-
Notifications
You must be signed in to change notification settings - Fork 15
Expand file tree
/
Copy path12-testing.html
More file actions
982 lines (862 loc) · 43.2 KB
/
Copy path12-testing.html
File metadata and controls
982 lines (862 loc) · 43.2 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>CS 4971: 12-testing slide set</title>
<meta name="description" content="A set of slides for UVa's Service Learning Practicum course">
<meta name="author" content="Aaron Bloomfield">
<meta name="apple-mobile-web-app-capable" content="yes" />
<meta name="apple-mobile-web-app-status-bar-style" content="black-translucent" />
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no, minimal-ui">
<link rel="stylesheet" href="../slides/reveal.js/css/reveal.css">
<link rel="stylesheet" href="../slides/reveal.js/css/theme/black.css" id="theme">
<link rel="stylesheet" href="../slides/css/slp.css">
<!-- Code syntax highlighting -->
<link rel="stylesheet" href="../slides/reveal.js/lib/css/zenburn.css">
<!-- Printing and PDF exports -->
<script>
var link = document.createElement( 'link' );
link.rel = 'stylesheet';
link.type = 'text/css';
link.href = window.location.search.match( /print-pdf/gi ) ? '../slides/reveal.js/css/print/pdf.css' : '../slides/reveal.js/css/print/paper.css';
document.getElementsByTagName( 'head' )[0].appendChild( link );
</script>
<!--[if lt IE 9]>
<script src="../slides/reveal.js/lib/js/html5shiv.js"></script>
<![endif]-->
<script type="text/javascript" src="../slides/js/dhtmlwindow.js"></script>
<script type="text/javascript" src="../slides/js/canvas.js"></script>
<link rel="stylesheet" href="../slides/css/dhtmlwindow.css" type="text/css">
</head>
<body>
<div id="dhtmlwindowholder"><span style="display:none"></span></div>
<div class="reveal">
<!-- Any section element inside of this container is displayed as a slide -->
<div class="slides">
<section data-markdown><script type="text/template">
# CS 4971
### Capstone Practicum II
<center><small>[Aaron Bloomfield](http://www.cs.virginia.edu/~asb) / [aaron@virginia.edu](mailto:aaron@virginia.edu) / [@bloomfieldaaron](http://twitter.com/bloomfieldaaron)</small></center>
<center><small>Repository: [github.com/aaronbloomfield/slp](http://github.com/aaronbloomfield/slp) / [↑](index.html) / <a href="12-testing.html?print-pdf"><img class="print" width="20" src="images/print-icon.png"></a></small></center>
## Testing and the Test Plan
</script></section>
<section data-markdown><script type="text/template">
#Contents
[Introduction](#/introduction)
[Requirements](#/requirements)
[Terminology](#/terminology)
[Test Types](#/specific)
[Master Test Plan](#/mastertestplan)
[Specific Test Plans](#/specificplans)
[Main Code Coverage](#/codecoverage)
[Javascript Code Coverage](#/jscodecoverage)
[Conclusions](#/conclusion)
</script></section>
<section>
<section id="introduction" data-markdown><script type="text/template">
# Introduction
</script></section>
<section data-markdown><script type="text/template">
## Testing is not absolute!
- You can never prove that a (non-trivial) program is bug free
- It's a corollary of the [halting problem](http://en.wikipedia.org/wiki/Halting_problem): prove that a program halts (successfully) on a given input
- Alan Turing proved that this was *undecidable* (meaning an algorithm cannot be constructed that could decide this)
- It's very easy to prove that a bug exists, though...
- But you can give yourself -- and your customer -- some sense of confidence that it will *likely* work correctly
</script></section>
<section data-markdown><script type="text/template">
## Bug rate
- A 1990 review estimated that there were 1 to 3 bugs per 100 executable statements
- Actually, programmers make many more (1.5 bugs per executable statement!), but they will catch 99% of those
- Your job is to catch the other 1%
</script></section>
<section data-markdown><script type="text/template">
## Costs of transition
- A much older study indicated that transition and maintenance takes 67% of the cost of a project
- And that doesn't include an additional 15% of testing during development!
