ae 1.8.0 → 1.8.1
Sign up to get free protection for your applications and to get access to all the features.
- data/.ruby +2 -2
- data/HISTORY.rdoc +9 -0
- data/QED.rdoc +736 -0
- data/lib/ae.yml +2 -2
- data/lib/ae/ansi.rb +26 -0
- data/qed/08_check.rdoc +63 -0
- metadata +11 -7
data/.ruby
CHANGED
@@ -40,9 +40,9 @@ revision: 0
|
|
40
40
|
created: '2008-08-17'
|
41
41
|
summary: Assertive Expressive
|
42
42
|
title: AE
|
43
|
-
version: 1.8.
|
43
|
+
version: 1.8.1
|
44
44
|
name: ae
|
45
45
|
description: ! "Assertive Expressive is an assertions library specifically designed
|
46
46
|
\nfor reuse by other test frameworks."
|
47
47
|
organization: Rubyworks
|
48
|
-
date: '2011-11-
|
48
|
+
date: '2011-11-04'
|
data/HISTORY.rdoc
CHANGED
@@ -1,5 +1,14 @@
|
|
1
1
|
= RELEASE HISTORY
|
2
2
|
|
3
|
+
== 1.8.1 / 2011-12-04
|
4
|
+
|
5
|
+
Fixed missing ae/ansi.rb file from distribution.
|
6
|
+
|
7
|
+
Changes:
|
8
|
+
|
9
|
+
* Update manifest, missing ae/ansi.rb
|
10
|
+
|
11
|
+
|
3
12
|
== 1.8.0 / 2011-12-03 / Checkered Flag
|
4
13
|
|
5
14
|
This new release improves support for Proc-based assertions and
|
data/QED.rdoc
ADDED
@@ -0,0 +1,736 @@
|
|
1
|
+
= Introduction
|
2
|
+
|
3
|
+
AE is an assertions framework for Ruby. It's designed
|
4
|
+
around the concept of an Assertor. The Assertor is an
|
5
|
+
Assertion Functor, or Higher-Order Function, which
|
6
|
+
reroutes method calls while monitoring them for failing
|
7
|
+
conditions.
|
8
|
+
|
9
|
+
|
10
|
+
== What AE Provides
|
11
|
+
|
12
|
+
Requiring the AE library.
|
13
|
+
|
14
|
+
require 'ae'
|
15
|
+
|
16
|
+
Loads two classes, +Assertion+ and +Assertor+, the Kernel
|
17
|
+
method +assert+ and it's antonyms +assert!+ and +refute+
|
18
|
+
and a set of core extensions that make writing certain types
|
19
|
+
of assertions easier.
|
20
|
+
|
21
|
+
|
22
|
+
== Assertion and Assertor Classes
|
23
|
+
|
24
|
+
The +Assertion+ class is at the heart of AE. All other AE
|
25
|
+
methods depend on it. The +Assertion+ class is a subclass
|
26
|
+
of Exception. When an assertion is made and fails, it is
|
27
|
+
an instance of Assertion that is raised.
|
28
|
+
|
29
|
+
expect Assertion do
|
30
|
+
msg = "my failure message"
|
31
|
+
assert false, msg
|
32
|
+
end
|
33
|
+
|
34
|
+
Like any raised exception, the last Assertion message is available
|
35
|
+
via <tt>$!</tt>.
|
36
|
+
|
37
|
+
(FYI, in Test::Unit the equivalent class was called +AssertionFailedError+.)
|
38
|
+
|
39
|
+
Assertions themselves are not generally used in creating tests or
|
40
|
+
behavior specifications. Rather they are used to create additional
|
41
|
+
types of assertion methods.
|
42
|
+
|
43
|
+
As mentioned above the +Assertor+ class is a type of Higher-Order
|
44
|
+
function, or Functor, which intercedes with a normal message
|
45
|
+
invocation to monitor for failed conditions, upon which is raises
|
46
|
+
Assertion exceptions.
|
47
|
+
|
48
|
+
|
49
|
+
== Assertion Methods
|
50
|
+
|
51
|
+
The three methods, +assert+, <tt>assert!</tt> and +refute+ all
|
52
|
+
return an Assertor instance when used fluidly, i.e. magic-dot
|
53
|
+
notation, higher-order notation, functor notation, whatever you
|
54
|
+
prefer to call it.
|
55
|
+
|
56
|
+
assert(AE::Assertor === assert)
|
57
|
+
|
58
|
+
Through the use of +method_missing+, the Assertor allows us to write
|
59
|
+
statements like:
|
60
|
+
|
61
|
+
1.assert == 1
|
62
|
+
|
63
|
+
If the operation evaluates to false or nil, then an Assertion error
|
64
|
+
is raised.
