Test Unit help

ruby-talk
1

I’m new to unit testing, and test-first development. Sometimes I
would like to run tests from within other tests rather than recoding
and maintaining a test in two places. Sometimes it helps to pass in
an argument as shown below. In Test::Unit 0.1.4, the first example
test_do_some_work(w=@worker) worked fine, but the second
test_do_more_work(w=@worker) failed to run. Now, in 0.1.8, both fail
to run. I’m using ruby 1.6.8 on Windows. Questions: Is this
behavior by design? Why? Is it a bad idea to do what I am trying?
Thanks,

require 'test/unit’
require ‘work.rb’

class TC_Work < Test::Unit::TestCase
def setup
@worker = Worker.new
end

def test_do_some_work(w=@worker)
s = 'a’
i = 25
assert_equal(‘abcdefghijklmnopqrstuvwxyz’, w.do_some_work(s, i))
end

def test_do_more_work(w=@worker)
s = 'abc’
i = 4
assert_equal(‘zbczbczbczbc’, w.do_more_work(s, i))
end
end

Jim Davis

2

I’m fairly new to unit testing too, but this could solve the problem:

def test_foo(w = nil)
w ||= @worker

w.bar(baz, quz, fubar)
end

I don’t feel OK when I see an instance var. as a default value, not sure
why.

···

On Sat, Mar 01, 2003 at 06:05:43AM +0900, James Davis wrote:

I’m new to unit testing, and test-first development. Sometimes I
would like to run tests from within other tests rather than recoding
and maintaining a test in two places. Sometimes it helps to pass in
an argument as shown below. In Test::Unit 0.1.4, the first example
test_do_some_work(w=@worker) worked fine, but the second
test_do_more_work(w=@worker) failed to run. Now, in 0.1.8, both fail
to run. I’m using ruby 1.6.8 on Windows. Questions: Is this
behavior by design? Why? Is it a bad idea to do what I am trying?

–
_ _

__ __ | | ___ _ __ ___ __ _ _ __
'_ \ / | __/ __| '_ _ \ / ` | ’ \
) | (| | |__ \ | | | | | (| | | | |
.__/ _,|_|/| || ||_,|| |_|
Running Debian GNU/Linux Sid (unstable)
batsman dot geo at yahoo dot com

Actually, typing random strings in the Finder does the equivalent of
filename completion.
– Discussion on file completion vs. the Mac Finder

3

Help is just a simple refactoring away. Since the test_xxx methods are
called by the framework, changing their signatures is a questionable
tactic (although one /would/ think that the default argument would
work).

Basically you just need to break the code into separate functions and
then call those functions from the test methods. For example …

class TC_Work < Test::Unit::TestCase
def setup
@worker = Worker.new
end

def test_do_some_work
check_some_work(@worker)
end

def check_some_work(w)
s = ‘a’
i = 25
assert_equal(‘abcdefghijklmnopqrstuvwxyz’, w.do_some_work(s, i))
end

def test_do_more_work
check_more_work(@worker)
end

def check_more_work(w)
s = ‘abc’
i = 4
assert_equal(‘zbczbczbczbc’, w.do_more_work(s, i))
end
end

···

On Fri, 2003-02-28 at 16:05, James Davis wrote:

I’m new to unit testing, and test-first development. Sometimes I
would like to run tests from within other tests rather than recoding
and maintaining a test in two places. Sometimes it helps to pass in
an argument as shown below. In Test::Unit 0.1.4, the first example
test_do_some_work(w=@worker) worked fine, but the second
test_do_more_work(w=@worker) failed to run. Now, in 0.1.8, both fail
to run. I’m using ruby 1.6.8 on Windows. Questions: Is this
behavior by design? Why? Is it a bad idea to do what I am trying?
Thanks,

require ‘test/unit’
require ‘work.rb’

class TC_Work < Test::Unit::TestCase
def setup
@worker = Worker.new
end

def test_do_some_work(w=@worker)
s = ‘a’
i = 25
assert_equal(‘abcdefghijklmnopqrstuvwxyz’, w.do_some_work(s, i))
end

def test_do_more_work(w=@worker)
s = ‘abc’
i = 4
assert_equal(‘zbczbczbczbc’, w.do_more_work(s, i))
end
end

–
– Jim Weirich jweirich@one.net http://w3.one.net/~jweirich

“Beware of bugs in the above code; I have only proved it correct,
not tried it.” – Donald Knuth (in a memo to Peter van Emde Boas)

4

Help is just a simple refactoring away. Since the test_xxx methods are
called by the framework, changing their signatures is a questionable
tactic (although one /would/ think that the default argument would
work).

Basically you just need to break the code into separate functions and
then call those functions from the test methods. For example …

class TC_Work < Test::Unit::TestCase
def setup
@worker = Worker.new
end

def test_do_some_work
check_some_work(@worker)
end

def check_some_work(w)
s = ‘a’
i = 25
assert_equal(‘abcdefghijklmnopqrstuvwxyz’, w.do_some_work(s, i))
end

end

Yes. Two important realizations here:

  1. Assertions need not be placed in ‘test_xxx’ methods.
  2. Refactor mercilessly!

Thanks,

  • Jim Davis -