How To Write Unit Tests in Python • Pytest Tutorial
HTML-код
- Опубликовано: 27 июн 2024
- Learn how to test your Python code by writing unit tests with the Pytest framework.
In this video I'll be covering how to write and run unit tests, how to check if an Error is raised, how to use test fixtures, and how to mock dependencies.
💻 Code: github.com/pixegami/simple-py...
👉 Pytest Documentation: docs.pytest.org/
👉 Unittest Mock: docs.python.org/3/library/uni...
📚 Chapters
00:00 Introduction
02:45 Install Pytest and setup the project
04:14 Running your first unit test
06:50 Implement your test (with "assert")
11:24 How to test that an Error is raised
18:30 How to run a specific unit test file/method
19:50 How to see print statements in pytest
20:27 Implementing "get_total_price" test
23:10 Using pytest fixtures to set up context for your tests
27:00 When your tests depends on another module
30:50 How to mock dependencies (using unittest Mock)
34:30 Wrap-up
#pytest #python
I just loved the way the entire tutorial was planned and executed. The build up was so apt. Loved every bit of it. Thanks for your time and effort. First time here - subscribed
Thanks! I did put a lot of thought into planning it, so that it covers all the topics with little wasted effort. I really want to respect my viewer's time - so I really appreciate your comment :)
I agree!! this video is best pytest tutorial
Agree 💯
This will rank in top 3 videos of all time in Python Unit Testing on RUclips. Had it been a category called 'Python Unit Test' in Academy Awards. It sure will win the Oscar.
Thank you! Glad you enjoyed it :)
One of the best videos about testing I've ever watched, that's excellent, straight to the point
Thanks! That's my goal with these videos :)
Most underrated channel on RUclips 🙏🏽
Thanks! I appreciate that.
Loved your straight-forward approach and building of functionality as you go. Excellent work. Thanks!
Thanks! I hope it helped!
@@pixegami There is only one thing missing from this video and pretty much everywhere I've looked on the Internet: How to import a function in a directory from a file in another directory at the same depth. This is import for testing packages with the directory structure /src/file with /test/test_file or when one package is accessing another one. A video on that would be terrific!
Very practical and accessible tutorial. I enjoy your concise and clear style. This is content with high sound to noise ratio. Subscribed!
Thanks! I know everyone is busy and want to learn as quickly as possible, so I aim to keep the content lean. Good to know that the effort is not wasted :)
really piqued my interest towards the end. great explanation. would love to see part 2 with more advanced test case scenarios. thanks
Solid tutorial, great explanation. I like the way it was the closest to hands on learning without having to actually be hands on.
Awesome, thank you!
In Such a short time you have covered so many features of Pytest...That's Amazing !!!!
Loved the video and explanation!
Pytest is indeed a powerful test framework !!
Glad you liked it :)
It was a great video it helped me a lot to get started with PYTEST, assert ,fixtures and mock. I just loved it
Thank you so much!! I used to think testing was hard and I was afraid to learn, but now I've changed that!
Great video🤩
Testing seems hard because I think there's not enough education about it! But I'm glad this video helped!
EXCELLENT Tutorial. Love your approach and style to teaching. Very clear, concise and to the point! Thank you for posting!👍
Glad you enjoyed it!
Thanks my friend. I can't thank you enough for the knowledge you just shared in just 35 minutes. I had absolutely 0 experience with Unit tests in general but now I really do. I really loved the way you started small and got a bit deeper. This is what most people just don't do. You probably know about Gall's law "All complex systems work, are built on top of simpler systems that worked" Hence, if you want to build a complex system, first build a simple system that works and build on top of that. Thank you. I really mean it. You just got a new subscriber. Abel
Thank you for the thoughtful comment :) Glad you found the content useful. Stay tuned, I hope to provide more useful content right here on this channel.
This is one of the best videos on pytest I have seen. Thank you for it.
Thank you, I appreciate your comment!
Nicely done. No bullshit background music, and silly-ass fancy graphics and sound effects. Just good technical knowledge, explained well.
Great video. Please make more. You explain things without overcomplicating the process.
Thanks! More to come. Feel free to suggest topics you'd like to see.
Thanks for this tutorial, I'm new to Pytest and this was very helpful.
Excellent video! You have a clear talent for teaching and being able to express your ideas thoughtfully and in a manner that is easy to understand.