|
65
|
+
|
66
|
+
expect Assertion do
|
67
|
+
1.assert == 2
|
68
|
+
end
|
69
|
+
|
70
|
+
The methods <tt>assert!</tt> and +refute+ are just like +assert+
|
71
|
+
expect they purport the negative condition. Patterned after Ruby's
|
72
|
+
own use of "<tt>!</tt>" as meaning +not+, <tt>assert!</tt> should be
|
73
|
+
read "assert not". While +refute+ exists for the sake of those who
|
74
|
+
find the use of a bang method for this purpose unsuited to them.
|
75
|
+
|
76
|
+
|
77
|
+
== How It Works
|
78
|
+
|
79
|
+
An Assertor essentially sits in wait for a method call (via
|
80
|
+
method_missing). When that happens it applies the method to the
|
81
|
+
original receiver, but wrapped in a clause that raises an
|
82
|
+
Assertion should the statement fail. If we wanted to be
|
83
|
+
pedantic, we could write our assertions like:
|
84
|
+
|
85
|
+
raise Assertion.new("1 != 1") unless 1 == 1
|
86
|
+
|
87
|
+
Instead of
|
88
|
+
|
89
|
+
1.assert == 1
|
90
|
+
|
91
|
+
Obviously using Assertor methods are whole lot more concise.
|
92
|
+
|
93
|
+
|
94
|
+
= Assertion Class
|
95
|
+
|
96
|
+
The Assertion class is a subclass of Exception and is the error raised when
|
97
|
+
and assertion fails.
|
98
|
+
|
99
|
+
= Assert Method
|
100
|
+
|
101
|
+
== Compatible with Test::Unit
|
102
|
+
|
103
|
+
The +assert+ method is designed to be backward compatible
|
104
|
+
with the same method in <tt>Test::Unit</tt>.
|
105
|
+
|
106
|
+
Using an argument, +assert+ will check that an argument evaluates
|
107
|
+
to true. Optionally one can send along a meaningful message should
|
108
|
+
the assertion fail.
|
109
|
+
|
110
|
+
assert(true, "Not true!")
|
111
|
+
|
112
|
+
expect Assertion do
|
113
|
+
assert(false, "Not true!")
|
114
|
+
end
|
115
|
+
|
116
|
+
|
117
|
+
== Assert with a Block
|
118
|
+
|
119
|
+
In addition +assert+ has been extended to accept a block. Like the case of the
|
120
|
+
argument, the block is expected to return something that evaluates as true.
|
121
|
+
|
122
|
+
assert do
|
123
|
+
true
|
124
|
+
end
|
125
|
+
|
126
|
+
Assertion.assert.raised? do
|
127
|
+
assert do
|
128
|
+
false
|
129
|
+
end
|
130
|
+
end
|
131
|
+
|
132
|
+
We should also mention that, while probably not very useful, since
|
133
|
+
the arity of a block can be checked, one can also pass the receiver
|
134
|
+
into the block as a block argument.
|
135
|
+
|
136
|
+
"hi".assert do |s|
|
137
|
+
/h/ =~ s
|
138
|
+
end
|
139
|
+
|
140
|
+
|
141
|
+
== Antonyms for Assert
|
142
|
+
|
143
|
+
We can state the opposite assertion using <tt>assert!</tt>.
|
144
|
+
|
145
|
+
10.assert! == 9
|
146
|
+
|
147
|
+
Or, because some people do not like the use of a bang method, +refute+.
|
148
|
+
|
149
|
+
10.refute == 9
|
150
|
+
|
151
|
+
These terms can be used just as +assert+ is used in all examples,
|
152
|
+
but with the opposite inference.
|
153
|
+
|
154
|
+
Another way to get the opposite inference, is to use +not+.
|
155
|
+
|
156
|
+
10.assert.not == 9
|
157
|
+
|
158
|
+
== Lambda Assertions
|
159
|
+
|
160
|
+
Passing +assert+ a `Proc` object, or any object that responds to `#call`,
|
161
|
+
will be used as if it were a block. This allows for a simple way to quickly
|
162
|
+
create reusable assertions.
|
163
|
+
|
164
|
+
palindrome = lambda{ |word| word == word.reverse }
|
165
|
+
|
166
|
+
"abracarba".assert palindrome
|
167
|
+
|
168
|
+
The message for a failed assertion will come from calling `#to_s` on the
|
169
|
+
object.
|
170
|
+
|
171
|
+
== RSpec-style Assertion Matchers
|
172
|
+
|
173
|
+
If an object passed to assert responds to `#matches?` then AE will handle
|
174
|
+
the object as an RSpec-style mather, the receiver will be passed to the
|
175
|
+
`#matches?` method to determine if the assertion passes and RSpec matcher
|
176
|
+
message methods will be used if they are defined.
|
177
|
+
|
178
|
+
palindrome = Object.new
|
179
|
+
|
180
|
+
def palindrome.matches?(word)
|
181
|
+
word == word.reverse
|
182
|
+
end
|
183
|
+
|
184
|
+
"abracarba".assert palindrome
|
185
|
+
|
186
|
+
|
187
|
+
== Identity Assertions
|
188
|
+
|
189
|
+
Rather then the general form.
