Develop Reference

Why do I only get a function as a return value by using a fixture (from pytest) in a test script?

I want to write test functions for my code and decided to use pytest. I had a look into this tutorial: https://semaphoreci.com/community/tutorials/testing-python-applications-with-pytest
My real code involves another script, written by me, so I made an example, which also creates the same problem, but does not rely on my other code.
#pytest.fixture()
def example():
value = 10
return value
def test_value(example):
print(example)
assert(example == 10)
test_value(example)
When I run my script with this toy example, the print returns a function:
<function example at 0x0391E540>
and the assertion fails.
If I try to call example() with the parenthesis, I get this:
Failed: Fixture "example_chunks" called directly. Fixtures are not meant to be called directly,
but are created automatically when test functions request them as parameters.
See https://docs.pytest.org/en/stable/fixture.html for more information about fixtures, and
https://docs.pytest.org/en/stable/deprecations.html#calling-fixtures-directly about how to update your code.
I am sure, I am missing something important here, but searching google did not help me, which is why I hope somebody here can provide some assistance.

Remove this line from your script
test_value(example)
Run your script file with pytest file.py
Fixtures will be automatically resolved by pytest
In your example you run code directly and fixtures are just simple functions

Py.test - Applying variables to decorators from a csv?

Please bear with me while I try to explain my predicament, I'm still a Python novice and so my terminology may not be correct. Also I'm sorry for the inevitable long-windedness of this post, but I'll try to expalin in as much relevant detail as possible.
A quick rundown:
I'm currently developing a suite of Selenium tests for a set of websites that are essentially the same in functionality, using py.test
Tests results are uploaded to TestRail, using the pytest plugin pytest-testrail.
Tests are tagged with the decorator #pytestrail.case(id) with a unique case ID
A typical test of mine looks like this:
#pytestrail.case('C100123') # associates the function with the relevant TR case
#pytest.mark.usefixtures()
def test_login():
# test code goes here
As I mentioned before, I'm aiming to create one set of code that handles a number of our websites with (virtually) identical functionality, so a hardcoded decorator in the example above won't work.
I tried a data driven approach with a csv and a list of the tests and their case IDs in TestRail.
Example:
website1.csv:
Case ID | Test name
C100123 | test_login
website2.csv:
Case ID | Test name
C222123 | test_login
The code I wrote would use the inspect module to find the name of the test running, find the relevant test ID and put that into a variable called test_id:
import csv
import inspect
class trp(object):
def __init__(self):
pass
with open(testcsv) as f: # testcsv could be website1.csv or website2.csv
reader = csv.reader(f)
next(reader) # skip header
tests = [r for r in reader]
def gettestcase(self):
self.current_test = inspect.stack()[3][3]
for row in trp.tests:
if self.current_test == row[2]:
self.test_id = (row[0])
print(self.test_id)
return self.test_id, self.current_test
def gettestid(self):
self.gettestcase()
The idea was that the decorator would change dynamically based on the csv that I was using at the time.
#pytestrail.case(test_id) # now a variable
#pytest.mark.usefixtures()
def test_login():
trp.gettestid()
# test code goes here
So if I ran test_login for website1, the decorator would look like:
#pytestrail.case('C100123')
and if I ran test_login for website2 the decorator would be:
#pytestrail.case('C222123')
I felt mighty proud of coming up with this solution by myself and tried it out...it didn't work. While the code does work by itself, I would get an exception because test_id is undefined (I understand why - gettestcase is executed after the decorator, so of course it would crash.
The only other way I can handle this is to apply the csv and testIDs before any test code is executed. My question is - how would I know how to associate the tests with their test IDs? What would an elegant, minimal solution to this be?
Sorry for the long winded question. I'll be watching closely to answer any questions if you need more explanation.

pytest is very good at doing all kind of metaprogramming stuff for the tests. If I understand your question correctly, the code below will do the dynamic tests marking with pytestrail.case marker. In the project root dir, create a file named conftest.py and place this code in it:
import csv
from pytest_testrail.plugin import pytestrail
with open('website1.csv') as f:
reader = csv.reader(f)
next(reader)
tests = [r for r in reader]
def pytest_collection_modifyitems(items):
for item in items:
for testid, testname in tests:
if item.name == testname:
item.add_marker(pytestrail.case(testid))
Now you don't need to mark the test with #pytestrail.case()at all - just write the rest of code and pytest will take care of the marking:
def test_login():
assert True
When pytest starts, the code above will read website1.csv and store the test IDs and names just as you did in your code. Before the test run starts, pytest_collection_modifyitems hook will execute, analyzing the collected tests - if a test has the same name as in csv file, pytest will add the pytestrail.case marker with the test ID to it.