- Granted, that was during the waterfall methodology craze...
- You are finished with this class when the semester ends
- If you want to help debug it later on, that's great -- but there is no requirement nor expectation for you to do so
- So we need to ensure that it is well tested before the semester concludes
</script></section>
<section data-markdown><script type="text/template">
## Things to keep in mind...
- We can *NOT* rely on the customer to complete tests!
- There will be *NO* new code features developed during iterations 12-14
- Only test cases, bug fixes, minor feature changes requested by the customer, etc.
- Your test plan must be significant, and must have real thought put into it
- You can start at the [Wikipedia Software Testing portal page](http://en.wikipedia.org/wiki/Portal:Software_Testing)
- If you wait until the last minute, then throw something half-baked together, I am going to get very cranky
</script></section>
<section data-markdown><script type="text/template">
## References
- Wikipedia, of course: the [portal on software testing](http://en.wikipedia.org/wiki/Portal:Software_Testing)
- [IEEE 829 (2008) test plan standard](http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=4578383)
- Free if viewed from UVa; it's also in Collab as [Resources/misc/IEEE-829-2008.pdf](https://collab.itc.virginia.edu/access/content/group/34417471-f045-46f1-a50f-738fa387fa54/misc/IEEE-829-2008.pdf) as well
- But see the warnings later in this slide set before you read through this
</script></section>
</section>
<section>
<section id="requirements" data-markdown><script type="text/template">
# Requirements
</script></section>
<section data-markdown><script type="text/template">
## Why requirements?
- We can't have a discussion about testing without having some set of requirements to test against
- Requirements are categorized into functional and non-functional
- Software attributes are described by the acronym FURPS
- Functional, Usability, Reliability, Performance, Supportability
</script></section>
<section data-markdown><script type="text/template">
## Functional & non-functional requirements
- Functional requirements define what a system is supposed to do ("system shall *do* X")
- Given an input, what is the desired outcome / output?
- These are *objective* tests and results
- Non-functional define how a system is supposed to be ("system shall *be* X")
- Rather than specific behaviors, it is criteria used to judge a system
- These are *subjective* tests and results
- Functional requirements are defined in the system design; non-functional in the system architecture
</script></section>
<section data-markdown><script type="text/template">
## Types of requirements
- Requirements are usually described by the FURPS acronym
- F: Funcionality
- U: Usability
- R: Reliability
- P: Performance
- S: Stability
</script></section>
<section data-markdown><script type="text/template">
## Furps: Functionality
- Feature set
- Capabilities
- Generality
- Security
</script></section>
<section data-markdown><script type="text/template">
## fUrps: Usability
- Human factors
- Aesthetics
- Consistency
- Documentation
</script></section>
<section data-markdown><script type="text/template">
## fuRps: Reliability
- Frequency/severity of failure
- Recoverability
- Predictability
- Accuracy
- Mean time to failure
</script></section>
<section data-markdown><script type="text/template">
## furPs: Performance
- Speed
- Efficiency
- Resource consumption
- Throughput
- Response time
</script></section>
<section data-markdown><script type="text/template">
## furpS: Supportability
- Testability
- Extensibility
- Adaptability
- Maintainability
- Compatibility
- Configurability
- Serviceability
- Installability
- Localizability
- Portability
</script></section>
<section data-markdown><script type="text/template">
## Analyzing requirement solicitation
- You've all seen SIS
- Which requirements did it have and meet?
- Which ones did not not have and/or not meet?
- Another example: the R25 room scheduler
</script></section>
<section data-markdown><script type="text/template">
## Requirements document
- Creating the test plan assumes that you have an updated requirements document
- If there are no requirements (or the requirements are unknown), then there is no point in testing!