I'm excited to be able to subscribe to a new channel that will soon blow up in popularity once more people see the quality of your content! I'll be able to say, _"I was there..."_ 😂
Subbed!
Thanks! I appreciate the support :)
This is by far one of the best tutorial lessons! It's like every time I'm thinking 'but why that' you continue with explaining exactly what I wanted to know!
Really glad to hear that! Thank you :)
Thank you for this video. You've done a really great job of explaining these concepts.
Thanks! I appreciate the comment :)
Best tutorial so far on pytest!
Thanks! Hope it was useful.
Very accessible and clear, thank you for that!
Glad it was helpful!
Very clear and well spoken instructions
Hey this was an EXCELLENT and concise intro to pytest! The explanations and examples were so clear, and went beyond the usual test video toy "x + y" drivel . I will strive to use the testing workflow model you demonstrated.
The almighty YT algo popped this video in after a recent freecodeacademy Pytest intro video - and honestly, this beats that one by far! And it is three times as long! In fact, I'm linking this video in that video's comments! You rock! And you got a new sub!
Thank you for your kind comments :) I try hard to structure the content to be as direct and useful as possible, so I appreciate hearing whenever it helps someone out.
Great Video. I finally understood most about the pytest framework. Thamks friend
Thank you for sharing your talents. It’s well organized, clear, and useful.🙏🏾👨🏽💻
Thank you! Glad you found it clear!
Excelent video. You are a natural speaker. Thx!
Thank you!
very cool Tutorial for pytest. Especially for beginners. Everything is crystal and clear.
Thank you! Glad it was helpful!
Thanks for a great and simple tutorial on PyTest!
Glad it was helpful!
Excellent! Thank you.
helpful video on unit testing! Thank you!
I can't thank you enough for this video🙌
This helped me so much!
Glad it helped!
Extremely well done and clear. Thanks so much for this!
You're very welcome!
Insightful, thank you!
What a great tutorial! Thank you!
Glad it was helpful!
This was a very helpful and clear tutorial, thank you very much!
Glad to hear it!
very nicely explained. Thank you
Thanks! Nice TDD tutorial.
Glad you liked it!
Excellent and comprehensive tutorial, very well explained and accessible. Best ever! Good Job!!
Perfectly explained an important topic!
Glad you liked it!
Wow, you covered everything that confused me as a beginner! Thank you
Glad it was helpful!
I got this video from google and assumed it was a popular channel. How is it only at 22k. Very well explained vs coming to the video with a bunch of testcases written
Very Nice video, Thanks man!
Glad you liked it!
Great, thanks you covered enough for me to get going.
You're welcome! Good luck.
Thank you for this amazing tutorial.
I hope it was helpful!
Great tutorial! Simple and effective. Thanks!
Glad it was helpful!
very informative. Thanks
This channel deserves more subscribers
Thanks! I'm glad you think so :)
Nice job! Great starter for pytest. Thanks
Glad you like it!
You are a good teacher, thank you
Thank you! 😃
A very useful video, thank you!
Glad it was helpful!
awesome! had everything for me to get started. side_effects ftw!
You're welcome!
Huge thanks for the vid.
Glad it helped
Excellent content!!!
Thank you so much, u've made my life easier
Awesome! Glad to hear that!
Great tutorial! Thanks a lot.
You're welcome!
Thank you! U totally domain the subject entirely
Glad it was helpful!
Really useful, thank you!
Glad it was helpful!
Awesome Video...never seen such a nice explaination of things..love it!!
Glad you liked it!
Amazing, I really enjoyed it
Awesome, thank you!
Well structured!
Thank you!
super helpful. Thank you! youtube should rank your video to the 1st position
Thanks!
keep up the good work!
Thanks for the encouragement!
Would love to see more videos on testing, and integration test to verify end to end testing, thanks 🙏🏼
That's a great idea - thanks for the feedback :) I can definitely cover more testing topics.
Yes, please!🙏
I rarely comments but this is a great video on PyTest!
very high quality video
Thanks. I appreciate it!
I hope you take this as a compliment. You are filling in the gap left by Corey Schafer!
Thank you! Corey Schafer is a legend and I'm humbled to be compared to them.
Very useful intro, thanks.
Glad it was helpful!