|
190
|
+
|
191
|
+
x = 10
|
192
|
+
x.assert.object_id == x.object_id
|
193
|
+
|
194
|
+
We can use Ruby's own <tt>equal?</tt> method.
|
195
|
+
|
196
|
+
x.assert.equal?(x)
|
197
|
+
|
198
|
+
AE provides <tt>identical?</tt> method as an alternative
|
199
|
+
to make it a bit more clear.
|
200
|
+
|
201
|
+
x.assert.identical?(x)
|
202
|
+
|
203
|
+
|
204
|
+
== Equality Assertions
|
205
|
+
|
206
|
+
The most common assertion is that of value equality (<tt>==</tt>),
|
207
|
+
as we have seen throughout this document. But other forms of
|
208
|
+
equality can be verified as easily. We have already mentioned
|
209
|
+
identity. In addition there is <i>type equality</i>.
|
210
|
+
|
211
|
+
17.assert.eql? 17
|
212
|
+
|
213
|
+
Assertion.assert.raised? do
|
214
|
+
17.assert.eql? 17.0
|
215
|
+
end
|
216
|
+
|
217
|
+
And there is <i>case equality</i>.
|
218
|
+
|
219
|
+
Numeric.assert === 3
|
220
|
+
|
221
|
+
|
222
|
+
== Checking Equality with a Block
|
223
|
+
|
224
|
+
Because operators can not take blocks, and at times blocks can
|
225
|
+
be convenient means of supplying a value to an assertion,
|
226
|
+
AE has defined alternate renditions of the equality methods.
|
227
|
+
For equal? and eql?, the method names are the same, they simply
|
228
|
+
can take a block in place of an argument if need be.
|
229
|
+
|
230
|
+
For <i>value equality</i> (<tt>==</tt>), the method is called <tt>eq?</tt>.
|
231
|
+
|
232
|
+
10.assert.eq? do
|
233
|
+
10.0
|
234
|
+
end
|
235
|
+
|
236
|
+
And should it fail...
|
237
|
+
|
238
|
+
Assertion.assert.raised? do
|
239
|
+
10.assert.eq? do
|
240
|
+
20
|
241
|
+
end
|
242
|
+
end
|
243
|
+
|
244
|
+
== Case Equality
|
245
|
+
|
246
|
+
For <i>case equality</i> (<tt>===</tt>), it is <tt>case?</tt>.
|
247
|
+
|
248
|
+
Numeric.assert.case? do
|
249
|
+
"3".to_i
|
250
|
+
end
|
251
|
+
|
252
|
+
Assertion.assert.raised? do
|
253
|
+
Numeric.assert.case? do
|
254
|
+
"3"
|
255
|
+
end
|
256
|
+
end
|
257
|
+
|
258
|
+
|
259
|
+
== Regular Expressions
|
260
|
+
|
261
|
+
Regular Expressions can be used to make assertions in much the same way as equality.
|
262
|
+
|
263
|
+
/i/.assert =~ "i"
|
264
|
+
|
265
|
+
Assertion.assert.raised? do
|
266
|
+
/i/.assert =~ "g"
|
267
|
+
end
|
268
|
+
|
269
|
+
Conversely the String class recognizes the #=~ method as well.
|
270
|
+
|
271
|
+
"i".assert =~ /i/
|
272
|
+
|
273
|
+
Assertion.assert.raised? do
|
274
|
+
"i".assert =~ /g/
|
275
|
+
end
|
276
|
+
|
277
|
+
|
278
|
+
== Exception Assertions
|
279
|
+
|
280
|
+
Validating errors is easy too, as has already been shown
|
281
|
+
in the document to verify assertion failures.
|
282
|
+
|
283
|
+
StandardError.assert.raised? do
|
284
|
+
unknown_method
|
285
|
+
end
|
286
|
+
|
287
|
+
|
288
|
+
== Assertions on Object State
|
289
|
+
|
290
|
+
While testing or specifying the internal state of an object is
|
291
|
+
generally considered poor form, there are times when it is
|
292
|
+
necessary. Assert combined with +instance_eval+ makes it easy too.
|
293
|
+
|
294
|
+
class X
|
295
|
+
attr :a
|
296
|
+
def initialize(a); @a = a; end
|
297
|
+
end
|
298
|
+
|
299
|
+
x = X.new(1)
|
300
|
+
|
301
|
+
x.assert.instance_eval do
|
302
|
+
@a == 1
|
303
|
+
end
|
304
|
+
|
305
|
+
|
306
|
+
== Catch/Try Assertions
|
307
|
+
|
308
|
+
Catch/Try throws can be tested via <tt>Symbol#thrown?</tt>.
|
309
|
+
|
310
|
+
:hookme.assert.thrown? do
|
311
|
+
throw :hookme
|
312
|
+
end
|
313
|
+
|
314
|
+
Alternatively, a lambda containing the potential throw
|
315
|
+
can be the receiver using <tt>throws?</tt>.