</script></section>
</section>
<section>
<section id="terminology" data-markdown><script type="text/template">
# Terminology
</script></section>
<section data-markdown><script type="text/template">
## What is a software "bug"?
- ... when the system crashes
- ... when the system has undesirable results
- ... when the system does not handle border cases
- ... when the system has invalid output
- ... when the system has calculation errors
- ... when the system has control flow errors
- ... when the system has race conditions
- ... when the system has errors interpreting data
- More [here](http://en.wikipedia.org/wiki/Software_bug#Common_types_of_computer_bugs)
</script></section>
<section data-markdown><script type="text/template">
## What is a software "defect"?
- Certainly any software bug
- When the software system does not conform to the requirement specification
- This assumes that the specification exists and is correct!
- Usability problems: too difficult to navigate, etc.
- Missing commands
- If it's not in the specification, then it's a specification error as well
- Performance issues
- Licensing issues!
- This is particularly relevant to us
</script></section>
<section data-markdown><script type="text/template">
## Terminology
- Black box testing: internal design is not considered, only external design
- White box testing: testing is based on detailed knowledge of how the code works
- Regression testing: testing the entire system after a modification has been made
- We are using Travis/Circle for this
- Acceptance testing: running the tests to make sure the system is in an 'acceptable' state (i.e., has passed all the tests)
</script></section>
<section data-markdown><script type="text/template">
## Alpha versus beta testing
- Alpha testing: done at the end of development to make sure the system works
- This is what you will be doing during development but BEFORE the end of iteration 11 (i.e., March 28th)
- Beta testing: an initial version is deployed, and tested "in-place" by the team, the customer, and possibly others
- This is what will happen in iterations 12-14
</script></section>
</section>
<section>
<section id="specific" data-markdown><script type="text/template">
# Test Types
</script></section>
<section data-markdown><script type="text/template">
## "On the fly" testing
- Try to do a bunch of things, and see if it doesn't work right
- This is the typical testing done on may types of projects (undergraduate course projects, short pieces of software, etc.)
- Good as an initial start, but this does not provide very good coverage...
</script></section>
<section data-markdown id="reqtesting"><script type="text/template">
## System or requirements testing
- Does the system handle the specified requirements?
- Go through each user case / user story, and see if all the parts are included as system functionality
- If not, then either the system has to be modified or the use case / user story needs to be updated
- Treat the system as a black box
- Once the user manual is written, you will need to ensure that it conforms to the usage of the system
</script></section>
<section data-markdown><script type="text/template">
## Unit testing
- A set of tests of each individual "unit"
- Method, procedure, class, etc.
- We have instructions on unit testing for the various platforms, which are good places to start
- The unit testing will need to cover ALL the aspects of the system
- There will be a significant amount of code for this!
</script></section>
<section data-markdown><script type="text/template">
## Installation testing
- Do the [installation instructions](02-deliverables.html#/installation) work?
- How do we know (meaning, how can we ascertain this)?
</script></section>
<section data-markdown><script type="text/template">
## Usability testing
- Most of these systems are intended to be used by specific, trained, individuals
- But not all!
- So while we don't need a perfect UI, it should still be relatively easy to use
- Can the user understand the application flow? Are the editing functions easy to use? Is there sufficient help documentation available (likely in the user manual)?
</script></section>
<section data-markdown><script type="text/template">
## Security testing
- Much of the information in these sites is protected under federal law!
- Can the system be hacked? Penetrated? Which pages can be accessed without credentials?
- How does the system handle authorization versus authentication?
</script></section>
<section data-markdown id="malformed"><script type="text/template">
## Malformed input testing
- How well does the system handle malformed input?
- For *each* type of input, what happens if the input is invalid?
- Likewise, how well does the system handle incorrect URLs?