I believe the reason this isn't working as you would expect has to do with how python reads and executes files. When python starts executing it reads in the linked python file(s) and executes each line one-by-one, in turn. For things at the 'root' indention level (function/class definitions, decorators, variable assignments, etc) this means they get run exactly one time as they are loaded in, and never again. In your case, the python interpreter reads in the pytest-testrail decorator, then the pytest decorator, and finally the function definition, executing each one once, ever.
(Side note, this is why you should never use mutable objects as function argument defaults: Common Gotchas)
Given that you want to first deduce the current test name, then associate that with a test case ID, and finally use that ID with the decorator, I'm not sure that is possible with pytest-testrail's current functionality. At least, not possible without some esoteric and difficult to debug/maintain hack using descriptors or the like.
I think you realistically have one option: use a different TestRail client and update your pytest structure to use the new client. Two python clients I can recommend are testrail-python and TRAW (TestRail Api Wrapper)(*)
It will take more work on your part to create the fixtures for starting a run, updating results, and closing the run, but I think in the end you will have a more portable suite of tests and better results reporting.
(*) full disclosure: I am the creator/maintainer of TRAW, and also made significant contributions to testrail-python

Python unittest, running only the tests that failed

I have a large python test file using unittest that I run from the command line. Some tests take a while to run. This is a mild pain point because I'm often only concerned with the last test I added. What I want is this:
add test.
run tests (one fails because I haven't written the code to make it pass)
implement the behaviour
run only the test that failed last time
fix the silly error I made when implementing the code
run only the failing test, which passes this time
run all the tests to find out what I broke.
Is it possible to do this from the command line?

(Not a fully automated solution, but better than the existing one)
If you pass the name of a test class as an argument to the test script, only that test will be run. For example, if you only want to run tests in the MyTest class in the script test_whatever.py:
python3 test_whatever.py MyTest
You can also specify an individual test as a member of that class. For example, suppose you want to run the test test_something in the class MyTest:
python3 test_whatever.py MyTest.test_something

Every test function is declared like:
def test_something_something(self):
If you add an underscore in front, like:
def _test_something_something(self):
that test will be ignored. One thing you can do is to do a quick find and replace in vim. Find all "test_"s and replace them with "_test_" and then find the one test that failed and remove the underscore.

Just run the test with --last-failed option (you might need pytest)

Calling python functions without running from the editor

Please excuse what I know is an incredibly basic question that I have nevertheless been unable to resolve on my own.
I'm trying to switch over my data analysis from Matlab to Python, and I'm struggling with something very basic: in Matlab, I write a function in the editor, and to use that function I simply call it from the command line, or within other functions. The function that I compose in the matlab editor is given a name at the function definition line, and it's generally best for the function name to match the .m file name to avoid confusion.
I don't understand how functions differ in Python, because I have not been successful translating the same approach there.
For instance, if I write a function in the Python editor (I'm using Python 2.7 and Spyder), simply saving the .py file and calling it by its name from the Python terminal does not work. I get a "function not defined" error. However, if I execute the function within Spyder's editor (using the "run file" button), not only does the code execute properly, from that point on the function is also call-able directly from the terminal.
So...what am I doing wrong? I fully appreciate that using Python isn't going to be identical to Matlab in every way, but it seems that what I'm trying to do isn't unreasonable. I simply want to be able to write functions and call them from the python command line, without having to run each and every one through the editor first. I'm sure my mistake here must be very simple, yet doing quite a lot of reading online hasn't led me to an answer.
Thanks for any information!

If you want to use functions defined in a particular file in Python you need to "import" that file first. This is similar to running the code in that file. Matlab doesn't require you to do this because it searches for files with a matching name and automagically reads in the code for you.
For example,
myFunction.py is a file containing
def myAdd(a, b):
return a + b
In order to access this function from the Python command line or another file I would type
from myFunction import myAdd
And then during this session I can type
myAdd(1, 2)
There are a couple of ways of using import, see here.

