Einheit means unit in German.
Einheit is a Nim unit testing library inspired by Python's unit tests. Nim's unittest library is good, but I wanted something a little more "Nim" feeling. I also really like Python's unittest module and thought it would be nice to have something similar in Nim. Also, unittest doesn't have much documentation on how to use it, and it's pretty bare bones, so I wanted a little more functionality and documentation.
The benefit of the macro style I chose means that you can document your tests nicely as well :)
testSuite is a compile-time macro that allows a user to easily define tests and run them.
Methods are used for inheritance, so if you want to derive a test suite, then you have to make sure the base suite uses methods for the tests that you want to derive.
If you don't want inheritance, you can just use procs.
Two special methods/procs are called setup() and tearDown(). The macro will inject these methods/procs if they don't exist and they will be called before and after running the test suite, respectively.
Test methods/procs to be run are prefixed with "test" in the method/proc name. This is so that you can write tests that call procs that do other things and won't be run as a test.
For each suite method/proc, an implicit variable called "self" is added. This lets you access the testSuite in an OO kind of way.
On failure, the macro gathers names and values of all arguments and functions and prints them out. It's really useful for debugging!
Install with nimble!
nimble install einheitnim c -r testing_script.nim
# Or no colors
nim c -r -d:noColors testing_script.nim
# No colors, quiet output
nim c -r -d:quiet -d:noColors testing_script.nim
# With Node.js as a target
nim js -d:nodejs -r testing_script.nim
import einheit
testSuite SuiteName of TestSuite:
var
suiteVar: string
testObj: int
method setup()=
## do setup code here
self.suiteVar = "Testing"
self.testObj = 90
method tearDown()=
## do tear down code here
self.suiteVar = nil
self.testObj = 0
method testAddingString()=
## adds a string to the suiteVar
self.suiteVar &= " 123"
self.check(self.suiteVar == "Testing 123")
proc raisesOs()=
# This proc won't be invoked as a test, it must begin with "test" in lowercase
raise newException(OSError, "Oh no! OS malfunction!")
method testRaises()=
# Two ways of checking
self.checkRaises OSError:
self.raisesOs()
self.checkRaises(OSError, self.raisesOs())
method testTestObj()=
self.check(self.testObj == 90)
method testMoreMore()=
self.check("String" == "String")
when isMainModule:
einheit.runTests()You can also find examples in the test.nim file, including inheritance.
Output of running
nim c -r test.nimis this:
[Running] UnitTests -----------------------------------------------------------
[OK] testForB
[Failed] testArrayAssert
Condition: check(self.testArray == [0, 1, 2])
Where:
self.testArray -> [0, 1, 2, 3]
Location: test.nim; line 27
[Failed] testForC
Condition: check(c == 1)
Where:
c -> 0
Location: test.nim; line 32
[1/3] tests passed for UnitTests. ----------------------------------------------
[Running] UnitTestsNew --------------------------------------------------------
[OK] testTestObj
[OK] testStuff
[Failed] testMore
Condition: check(more == 1)
Where:
more -> 23
Location: test.nim; line 56
[Failed] testMoreMore
Condition: check(self.returnTrue())
Where:
self.returnTrue() -> false
self -> ref UnitTestsNew(testObj: 90, name: UnitTestsNew, currentTestName: testMoreMore, testsPassed: 2, numTests: 4)
Location: test.nim; line 59
[2/4] tests passed for UnitTestsNew. -------------------------------------------
[Running] TestInherit ---------------------------------------------------------
[OK] testTestObj
[OK] testStuff
[Failed] testMore
Condition: check(more == 1)
Where:
more -> 23
Location: test.nim; line 56
[Failed] testMoreMore
Condition: check(self.returnTrue())
Where:
self.returnTrue() -> false
self -> ref UnitTestsNew(testObj: 90, name: TestInherit, currentTestName: testMoreMore, testsPassed: 2, numTests: 4)
Location: test.nim; line 59
[Failed] testRaises
Condition: checkRaises(OSError, self.raisesOs())
Where:
self.raisesOs() -> SystemError
Location: test.nim; line 72
[2/5] tests passed for TestInherit. --------------------------------------------
[Running] MoreInheritance -----------------------------------------------------
[Failed] testTestObj
Condition: check(self.testObj == 90)
Where:
self.testObj -> 12345
Location: test.nim; line 46
[OK] testStuff
[Failed] testMore
Condition: check(more == 1)
Where:
more -> 23
Location: test.nim; line 56
[Failed] testMoreMore
Condition: check(self.returnTrue())
Where:
self.returnTrue() -> false
self -> ref UnitTestsNew(testObj: 12345, name: MoreInheritance, currentTestName: testMoreMore, testsPassed: 1, numTests: 4)
Location: test.nim; line 59
[Failed] testRaises
Condition: checkRaises(OSError, self.raisesOs())
Where:
self.raisesOs() -> SystemError
Location: test.nim; line 72
[OK] testTestObj
[OK] testNewObj
[Failed] testRefObject
Condition: check(d == k)
Where:
k -> ref TestObj(t: 30)
d -> ref TestObj(t: 3)
Location: test.nim; line 123
[Failed] testObject
Condition: check(d == k)
Where:
k -> TestObj(t: 30)
d -> TestObj(t: 3)
Location: test.nim; line 138
[Failed] testComplexObject
Condition: check(x.isObj(p))
Where:
x.isObj(p) -> false
p -> 4
x -> Obj2(d: Obj1(e: Hey))
Location: test.nim; line 150
[Failed] testTuple
Condition: check(t == r)
Where:
r -> tuple Person(name: P, age: 3)
t -> tuple Person(name: Peter, age: 30)
Location: test.nim; line 161
[Failed] testComplex
Condition: check(self.doStuff(a, s) == "5stuff" and self.doStuff(a, self.doStuff(a, self.doStuff(y, s))) == "something?")
Where:
self.doStuff(a, s) -> 5stuff
a -> 5
self -> ref MoreInheritance(testObj: 12345, name: MoreInheritance, currentTestName: testComplex, testsPassed: 3, numTests: 12)
self.doStuff(a, self.doStuff(a, self.doStuff(y, s))) -> 5545stuff
y -> 45
s -> stuff
self.doStuff(y, s) -> 45stuff
self.doStuff(a, self.doStuff(y, s)) -> 545stuff
Location: test.nim; line 169
[3/12] tests passed for MoreInheritance. ---------------------------------------
[Summary]
[8/24] tests passed.
Notice that on failure, the test runner gives some useful information about the test in question. This is useful for determining why the test failed.