|
316
|
+
|
317
|
+
hook = lambda{ throw :hookme }
|
318
|
+
|
319
|
+
hook.assert.throws?(:hookme)
|
320
|
+
|
321
|
+
|
322
|
+
== Assertions on Proc Changes
|
323
|
+
|
324
|
+
I have to admit I'm not sure how this is useful,
|
325
|
+
but I found it in the Bacon API and ported it over
|
326
|
+
just for sake of thoroughness.
|
327
|
+
|
328
|
+
a = 0
|
329
|
+
|
330
|
+
l = lambda{ a }
|
331
|
+
|
332
|
+
l.assert.change?{ a +=1 }
|
333
|
+
|
334
|
+
|
335
|
+
== Assertion on literal True, False and Nil
|
336
|
+
|
337
|
+
Ruby already provides the #nil? method.
|
338
|
+
|
339
|
+
nil.assert.nil?
|
340
|
+
|
341
|
+
AE adds <tt>true?</tt> and <tt>false?</tt> which acts accordingly.
|
342
|
+
|
343
|
+
true.assert.true?
|
344
|
+
false.assert.false?
|
345
|
+
|
346
|
+
|
347
|
+
== Send Assertions
|
348
|
+
|
349
|
+
Assert that a method can be successfully called.
|
350
|
+
|
351
|
+
"STRING".assert.send?(:upcase)
|
352
|
+
|
353
|
+
|
354
|
+
== Numeric Delta and Epsilon
|
355
|
+
|
356
|
+
You may wish to assert that a numeric value is with some
|
357
|
+
range.
|
358
|
+
|
359
|
+
3.in_delta?(1,5)
|
360
|
+
|
361
|
+
Or minimum range.
|
362
|
+
|
363
|
+
3.in_epsilon?(3,5)
|
364
|
+
|
365
|
+
|
366
|
+
== Verifying Object State
|
367
|
+
|
368
|
+
Not surprisingly if underlying object state needs to be verified, +instance_eval+
|
369
|
+
can be used in conjunction with +assert+.
|
370
|
+
|
371
|
+
class X
|
372
|
+
attr :a
|
373
|
+
def initialize(a); @a = a; end
|
374
|
+
end
|
375
|
+
|
376
|
+
x = X.new(4)
|
377
|
+
|
378
|
+
x.instance_eval do
|
379
|
+
@a.assert == 4
|
380
|
+
end
|
381
|
+
|
382
|
+
However #instance_eval is a reserved method for the underlying Assertor class,
|
383
|
+
so it cannot be used on #assert, e.g.
|
384
|
+
|
385
|
+
x.assert.instance_eval do
|
386
|
+
@a == "obvisouly wrong"
|
387
|
+
end
|
388
|
+
|
389
|
+
AE offers an optional helper method for times when testing underlying private
|
390
|
+
or protected methods is important, called #pry. See the QED on pry for more
|
391
|
+
information.
|
392
|
+
|
393
|
+
For some testing underlying implementation might be considered poor
|
394
|
+
form. You will get no argument here. It should be used thoughtfully,
|
395
|
+
but I would not bet against there being occasions when such validations
|
396
|
+
might be needed.
|
397
|
+
|
398
|
+
= Subjunctives
|
399
|
+
|
400
|
+
Okay. I can hear the BDDers rumbling, "where's the *should?*"
|
401
|
+
AE has nothing against "should", but there are different
|
402
|
+
approaches for utilizing should nomenclature in specifications,
|
403
|
+
and AE wants to be open to these techniques. One of which
|
404
|
+
is how Shoulda (http://shoulda.rubyforge.org) utilizes
|
405
|
+
+should+ in a way analogous to RSpec's use of +it+.
|
406
|
+
|
407
|
+
Even so, AE provides an optional mixin called +Subjunctive+ which
|
408
|
+
can be used to create assertor methods with English subjunctive
|
409
|
+
terms, such as +should+, or +must+, +shall+ and +will+.
|
410
|
+
To load this library use:
|
411
|
+
|
412
|
+
require 'ae/subjunctive'
|
413
|
+
|
414
|
+
Then all that is required it to define a subjunctive method for all
|
415
|
+
objects. For example:
|
416
|
+
|
417
|
+
def will(*args, &block)
|
418
|
+
Assertor.new(self, :backtrace=>caller).be(*args,&block)
|
419
|
+
end
|
420
|
+
|
421
|
+
It's that easy. Because of their commonality AE provides two such terms,
|
422
|
+
+should+ and +must+ as optional add-ons out-of-the-box.
|
423
|
+
|
424
|
+
require 'ae/should'
|
425
|
+
require 'ae/must'
|
426
|
+
|
427
|
+
We will use these two methods interchangeable for the rest of this
|
428
|
+
demonstration, but to be clear they both work exactly the same way,
|
429
|
+
and almost exactly like +assert+.
