Package 'mockery'

Expectation: does the given call match the expected?

Description

Together with mock can be used to verify whether the call expression (expect_call) and/or argument values (expect_args) match the expected.

Usage

expect_call(mock_object, n, expected_call)

expect_args(mock_object, n, ...)

expect_called(mock_object, n)

Arguments

mock_object	A mock object.
n	Call number or total number of calls.
expected_call	Expected call expression; will be compared unevaluated.
...	Arguments as passed in a call.

Details

With expect_called you can check how many times has the mock object been called.

Examples

library(testthat)

# expect call expression (signature)
m <- mock()
with_mock(summary = m, summary(iris))

# it has been called once
expect_called(m, 1)

# the first (and only) call's arguments matches summary(iris)
expect_call(m, 1, summary(iris))

# expect argument value
m <- mock()
a <- iris
with_mock(summary = m, summary(object = a))
expect_args(m, 1, object = a)
# is an equivalent to ...
expect_equal(mock_args(m)[[1]], list(object = a))

---

Create and query a mocked function.

Description

Mock object's primary use is to record calls that are made on the mocked function.

Usage

mock(..., cycle = FALSE, envir = parent.frame())

mock_args(m)

mock_calls(m)

## S3 method for class 'mock'
length(x)

Arguments

...	Values returned upon subsequent calls.
cycle	Whether to cycle over the return values. If FALSE, will fail if called too many times.
envir	Where to evaluate the expressions being returned.
m	A mocked function.
x	A mocked function.

Details

Optionally values/expressions can be passed via ... for the mock object to return them upon subsequent calls. Expressions are evaluated in environment envir before being returned. If no value is passed in ... then NULL is returned.

Passing an expression or a function call via ... is also a way to implement side effects: keep track of the state of code under testing, throw an exception when a condition is met, etc.

mock_calls and mock_args can be used to access the list of calls made on a mocked function and a respective list of values of arguments passed to each of these calls.

Value

mock() returns a mocked function which can be then used with with_mock.

mock_args() returns a list of lists of argument values.

mock_calls() returns a list of calls.

length.mock() returns the number of calls invoked on m.

Examples

library(testthat)

m <- mock(1)
with_mock(summary = m, {
  expect_equal(summary(iris), 1)
  expect_called(m, 1)
  expect_call(m, 1, summary(iris))
  expect_args(m, 1, iris)
})

# multiple return values
m <- mock(1, "a", sqrt(3))
with_mock(summary = m, {
  expect_equal(summary(iris), 1)
  expect_equal(summary(iris), "a")
  expect_equal(summary(iris), 1.73, tolerance = .01)
})

# side effects
m <- mock(1, 2, stop("error"))
with_mock(summary = m, {
  expect_equal(summary(iris), 1)
  expect_equal(summary(iris), 2)
  expect_error(summary(iris), "error")
})

# accessing call expressions
m <- mock()
m(x = 1)
m(y = 2)
expect_equal(length(m), 2)
calls <- mock_calls(m)
expect_equal(calls[[1]], quote(m(x = 1)))
expect_equal(calls[[2]], quote(m(y = 2)))

# accessing values of arguments
m <- mock()
m(x = 1)
m(y = 2)
expect_equal(length(m), 2)
args <- mock_args(m)
expect_equal(args[[1]], list(x = 1))
expect_equal(args[[2]], list(y = 2))

---

R package to make mocking easier

Description

There are great tools for unit testing in R out there already but they don't come with a lot of support for mock objects. This package aims at fixing that.

Examples

library(mockery)

m <- mock(TRUE, FALSE, TRUE)

# this will make summary call our mock function rather then
# UseMethod; thus, summary() will return values as above
stub(summary, 'UseMethod', m)

summary(iris) # returns TRUE
summary(cars) # returns FALSE
summary(co2)  # returns TRUE

## Not run: 
library(testthat)

m <- mock(TRUE)
f <- function() read.csv('data.csv')

with_mock(read.csv = m, {
  f()
  expect_call(m, 1, read.csv('data.csv'))
})

## End(Not run)

---

Replace a function with a stub.

Description

The result of calling stub is that, when where is invoked and when it internally makes a call to what, how is going to be called instead.

Usage

stub(where, what, how, depth = 1)

Arguments

where	Function to be called that will in turn call what.
what	Name of the function you want to stub out (a character string).
how	Replacement function (also a mock function) or a return value for which a function will be created automatically.
depth	Specifies the depth to which the function should be stubbed

Details

This is much more limited in scope in comparison to with_mock which effectively replaces what everywhere. In other words, when using with_mock and regardless of the number of intermediate calls, how is always called instead of what. However, using this API, the replacement takes place only for a single function where and only for calls originating in that function.

Examples

f <- function() TRUE
g <- function() f()
stub(g, 'f', FALSE)

# now g() returns FALSE because f() has been stubbed out
g()

# you can stub multiple functions by calling stub() multiple times 
f <- function() TRUE
g <- function() TRUE
h <- function() any(f(), g())
stub(h, 'f', FALSE)
stub(h, 'g', FALSE)

# now h() returns FALSE because both f() and g() have been stubbed out
h()

---