Thanks very useful :)
Glad it was helpful!
Thanks for your time and effort
awesome video!!!!
Thank you, I appreciate your comment!
Perfect!
I hope it's useful!
Great video. Could you make a part 2 of this? For example: how to work with context across multiple tests, how to work with scopes, how to work with test data for AI, different types of exceptions etc.
Interesting - I hadn't considered a part 2, but I'm keen to hear your use cases for the things you mentioned (particularly about AI testing - do you have a use case in mind for that)?
@@pixegami I worked on a personal project where a pokerbot would receive real images of pokercards and then used a simple CNN to decide which card it was. In order to test the flow I had to import the model, retrieve some test images etc. But I also just wondering in general how to combine unit testing and data science. I really enjoy developing well tested code and these jupyter notebooks are a nightmare to work with when I do anything AI related
good video. You earn a new follower
Thanks and welcome!
wow, helped me a lot
Glad it helped :)
nice information.
Thanks for the feedback. I've reduced the ad settings so now it should be a smoother viewing experience.
You're a great teacher.
Wonder if you have a tutorial on app structure too?
Not yet, but I'll note it down as an idea that my audience wants! Thank you.
Excellent video! it's quite clear and concise. By the way, I am curious what VSCode colorscheme you are using in the video?
Thank you! I use monokai.pro/ - I've tried a lot of different themes, but I always come back to this one.
Really nice intro.
Thanks!
Excellent tutorial, very helpful! Just one request, is there a way you could position your mic so your keyboard is not so loud (particularly the Return-Key)? Maybe use a 'unidirectional' mic and some type of dampener under your keyboard?
Thanks for the feedback! Glad you enjoyed the video. And yes, I agree the keyboard is too loud. Sorry! I've actually upgraded my Mic to a vocal mic, and I'm using a different (softer) keyboard now in my newer videos so I hope it'll feel/sound better for you.
20:07 You can also use "pytest -k test_can_get_total_price"
Ah, thanks for calling out!
Thanks alot
Hello, This is really very good content. It helped me a lot to implement unit test cases. I am trying to implement some retry mechanism (using HTTPadapter , retry package) when the given http errors (500,502,503,504) occur. Now I am not able to implement the retry mechanism as well as the unit testcases for the same. It would be really helpful if you create some video on the same topic.
at 8:20 here you are teasing one function of the instance(add) but also depends on another function of that instance(size). I thought that we are suppose to isolate each test.
great video! what are you using as a terminal ? I love the way it displays the cwd as shadow instead of showing the full path , can you please share the theme name ? thanks !
May I ask what operating system you're using? Great tutorial by the way.
Thank you! In this video I'm using Ubuntu Linux. It's probably my favourite programming environment, although these days I'm using a Mac because I find I also need a lot of creative software (to make the actual videos).
Great explanation and flow in the video:
I find myself sitting with two questions at the end though:
1. Your implementation of the fixture returned one instanciated class. What about if I want to return several values, e.g. environment values, connenction details to databases. How would I return that? Would I return a list of values? Or can I somehow bake it into the function arguments, what information I want to retrieve?
2. I find myself confused on the last part with the Mock module. The way I see it:
1. You instanciate an empty class
2. You call an empty method of this class (none of these fail because both are in fact defined, just have no content)
3. You call Mock to get a value, and you pass in the function to handle the logic.
However, the mock_get_item requires an item argument passed to it, yet I do not see you pass anything. Should mock_get_item not receive a list os something that contains the items added to the cart?
And in the assertion, you pass in item_database, but the way I read the code, the item_database should still be an empty class since the .get method does nothing yet..
Great questions!
