed testing approach. 

Requirements
------------

Local requirements: 

* ssh client
* python

requirements for remote machines:

* ssh daemon running
* ssh keys setup to allow login without a password
* python (2.3 or 2.4 should work)
* unix like machine (reliance on ``os.fork``)

Hot to use it
-----------------------

When you issue ``py.test --session=R`` then your computer becomes
the distributor of tests ("master") and will start collecting
and distributing tests to several machines.  The machines
need to be specified in a ``conftest.py`` file.  

At start up, the master connects to each node using `py.execnet.SshGateway`_ 
and *rsyncs* all specified python packages to all nodes. 
Then the master collects all of the tests and immediately sends test item
descriptions to its connected nodes. Each node has a local queue of tests 
to run and begins to execute the tests, following the setup and teardown 
semantics.   The test are distributed at function and method level. 
When a test run on a node is completed it reports back the result
to the master. 

The master can run one of three reporters to process the events 
from the testing nodes: command line, rest output and ajaxy web based. 

.. _`py.execnet`: execnet.html
.. _`py.execnet.SshGateway`: execnet.html

Differences from local tests
----------------------------

* Test order is *not* guaranteed.
* Hanging nodes or tests are not detected properly.
* ``conftest.py`` cannot reference files outside of the copied packages.

Configuration
-------------

You must create a conftest.py in any parent directory above your tests.

The options that you need to specify in that conftest.py file are:

* `dist_hosts` - a list of ssh addresses (including a specific path if it 
  should be different than the default: ``$HOME/pytestcache-hostname``)
* `dist_rsync_roots` - a list of packages to copy to the remote machines.
* `dist_remotepython` - the remote python to run.
* `SessionOptions` - containing some specific tuning options

Sample configuration::

    dist_hosts = ['localhost', 'user@someserver:/tmp/somedir']
    dist_rsync_roots = ['pypy', 'py']
    dist_remotepython = 'python2.4'
    dist_nicelevel = 10 
    dist_boxing = True
    dist_maxwait = 100 
    dist_taskspernode = 10

Running server is done by ``-w`` command line option or ``--startserver``
(the former might change at some point due to conflicts).

Possible `SessionOptions` arguments:

* `nice_level` - Level of nice under which tests are run
* `runner_policy` - (for `LSession` only) - contains policy for the test boxig.
  `"plain_runner"` means no boxing, `"box_runner"` means boxing.
* `waittime` - Default maximum waiting time for remote host to execute
  one test.
* `max_tasks_per_node` - Maximum number of tasks which can be send to one node.
* `import_pypy` - Flag to control pypy importing for js regeneration (defaults
  to False)

Development Notes
-----------------

Changing the behavior of the web based reporter requires `pypy`_ since the
javascript is actually generated fom rpython source.

There exists as well `L/Rsession document`_ which discusses in more details
unique features and developement notes.

.. _`pypy`: http://codespeak.net/pypy
.. _`L/Rsession document`: test-distributed.html

Future/Planned Features of py.test 
==================================

Please note that the following descriptions of future features
sound factual although they aren't implemented yet.  This
allows easy migration to real documentation later.
Nevertheless, none of the described planned features is
set in stone, yet. In fact, they are open to discussion on
py-dev at codespeak dot net. 

Hey, if you want to suggest new features or command line options 
for py.test it would be great if you could do it by providing 
documentation for the feature.  Welcome to documentation driven 
development :-) 

selecting tests by queries/full text search 
-------------------------------------------

    Note: there already is experimental support for test `selection by keyword`_. 
    Otherwise the following is not yet implemented

You can selectively run tests by specifiying words 
on the command line in a google-like way. Example:: 

    py.test simple 

will run all tests that are found from the current directory 
and that have the word "simple" somewhere in their `test address`_.  
``test_simple1`` and ``TestSomething.test_whatever_simpleton`` would both 
qualify.  If you want to exclude the latter test method you could say:: 

    py.test -- simple -simpleton 

Note that the doubledash "--" signals the end of option parsing so
that "-simpleton" will not be misinterpreted as a command line option. 

Interpreting positional arguments as specifying search queries 
means that you can only restrict the set of tests. There is no way to 
say "run all 'simple' in addition to all 'complex' tests".  If this proves
to be a problem we can probably come up with a command line option 
that allows to specify multiple queries which all add to the set of
tests-to-consider. 

.. _`test address`: 

the concept of a test address
----------------------------- 

For specifiying tests it is convenient to define the notion
of a *test address*, representable as a filesystem path and a 
list of names leading to a test item.  If represented as a single 
string the path and names are separated by a `/` character, for example: 

    ``somedir/somepath.py/TestClass/test_method``

Such representations can be used to memoize failing tests 
by writing them out in a file or communicating them across
process and computer boundaries.