You need to a check for __main__ to your python script
def myFunction():
pass
if __name__ == "__main__":
myFunction()
then you can run your script from terminal like this
python myscript.py
Also if your function is in another file you need to import it
from myFunctions import myFunction
myFunction()

Python doesn't have MATLAB's "one function per file" limitation. You can have as many functions as you want in a given file, and all of them can be accessed from the command line or from other functions.
Python also doesn't follow MATLAB's practice of always automatically making every function it can find usable all the time, which tends to lead to function name collisions (two functions with the same name).
Instead, Python uses the concept of a "module". A module is just a file (your .py file). That file can have zero or more functions, zero or more variables, and zero or more classes. When you want to use something from that file, you just import it.
So say you have a file 'mystuff.py':
X = 1
Y = 2
def myfunc1(a, b):
do_something
def myfunc2(c, d):
do_something
And you want to use it, you can just type import mystuff. You can then access any of the variables or functions in mystuff. To call myfunc2, you can just do mystuff.myfunc2(z, w).
What basically happens is that when you type import mystuff, it just executes the code in the file, and makes all the variables that result available from mystuff.<varname>, where <varname> is the name of the variable. Unlike in MATLAB, Python functions are treated like any other variable, so they can be accessed just like any other variable. The same is true with classes.
There are other ways to import, too, such as from mystuff import myfunc.

You run python programs by running them with
python program.py

Python failure injection

Is there a neat way to inject failures in a Python script? I'd like to avoid sprinkling the source code with stuff like:
failure_ABC = True
failure_XYZ = True
def inject_failure_ABC():
raise Exception('ha! a fake error')
def inject_failure_XYZ():
# delete some critical file
pass
# some real code
if failure_ABC:
inject_failure_ABC()
# some more real code
if failure_XYZ:
inject_failure_XYZ()
# even more real code
Edit:
I have the following idea: insert "failure points" as specially-crafted comments. The write a simple parser that will be called before the Python interpreter, and will produce the actual instrumented Python script with the actual failure code. E.g:
#!/usr/bin/parser_script_producing_actual_code_and_calls python
# some real code
# FAIL_123
if foo():
# FAIL_ABC
execute_some_real_code()
else:
# FAIL_XYZ
execute_some_other_real_code()
Anything starting with FAIL_ is considered as a failure point by the script, and depending on a configuration file the failure is enabled/disabled. What do you think?

You could use mocking libraries, for example unittest.mock, there also exist many third party ones as well. You can then mock some object used by your code such that it throws your exception or behaves in whatever way you want it to.

When testing error handling, the best approach is to isolate the code that can throw errors in a new method which you can override in a test:
class ToTest:
def foo(...):
try:
self.bar() # We want to test the error handling in foo()
except:
....
def bar(self):
... production code ...
In your test case, you can extend ToTest and override bar() with code that throws the exceptions that you want to test.
EDIT You should really consider splitting large methods into smaller ones. It will make the code easier to test, to understand and to maintain. Have a look at Test Driven Development for some ideas how to change your development process.
Regarding your idea to use "Failure Comments". This looks like a good solution. There is one small problem: You will have to write your own Python parser because Python doesn't keep comments when it produces bytecode.
So you can either spend a couple of weeks to write this or a couple of weeks to make your code easier to test.
There is one difference, though: If you don't go all the way, the parser will be useless. Also, the time spent won't have improved one bit of your code. Most of the effort will go into the parser and tools. So after all that time, you will still have to improve the code, add failure comments and write the tests.
With refactoring the code, you can stop whenever you want but the time spent so far will be meaningful and not wasted. Your code will start to get better with the first change you make and it will keep improving.
Writing a complex tool takes time and it will have it's own bugs which need to fix or work around. None of this will improve your situation in the short term and you don't have a guarantee that it will improve the long term.

If you only want to stop your code at some point, and fall back to interactive interpreter, one can use:
assert 1==0
But this only works if you do not run python with -O
Edit
Actually, my first answer was to quick, without really understanding what you want to do, sorry.
Maybe your code becomes already more readable if you do parameterization through parameters, not through variable/function suffices. Something like
failure = {"ABC": False, "XYZ":False}
#Do something, maybe set failure
def inject_failure(failure):
if not any(failure.values()):
return
if failure["ABC"]:
raise Exception('ha! a fake error')
elif failure["XYZ"]:
# delete some critical file
pass
inject_failure(failure)