What you're describing is a very limited subset of testing, which presumably is fine for the projects you work on, but that experience does not generalise well.
Integration testing is of course useful, but generally one would want to create unit tests for every part of the code, and by definition it's not a unit test if hits multiple parts of the code simultaneously.
Apart from that, databases and file access may be fast but they still take resources and time to spin up; beyond a certain project and team size, it's far cheaper to mock those things. With a mock you can also easily simulate failure cases, bad data, etc. - how do you test for file access issues, or the database server being offline?
Using mocks properly is a sign of a well-factored codebase.
The common pitfall with this style of testing is that you end up testing implementation details and couple your tests to your code and not the interfaces at the boundaries of your code.
I prefer the boundary between unit and integration tests to be the process itself. Meaning, if I have a dependency outside the main process (eg database, HTTP API etc) then it warrants an integration test where i mock this dependency somehow. Otherwise, unit tests test the interfaces with as much coverage of actual code execution as possible. In unit tests, out of process dependencies are swapped with a fake implementation like an in-memory store instead of a full of fledged one that covers only part of interface that i use. This results in much more robust tests that I can rely on during refactoring as opposed to “every method or function is a unit, so unit tests should test these individual methods”.
You seem to say it's not worth writing a lot of tests, but then you talk about tests breaking due to bad changes - if you don't write those tests in the first place, then how do you get into that situation?
I didn't word my earlier comment very well - I don't mean to advocate for 100% coverage, which I personally think is a waste of time at best, and a false comfort at worst. Is this what you're talking about? What I wanted to say is that unit tests should be written for every part of the code that you're testing, i.e. break it into bits rather than test the whole thing in one lump, or better, do both - unit tests and integration tests.
This is what I mean about a well-factored code base; separate things are separate, and can be tested independently.
> Unit tests have proven more harmful than helpful
Why? I find them very useful. If I change a method and inadvertently break it somehow, I have a battery of tests testing that method, and some of them fail. Nothing else fails because I didn't change anything else, so I don't need to dive into working out why those other tests are failing. It's separation of concerns.
i have concluded a unit needs to be large. Not a single class or function but a large collection of them. When 'archtecture astronaughts' draw their boxes they ar drawing units. Often thousands of functions belong to a unit. even then though often it is easier to use the real other unit than a test double.
If your unit is thousands of functions wide then you have a monolith, and there are widely discussed reasons why we try to avoid those.
Well-factored codebase doesn’t need mocks.
now I do take care to avoid writing tests that depend on other code's results that would be likely to change. This hasn't proven to be the problem I've so often been warned about though.
...or, you tested each thing individually, so only related tests break and you can quickly zero in on the problem. Isn't that easier?
Even funnier, this was all hypothetical and yet taken as gospel. We hadn't even written the tests yet, so it was impossible to say whether they were slow or not. Nothing had been measured, no performance budget had been defined, no prototype of the supposedly slow tests had been written to demonstrate the point.
We ended up writing - no joke - less than 100 tests total, almost all of which hit the database, including some full integration tests, and the entire test suite finished in a few seconds.
I'm all for building in a way that respects performance as an engineering value, but we got lost somewhere along the way.
Why fake it when an integration test tests the real thing.
I’ve seen what you clearly have. Mocked ResultSets, mocked JDBC templates. “When you get SQL, it should be this string. These parameters should be set. Blah blah.”
It’s so much work. And it’s useless. Where does that SQL to check come from? Copy and paste, so it won’t catch a syntax error.
Test data is faked in each result. You can’t test foreign keys.
Just a bad idea. You’re so right. I find it odd some people are so anti-mock. Yeah it gets abused but that’s not the tool’s fault.
But DB calls are not a good spot to mock out.
Hopefully one day you'll back at that, and realise what an immature attitude that was.
Unless that whole paragraph was a figure of speech, in which case what I said doesn't apply, and we can both go about our days.
More broadly, suppose foo() has an implementation that depends on Bar, but Bar is complicated to instantiate because it needs to know about 5 external services. Fortunately foo() only depends on a narrow sliver of Bar's functionality. Why not wrap Bar in a narrow interface—only the bits foo() depends on—and fake it?
I'm not a maximalist about test doubles. I prefer to factor out my I/O until it's high-level enough that it doesn't need unit tests. But that's not always an option, and I'd rather be flexible and use a test double than burden all my unit tests with the full weight of their production dependencies.
you could extend this to say 85% of the tome just write the code directly to prod and dont have any tests. if you broke something, an alarm will go off