|
430
|
+
|
431
|
+
Keep in mind, AE "conical" functionality does not entail the subjunctive
|
432
|
+
forms. These are simply options you can load via your <tt>test_helper.rb</tt>,
|
433
|
+
or similar script, if you prefer these nomenclatures.
|
434
|
+
|
435
|
+
|
436
|
+
== Fluent Notation and Antonyms
|
437
|
+
|
438
|
+
Like +assert+, +should+ and +must+ can be used as higher order functions.
|
439
|
+
|
440
|
+
4.should == 4
|
441
|
+
4.must == 4
|
442
|
+
|
443
|
+
Antonyms provided for +should+ as <tt>should!</tt> (read "should not") and +shouldnt+.
|
444
|
+
For +must+ +must+, they are <tt>must!</tt> and +wont+.
|
445
|
+
|
446
|
+
4.should! == 5
|
447
|
+
4.shouldnt == 5
|
448
|
+
|
449
|
+
4.must! == 5
|
450
|
+
4.wont == 5
|
451
|
+
|
452
|
+
|
453
|
+
== To Be
|
454
|
+
|
455
|
+
On occasions where the English readability of a specification is hindered,
|
456
|
+
+be+ can be used.
|
457
|
+
|
458
|
+
StandardError.must.be.raised? do
|
459
|
+
unknown_method
|
460
|
+
end
|
461
|
+
|
462
|
+
The +be+ method is the same as +assert+ with the single exception
|
463
|
+
that it will compare a lone argument to the receiver using +equate?+,
|
464
|
+
unlike +assert+ which simply checks to see that the argument evaluates
|
465
|
+
as true.
|
466
|
+
|
467
|
+
10.should.be 10
|
468
|
+
10.should.be 10.0
|
469
|
+
10.should.be Numeric
|
470
|
+
|
471
|
+
Assertion.assert.raised? do
|
472
|
+
10.should.be "40"
|
473
|
+
end
|
474
|
+
|
475
|
+
|
476
|
+
== Indefinite Articles
|
477
|
+
|
478
|
+
Additional English forms are +a+ and +an+, equivalent to +be+ except
|
479
|
+
that they use <tt>case?</tt> (same as <tt>#===</tt>) instead of
|
480
|
+
<tt>equate?</tt> when acting on a single argument.
|
481
|
+
|
482
|
+
"hi".must.be.a String
|
483
|
+
|
484
|
+
Assertion.assert.raised? do
|
485
|
+
/x/.must.be.a /x/
|
486
|
+
end
|
487
|
+
|
488
|
+
Otherwise they are interchangeable.
|
489
|
+
|
490
|
+
"hi".must.be.an.instance_of?(String)
|
491
|
+
|
492
|
+
The indefinite articles work well when a noun follows as an arguments.
|
493
|
+
|
494
|
+
palindrome = lambda{ |x| x == x.reverse }
|
495
|
+
|
496
|
+
"abracarba".must.be.a palindrome
|
497
|
+
|
498
|
+
|
499
|
+
= Expect Method
|
500
|
+
|
501
|
+
Expect is another assertion nomenclature available for use in your
|
502
|
+
tests or specifications. Inspired by Jay Fields' Expectations library,
|
503
|
+
it provides convenient syntax for creating exception and case equality
|
504
|
+
assertions.
|
505
|
+
|
506
|
+
require 'ae/expect'
|
507
|
+
|
508
|
+
|
509
|
+
== Underlying Comparison
|
510
|
+
|
511
|
+
Expect uses #=== for comparison. So providing an argument and a block to
|
512
|
+
#expect we can test for a somewhat broader range of compassion than #assert.
|
513
|
+
For example we can test for a subclass.
|
514
|
+
|
515
|
+
expect Numeric do
|
516
|
+
3
|
517
|
+
end
|
518
|
+
|
519
|
+
Assertion.assert.raised? do
|
520
|
+
expect Numeric do
|
521
|
+
"3"
|
522
|
+
end
|
523
|
+
end
|
524
|
+
|
525
|
+
|
526
|
+
== Exception Expectation
|
527
|
+
|
528
|
+
If the comparator is an Exception class or a instance of an Exception class,
|
529
|
+
then #expect will check to see if the block raises that kind of exception.
|
530
|
+
|
531
|
+
expect StandardError do
|
532
|
+
some_undefined_method
|
533
|
+
end
|
534
|
+
|
535
|
+
expect Assertion do
|
536
|
+
expect(nil)
|
537
|
+
end
|
538
|
+
|
539
|
+
This is an important distinction to note because it means #expect can not be used
|
540
|
+
if verify instances of Exception classes.
|
541
|
+
|
542
|
+
Assertion.assert.raised? do
|
543
|
+
expect Exception do
|
544
|
+
Exception.new
|
545
|
+
end
|
546
|
+
end
|
547
|
+
|
548
|
+
|
549
|
+
== Regex Expectations
|
550
|
+
|
551
|
+
That #expect entails #=== also means we can check for Regexp matches.