- CakePHP can set the various 'debug' option in core.php to control how this is handled
- Rails and Django can do something similar
- But you may need to add a 404 page as well
- This may be included in the unit testing (for malformed input) or system testing (for malformed URLs)
</script></section>
<section data-markdown><script type="text/template">
## Testing we likely won't be doing...
- Compatibility testing: our system is only going to run on one type of configuration
- Load, stress, and performance testing: one or two groups may need this, but most won't
- Recovery testing: our system is not the type that can "crash" very often
</script></section>
<section data-markdown><script type="text/template">
## What about other test types?
- If your system requires other types of tests, then you should definitely include them!
</script></section>
</section>
<section>
<section id="mastertestplan" data-markdown><script type="text/template">
# Master Test Plan
</script></section>
<section data-markdown><script type="text/template">
## Master test plan
- Due at the end of iteration 9 (February 27th)
- This is a high-level overview of the testing
</script></section>
<section data-markdown><script type="text/template">
## IEEE standard
- [IEEE 829 (2008): Standard for Software and System Test Documentation](http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=4578383)
- Free if viewed from UVa; it's also in Collab as [Resources/misc/IEEE-829-2008.pdf](https://collab.itc.virginia.edu/access/content/group/54bcbf62-81cb-45bc-bf8c-c7d46467bba5/misc/IEEE-829-2008.pdf)
- And on [Wikipedia](http://en.wikipedia.org/wiki/IEEE_829)
- Section 8 concerns the Master Test Plan ("MTP" in the document)
- Some of it is fluff-inducing, so we'll take it with a ***HUGE*** grain of salt
- Basically, if something doesn't apply, then state so rather than make something up
- The listing there is meant to provide a structure, not cause you to BS your way through the test plan
</script></section>
<section data-markdown><script type="text/template">
## Relevant parts of the IEEE standard
- Some sections are where the real content is going
- Section 8.1.5.6: tools, techniques, methods, and metrics
- Section 8.2.1: Test processes including definition of test levels
- (well, less on the test levels...)
- That being said, the following slides are things that should go somewhere in the test plan...
- And questions to ask yourself -- the goal is for you to *think* about the testing process
</script></section>
<section data-markdown><script type="text/template">
## What *parts* of the system will be tested?
- Obviously not the framework code itself
- Probably not all the code of the gems (for Rails), packages (for Django), or plugins (for CakePHP) that you are using
- But likely your code to interact with them
- And certainly their functionality
- So saying "everything else" is missing the point
- List the specific methods and functionality that are going to be tested
- You can certainly use wildcards, etc., for this
- You have MVC components, and you will need to test all those parts
</script></section>
<section data-markdown><script type="text/template">
## Who will write the tests?
- Is it going to be all one person? (this is a BAD idea!)
- Will it be a small group within the overall group?
- Will it be everybody?
- If more than one person, who will write the tests for what?
</script></section>
<section data-markdown><script type="text/template">
## How will you determine what cases to test?
- Corner cases, malformed input cases, first to perform an action, scheduling in a full calendar, etc.
- This will require creating lots of *fixtures*, which are the sample data in the DB when the unit test is run
</script></section>
<section data-markdown><script type="text/template">
## How will you determine what tests to write?
- There are a lot of sites online for "how to test a X", where 'X' is a various part of your system
- Model, View, Controller, etc.
- See [here](http://book.cakephp.org/2.0/en/development/testing.html) for CakePHP, [here](http://guides.rubyonrails.org/testing.html) for Rails, and [here](https://docs.djangoproject.com/en/1.6/topics/testing/) for Django
</script></section>
<section data-markdown><script type="text/template">
## How will you indicate what is being tested?
- From the point of view of the tests, you will need to indicate what requirement is being tested
- This can be a one-line comment in the unit test
- From the point of view of the requirements, what tests are validating that those requirements work?
</script></section>
<section data-markdown><script type="text/template">
## What *other* types of tests (other than unit tests) will you be performing?
- Hint: look at the [previous section](#/specific) for details
- This includes, among other things: installation testing, usability testing, security testing, FURPS, etc.
