Skip to main content
Flask EZ Test
=============

Making flask app's that serve html and use a flasksqlalchemy backend engine easier to test

Setup
-----
.. code::
$ pip install flaskeztest

Make sure you set the environemnt variable FLASK_APP to the file path of your main app module

.. code::
$ export FLASK_APP=myapp/app.py

flaskeztest will expect there to be a variable named 'app' under this module and another named 'db'.
'app' should be a Flask object and 'db' should be an SQLAlchemy object.

.. code:: python

# myapp/app.py
from model import db
app = Flask(__name__)
db.init_app(app)
...

# myapp/model.py
from flask_sqlalchemy import SQLAlchemy
db = SQLAlchemy()
...

Then run all test suites with eztest command

.. code::
$ eztest

Other options
~~~~~~~~~~~~~

for running one testcase under a suite

.. code::

$ eztest [suite] [testcase]

or

for running one whole suite

.. code::

$ eztest [suite]

where suite is the same as a suite module without .py and testcase is the same as the classname of the testcase

In General, the eztest object get's its settings from a special test config module which is set in app.config
as EZTEST_CONFIG_MODULE

i.e:

.. code:: python

# test/config.py

EZTEST_FIXTURES_DIR = 'test/fixtures'
EZTEST_SUITES_PACKAGE = 'test/suites'

Make sure that the suites package has a ```suite_names``` list in its __init__.py'

.. code:: python
# suites/__init__.py
suite_names = ['index']

And then in normal config module

.. code:: python
# config.py

EZTEST_CONFIG_MODULE='test/config'
...

Then in flask app module set config from normal config module

.. code:: python
# myapp.py
app = Flask(__name__)
app.from_object('config')


Developing Testcases
--------------------

Test cases are pacakged in modules that expose a 'suite' object in its global scope

Example suite 'index':

.. code:: python

# test/suites/index.py
from flaskeztest import EZTestSuite, EZTestCase

class SimpleTestCase(EZTestCase):

FIXTURE = 'myfixture' # json filename without .json

def runTest(self):
self.assertTrue(...the truth...)

...

suite = EZTestSuite('index', __name__)

Using Fixtures
--------------

Fixtures are a json file representing data to insert into the database before a test case runs

Example:

myfix.json

.. code::

[
{
"model": "User",
"row": {
"name": "Bob",
"email: "bob@example.com"
}
}
]

or with multiple entries for one model:

myfix2.json

.. code::

[
{
"model": "User",
"rows": [
{
"name": "Bob",
"email: "bob@example.com"
},
{
"name": "Alice",
"email: "alice@example.com"
}
]
}
]

The EZTestCase class
--------------------

EZTestCase's setUp method loads the fixture named in its FIXTURE class variable which places the data for each field
into the expected_models dict.

For instance:

.. code:: python

class TC(EZTestCase):

FIXTURE = 'myfix' # Referring to the first fixture above

def runTest(self):

# Passes
self.assertEqual(self.expected_models['User']['name'], 'Bob')

.. code:: python

class TC2(EZTestCase):

FIXTURE = 'myfix2' # Referring to the second fixture above

def runTest(self):

# Passes as well, but would give index error on first test case
self.assertEqual(self.expected_models['User'][1]['name], 'Alice')

Two other useful method of the EZTestCase class is 'get_endpoint' and 'does_field_exist'.

Example:

.. code:: python

class TC(EZTestCase):
FIXTURE = 'myfix2'
def runTest(self):
# Pull up user details for Alice
self.get_endpoint('index.users', user_id=2) # Assume that index.users take user primary key as argument

# Assert we see second user on page
self.assertTrue(self.does_field_exist('User', 'name', 1))

In general most of the methods defined for EZTestCase are useful to include in your own test case classes.
Check out flaskeztest/eztestcase.py for more of them.


Using capybara
--------------

Flaskeztest allows for querying the html returned from pages using capybara.
Check out the docs at https://elliterate.github.io/capybara.py/ for all the useful methods that can be applied to the
the EZTestCase.page object.

Running with a reflected SQL database
-------------------------------------

Sometimes flaskeztest may be used with a flask app that is not using declarative models with sqlalchemy, but instead
want to simply use a database whose schema is contained within the remote database itself.

One can specify the remote database that flask-eztest should reflect by setting the EZTEST_REFLECTED_DB_URI in the test config module

i.e

.. code:: python

# test/config.py
...
EZTEST_REFLECTED_DB_URI = 'mysql://..."

Metadata

Release files for flaskeztest 0.2.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distribution (wheel)

Table of built distributions (wheels) for flaskeztest 0.2.2
File Interpreter ABI Platform Actions
flaskeztest-0.2.2-py2-none-any.whl Python 2 none any Details for flaskeztest-0.2.2-py2-none-any.whl

Release files / flaskeztest-0.2.2-py2-none-any.whl

Download URL flaskeztest-0.2.2-py2-none-any.whl
Size 9.7 kB
Tags Python 2
SHA-256 checksum
How to use checksums
0c3a169c39d80001fc087640f86a902e1854ae9512ba95d6d1257b62d76fdc14
BLAKE2b-256 checksum
How to use checksums
328ae5ff74afb253e9d8d8bda74b3371d01ffc2d2db2e9a9f691a66691ee34e7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/1.12.1 pkginfo/1.4.2 requests/2.19.1 setuptools/40.2.0 requests-toolbelt/0.8.0 tqdm/4.26.0 CPython/2.7.15rc1

Security for flaskeztest 0.2.2

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page