|
552
|
+
|
553
|
+
expect /x/ do
|
554
|
+
"oooxooo"
|
555
|
+
end
|
556
|
+
|
557
|
+
|
558
|
+
== Expected Method
|
559
|
+
|
560
|
+
We can use #expected to make the receiver the object of expectation.
|
561
|
+
|
562
|
+
x = "dummy"
|
563
|
+
|
564
|
+
/x/.expected do
|
565
|
+
"x"
|
566
|
+
end
|
567
|
+
|
568
|
+
|
569
|
+
== Without Block
|
570
|
+
|
571
|
+
Without a block, the receiver is compared to the argument.
|
572
|
+
|
573
|
+
x.expect String
|
574
|
+
|
575
|
+
|
576
|
+
== Negative Forms
|
577
|
+
|
578
|
+
Like #assert, #expect has a negated form called #expect!
|
579
|
+
|
580
|
+
expect! /x/ do
|
581
|
+
"o"
|
582
|
+
end
|
583
|
+
|
584
|
+
The pure word form for those who do not like the clever use of the
|
585
|
+
explimation mark is #forbid.
|
586
|
+
|
587
|
+
forbid /x/ do
|
588
|
+
"o"
|
589
|
+
end
|
590
|
+
|
591
|
+
|
592
|
+
== Functor, or Higher Order Function
|
593
|
+
|
594
|
+
Like #assert, #expect can be used used as a *fluid* notation.
|
595
|
+
|
596
|
+
10.expect == 10
|
597
|
+
|
598
|
+
In which case it works just like #assert, including negative forms.
|
599
|
+
|
600
|
+
10.expect! == 11
|
601
|
+
10.forbid == 11
|
602
|
+
|
603
|
+
|
604
|
+
= Assertion Counts
|
605
|
+
|
606
|
+
AE tracks the number of assertions made and the number that failed to pass.
|
607
|
+
We can reset the count using the +recount+ class method.
|
608
|
+
|
609
|
+
old_counts = AE::Assertor.recount
|
610
|
+
|
611
|
+
For example if we have one assertion pass and another fail.
|
612
|
+
|
613
|
+
assert(true)
|
614
|
+
|
615
|
+
expect Assertion do
|
616
|
+
assert(false)
|
617
|
+
end
|
618
|
+
|
619
|
+
We will see that AE counted three assertions and one failure.
|
620
|
+
|
621
|
+
counts = AE::Assertor.counts.dup
|
622
|
+
|
623
|
+
counts[:total].assert == 3
|
624
|
+
counts[:pass].assert == 2
|
625
|
+
counts[:fail].assert == 1
|
626
|
+
|
627
|
+
The #expect call is an assertion too, which is why the total count is 3
|
628
|
+
rather than 2.
|
629
|
+
|
630
|
+
Now that we are done checking counts we will restore them so that
|
631
|
+
any other demos being run with this will tally correctly.
|
632
|
+
|
633
|
+
AE::Assertor.recount(old_counts)
|
634
|
+
AE::Assertor.counts[:total] += 3
|
635
|
+
AE::Assertor.counts[:pass] += 3
|
636
|
+
|
637
|
+
|
638
|
+
= Matchers
|
639
|
+
|
640
|
+
Matchers are simply Procs or objects with the proper interface that can be
|
641
|
+
passed to #assert or #refute (or other Assertor) as an ecapsulated test.
|
642
|
+
|
643
|
+
== Proc or #to_proc
|
644
|
+
|
645
|
+
Passing a Proc object or an object that responds to :to_proc, will use it
|
646
|
+
as if it were a block of the method. This allows for a simple way to quickly
|
647
|
+
create reusable assertions.
|
648
|
+
|
649
|
+
palindrome = lambda{ |word| word == word.reverse }
|
650
|
+
|
651
|
+
"abracarba".assert palindrome
|
652
|
+
|
653
|
+
== #matches?
|
654
|
+
|
655
|
+
Additionally if an object responds to #matches? then the receiver
|
656
|
+
will be passed to this method to determine if the assertion passes.
|
657
|
+
|
658
|
+
palindrome = Object.new
|
659
|
+
|
660
|
+
def palindrome.matches?(word)
|
661
|
+
word == word.reverse
|
662
|
+
end
|
663
|
+
|
664
|
+
"abracarba".assert palindrome
|
665
|
+
|
666
|
+
== RSpec, Shoulda and other 3rd-Party Matchers
|
667
|
+
|
668
|
+
With tha addition of #matches?, AE supports the same interface for matchers
|
669
|
+
as RSpec. Any matcher library designed for use with RSpec should therefore
|
670
|
+
be usable with AE as well. This includes RSpecs own matchers and Shoulda's
|
671
|
+
excellent Rails matchers.
|
672
|
+
|
673
|
+
== Check Ok/No
|
674
|
+
|
675
|
+
The Check library is an optional library that can be used
|
676
|
+
to conveniently speed-up construction of reptitive assertions.