1. How do I return a list of values or (anything else) from a fixture? Well, a fixture can return any type of value. So you could have a fixture that returns a tuple. Or you might choose to have a Class or a module that groups together a bunch of different values you want to use together. You are also not limited to using one fixture per test. If you made another fixture called "cart2", then you should be able to use both fixtures in the test by writing: `def test_can_get_total_price(cart, cart2)`. Take a look here for more examples: docs.pytest.org/en/6.2.x/fixture.html
2. You don't see anything being passed in to `mock_get_item` because we never use it directly in the test. We *assign* it to our item_database's "get()" function: "item_database.get = Mock(side_effect=mock_get_item)". Then we call `cart.get_total_price(...)`, which then goes on to call the item_database.get() with our arguments: "total_price += price_map.get(item)":
github.com/pixegami/simple-pytest-tutorial/blob/aa65d4b457039dddaeb0a75a359f223a25ec3d09/shopping_cart.py#L23
So when you say " item_database should still be an empty class since the .get method does nothing" - it's actually not true because we *mocked* the get() method to do something for us with this line. We basically 'replaced' the functionality of `item_database.get()`:
github.com/pixegami/simple-pytest-tutorial/blob/aa65d4b457039dddaeb0a75a359f223a25ec3d09/test_shopping_cart.py#L42
Hope this helps!
Great video. Just learnt pytest. Mind sharing your vs code theme setup?
Thanks! For VSCode I use the paid "Monokai Pro" theme, but they do have a free trail. Other themes I've enjoyed in the past are "Nord" and "Ayu".
What is your keyboard? It sounds great!
In this video I just used a DURAGOD keyboard with blue-key switches. I've since changed now to a Moonlander keyboard with brown switches.
They keyboards don't really make too much difference to me as long as they are the thicker mechanical types.
What's your vscode theme? It 's great!
Hi, great video thanks! Could you explain why you didnt or when to use mock.patch instead of what you did here creating an instance of the db and return a mock.
Great suggestion! `mock.patch` works too, and is probably a good way to do it. I didn't choose to use it here mainly because I wanted to make the relationship between the mock and the dependency clearer. Introducing mock.patch would introduce some new mental concepts (with it being a decorator), and I didn't want that at this stage in the tutorial.
Now, in actual real projects I've used mock.patch before, but it started to become annoying when there's maybe 5-6 things I needed to patch for one test. A lot of nesting, ordinal arguments, and a lot of repetition across various test cases. I am now learning towards fixtures instead, to set up more complex mock behavior.
Which ZSH theme is that ? I would like to install the same. Looks amazing,
I have a video and a GitHub repo covering the Ubuntu ZSH theme I use: github.com/pixegami/terminal-profile
That keyboard sounds good 👌 specs?
Haha I've used a lot of keyboards so I'm not sure which one specifically I used in this video. I think it was probably this one, with the blue Cherry switches: amzn.asia/d/3uXqhmU
Very helpful to a beginner like me,thanks. can you help me on how to integrate zoom in a django project?
Sorry I haven't done any Django projects yet but it's on my list!
Cool@@pixegami
Hi Pixegami,
Python is totally new to me. I am very interested in learning it. Would you mind sending me a link to or show me how to download/install Python to my laptop? Thanks in advance.
Python is a great programming language to learn! I have a complete playlist on a learning roadmap and how to install it and learn all the basics here: ruclips.net/p/PLZJBfja3V3Rsbiz84Z63IXnTQZH_Rnfuo&si=-hMurnRNSf2s7J_m
Wish he did a more advanced pytest course
17:19 what you should actually do, is write another test that checks if you can add 5 items and nothing happens. These two tests would be sufficient.
Great tutorial, but I still dont seem to get how the item database is getting populated. we never passed the cart values to it, so how did it even know ?
Yup! In unit tests, we don't want to use real databases or values, so we "mocK' them. We're saying "pretend I have a database, and when I query for 'x', give me '5'."
This way we can just test that specific unit of the code (usually just a class or a function) without having to worry about the other parts of app.
what are flat directories? how can i make it so i can just directly reference the file like that
Here's a section in the doc on how to specify which tests to run: docs.pytest.org/en/latest/how-to/usage.html#specifying-which-tests-to-run
E.g. for a specific function you can run `pytest tests/test_mod.py::test_func`.
Great video but question, why didnt you use monkeypatch?
Good call. I wanted to start off with more primitive tools, and more generalisable knowledge. Mocking and stubbing is pretty widely used, but I think patching is more prominent in Python testing.
I'm also not a fan of positional arguments for the mocked items, I think fixtures afford more precision and flexibility.
set up a run configuration for pytest but it gave DeprecationWarning: pkg_resources is deprecated as an API. How to fix?
Hmm, not sure where that is coming from. I'd probably recommend just tracing back the error to find where it's coming from, and also just updating all the dependencies so they are all on the latest version.
But if it's just a warning, I don't think it should impact you.