|
677
|
+
|
678
|
+
To use it, first require the library, then include the mixin
|
679
|
+
into the namespace in which you will utilize it.
|
680
|
+
|
681
|
+
require 'ae/check'
|
682
|
+
|
683
|
+
include AE::Check
|
684
|
+
|
685
|
+
Now we can define ok/no check procedures. A one-off procedure is
|
686
|
+
defined with a block only.
|
687
|
+
|
688
|
+
check do |x, y|
|
689
|
+
x == y
|
690
|
+
end
|
691
|
+
|
692
|
+
ok 1,1
|
693
|
+
|
694
|
+
no 1,2
|
695
|
+
|
696
|
+
To define reusable check procedures, give the procedure a name.
|
697
|
+
|
698
|
+
check :palindrome do |x|
|
699
|
+
x.reverse == x
|
700
|
+
end
|
701
|
+
|
702
|
+
This will also cause the current check method to be set.
|
703
|
+
Later in the code, the check procedure can be reset to this
|
704
|
+
by just passing the name.
|
705
|
+
|
706
|
+
check :palindrome
|
707
|
+
|
708
|
+
ok 'abracarba'
|
709
|
+
|
710
|
+
no 'foolishness'
|
711
|
+
|
712
|
+
The Check mixin comes preloaded with a few standard checks.
|
713
|
+
By default the `:equality` procedure is used.
|
714
|
+
|
715
|
+
check :equality
|
716
|
+
|
717
|
+
ok 1=>1.0
|
718
|
+
|
719
|
+
Notice the use of the hash argument here. This is a useful construct for
|
720
|
+
many check procedures becuase it it akin to the `#=>` pattern we often
|
721
|
+
see in code examples and it also allows for multiple assertions in one call.
|
722
|
+
For instance, in the case of `:equality`, multiple entries convey a
|
723
|
+
meaning of logical-or.
|
724
|
+
|
725
|
+
ok 1=>2, 1=>1
|
726
|
+
|
727
|
+
This would pass becuase the second assertion of equality is true.
|
728
|
+
|
729
|
+
Another built in check is `:case_equality` which uses `#===` instead of `#==`
|
730
|
+
to make the comparison.
|
731
|
+
|
732
|
+
check :case_equality
|
733
|
+
|
734
|
+
ok 1=>Integer
|
735
|
+
|
736
|
+
|
data/lib/ae.yml
CHANGED
@@ -40,9 +40,9 @@ revision: 0
|
|
40
40
|
created: '2008-08-17'
|
41
41
|
summary: Assertive Expressive
|
42
42
|
title: AE
|
43
|
-
version: 1.8.
|
43
|
+
version: 1.8.1
|
44
44
|
name: ae
|
45
45
|
description: ! "Assertive Expressive is an assertions library specifically designed
|
46
46
|
\nfor reuse by other test frameworks."
|
47
47
|
organization: Rubyworks
|
48
|
-
date: '2011-11-
|
48
|
+
date: '2011-11-04'
|
data/lib/ae/ansi.rb
ADDED
@@ -0,0 +1,26 @@
|
|
1
|
+
require 'ansi/diff'
|
2
|
+
|
3
|
+
module AE
|
4
|
+
|
5
|
+
# Default ANSI mode is "on".
|
6
|
+
@ansi = true
|
7
|
+
|
8
|
+
# ANSI mode.
|
9
|
+
#
|
10
|
+
# @return [Boolean] ANSI mode.
|
11
|
+
def self.ansi?
|
12
|
+
@ansi
|
13
|
+
end
|
14
|
+
|
15
|
+
# To turn of ANSI colorized error messages off, set
|
16
|
+
# ansi to +false+ in your test helper.
|
17
|
+
#
|
18
|
+
# @example
|
19
|
+
# AE.ansi = false
|
20
|
+
#
|
21
|
+
def self.ansi=(boolean)
|
22
|
+
@ansi = boolean
|
23
|
+
end
|
24
|
+
|
25
|
+
end
|
26
|
+
|
data/qed/08_check.rdoc
ADDED
@@ -0,0 +1,63 @@
|
|
1
|
+
== Check Ok/No
|
2
|
+
|
3
|
+
The Check library is an optional library that can be used
|
4
|
+
to conveniently speed-up construction of reptitive assertions.
|
5
|
+
|
6
|
+
To use it, first require the library, then include the mixin
|
7
|
+
into the namespace in which you will utilize it.
|
8
|
+
|
9
|
+
require 'ae/check'
|
10
|
+
|
11
|
+
include AE::Check
|
12
|
+
|
13
|
+
Now we can define ok/no check procedures. A one-off procedure is
|
14
|
+
defined with a block only.
|
15
|
+
|
16
|
+
check do |x, y|
|
17
|
+
x == y
|
18
|
+
end
|
19
|
+
|
20
|
+
ok 1,1
|
21
|
+
|
22
|
+
no 1,2
|
23
|
+
|
24
|
+
To define reusable check procedures, give the procedure a name.
|
25
|
+
|
26
|
+
check :palindrome do |x|
|
27
|
+
x.reverse == x
|
28
|
+
end
|
29
|
+
|
30
|
+
This will also cause the current check method to be set.
|
31
|
+
Later in the code, the check procedure can be reset to this
|
32
|
+
by just passing the name.
|
33
|
+
|
34
|
+
check :palindrome
|
35
|
+
|
36
|
+
ok 'abracarba'
|
37
|
+
|
38
|
+
no 'foolishness'
|
39
|
+
|
40
|
+
The Check mixin comes preloaded with a few standard checks.
|
41
|
+
By default the `:equality` procedure is used.
|
42
|
+
|
43
|
+
check :equality
|
44
|
+
|
45
|
+
ok 1=>1.0
|
46
|
+
|
47
|
+
Notice the use of the hash argument here. This is a useful construct for
|
48
|
+
many check procedures becuase it it akin to the `#=>` pattern we often
|
49
|
+
see in code examples and it also allows for multiple assertions in one call.
|
50
|
+
For instance, in the case of `:equality`, multiple entries convey a
|
51
|
+
meaning of logical-or.
|
52
|
+
|
53
|
+
ok 1=>2, 1=>1
|
54
|
+
|
55
|
+
This would pass becuase the second assertion of equality is true.
|
56
|
+
|
57
|
+
Another built in check is `:case_equality` which uses `#===` instead of `#==`
|
58
|
+
to make the comparison.
|
59
|
+
|
60
|
+
check :case_equality
|
61
|
+
|
62
|
+
ok 1=>Integer
|
63
|
+
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: ae
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 1.8.
|
4
|
+
version: 1.8.1
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -13,7 +13,7 @@ date: 2011-11-04 00:00:00.000000000 Z
|
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: ansi
|
16
|
-
requirement: &
|
16
|
+
requirement: &31611960 !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ! '>='
|
@@ -21,10 +21,10 @@ dependencies:
|
|
21
21
|
version: '0'
|
22
22
|
type: :runtime
|
23
23
|
prerelease: false
|
24
|
-
version_requirements: *
|
24
|
+
version_requirements: *31611960
|
25
25
|
- !ruby/object:Gem::Dependency
|
26
26
|
name: detroit
|
27
|
-
requirement: &
|
27
|
+
requirement: &31611420 !ruby/object:Gem::Requirement
|
28
28
|
none: false
|
29
29
|
requirements:
|
30
30
|
- - ! '>='
|
@@ -32,10 +32,10 @@ dependencies:
|
|
32
32
|
version: '0'
|
33
33
|
type: :development
|
34
34
|
prerelease: false
|
35
|
-
version_requirements: *
|
35
|
+
version_requirements: *31611420
|
36
36
|
- !ruby/object:Gem::Dependency
|
37
37
|
name: qed
|
38
|
-
requirement: &
|
38
|
+
requirement: &31610920 !ruby/object:Gem::Requirement
|
39
39
|
none: false
|
40
40
|
requirements:
|
41
41
|
- - ! '>='
|
@@ -43,7 +43,7 @@ dependencies:
|
|
43
43
|
version: '0'
|
44
44
|
type: :development
|
45
45
|
prerelease: false
|
46
|
-
version_requirements: *
|
46
|
+
version_requirements: *31610920
|
47
47
|
description: ! "Assertive Expressive is an assertions library specifically designed
|
48
48
|
\nfor reuse by other test frameworks."
|
49
49
|
email:
|
@@ -53,6 +53,7 @@ extensions: []
|
|
53
53
|
extra_rdoc_files:
|
54
54
|
- HISTORY.rdoc
|
55
55
|
- README.rdoc
|
56
|
+
- QED.rdoc
|
56
57
|
- NOTICE.rdoc
|
57
58
|
files:
|
58
59
|
- .ruby
|
@@ -61,6 +62,7 @@ files:
|
|
61
62
|
- lib/ae/adapters/minitest.rb
|
62
63
|
- lib/ae/adapters/rspec.rb
|
63
64
|
- lib/ae/adapters/testunit.rb
|
65
|
+
- lib/ae/ansi.rb
|
64
66
|
- lib/ae/assert.rb
|
65
67
|
- lib/ae/assertion.rb
|
66
68
|
- lib/ae/assertor.rb
|
@@ -85,8 +87,10 @@ files:
|
|
85
87
|
- qed/05_expect.rdoc
|
86
88
|
- qed/06_counts.rdoc
|
87
89
|
- qed/07_matchers.rdoc
|
90
|
+
- qed/08_check.rdoc
|
88
91
|
- HISTORY.rdoc
|
89
92
|
- README.rdoc
|
93
|
+
- QED.rdoc
|
90
94
|
- NOTICE.rdoc
|
91
95
|
homepage: http://rubyworks.github.com/ae
|
92
96
|
licenses:
|