- You will need to give an overview (following the concepts discussed in the previous few slides) for these tests as well
</script></section>
<section data-markdown><script type="text/template">
## Remember!
- If something the IEEE docuemnt doesn't make sense, doesn't apply, or is confusing, or seems to be in another language, you can ignore it
- That document is meant to provide *structure*, not to force you to create a lot of fluff
- I really hate fluff
</script></section>
<section data-markdown><script type="text/template">
## Tests that are elsewhere
- The master test plan will *briefly* describe other tests to be performed
- Installation test, usability test, security test, requirements test, etc.
- These will be completed in a separate iteration (that's the [speicifc test plan](#/specifictests))
- But you need a brief description of each in your MTP; see next slide
</script></section>
<section data-markdown><script type="text/template">
## MTP contents: the specific tests
- A brief description of the specific test types that you plan on performing
- What requirement each test (or test type) is validating
- How you plan on doing those tests. Some examples:
- Requirements testing may mean that you go through each of the requirements (which must be linked!) to ensure that they match
- Security testing will require a specific set of "hacks" against the system, such as X, Y, and Z
- Unit tests: what parts of the system are you testing? With which unit tests? How will you determine which unit tests to write?
- The point of this part is to think of the testing plan from the point of view of the tests that are going to be performed
- No details yet -- that's done later
</script></section>
<section data-markdown><script type="text/template">
## What should NOT be in the MTP (yet)
- The specific unit tests that you are going to be performing
- You should give overall ideas, but the specific unit tests are in the source code
- Full details about the specific test plan tests
- Example: the "installation test" can be a short paragraph stating that the tester(s) will follow the installation manual from an unconfigured machine
- Since there is no installation manual yet, there is no need to go into more detail at this point
</script></section>
<section data-markdown><script type="text/template">
## Sample MTP Outline
- Basically, the slides in this section describe the various sections of the MTP
- Relevant parts of the IEEE MTP document
- A listing of the parts of the system to be tested
- A description of who will write the tests
- How you will determine what cases to test?
- How will you determine what unit tests to write?
- How will you indicate what is being tested?
- What other (non-unit) tests will you be performing? This is the brief outline of the specific test plan
</script></section>
</section>
<section>
<section data-markdown id="specificplans"><script type="text/template">
# Specific test plans
</script></section>
<section data-markdown><script type="text/template">
## Specific test plans
- Whereas the master test plan (MTP) is the high-level overview...
- ... we need some specific details about the tests
- These are the specific test plans
- Installation test (does it install properly?)
- Usability test (is it usable?)
- Security testing (is it secure?)
- Requirements testing (does it meet the requirements?)
- Compability testing (does it work on multiple browsers?)
- More, as necessary
</script></section>
<section data-markdown><script type="text/template">
## Submission requirements
- A brief outline of the specific test plan will be part of the Master Test Plan
- The detailed version (due later) should be combined into a single file (with multiple sections, one for each test type) called `docs/specific-test-plans.md`
- Note that you may not need all of the tests described here
- If you don't need one, then mention so in the specific-test-plans.md file so we know you intentionally didn't include it (as opposed to you forgot it)
</script></section>
<section data-markdown><script type="text/template">
## Due dates
- Preliminary versions of these are due with the master test plan (iteration 9 (February 27th))
- Just a high-level outline so we know which specific test plans you will writing -- you only need to include a paragraph for each specific test plan
- This is *part* of the MTP
- The full versions of the tests are due at the end of iteration 12 (April 10th)
- That's one iteration after development ends
- Some of the tests need not be *compelted* then, even though the test plans are due then
- Completed tests by the end of iteration 13 (April 24th)
- That's two iterations after development ends
</script></section>
<section data-markdown><script type="text/template">
## Installation test
- You are testing that the [installation instructions](02-deliverables.html#/installation) work
- This can be a single paragraph stating:
- What the system is that you will start with
- Perhaps a VirtualBox image without apache installed
- You can assume a package manager such as apt-get
- How you will know if the test works or not
- You can't just say "if it works" -- that's not a measurable goal
- Instead, perhaps you know it works because they can do functions A, B, and C on the system, and the unit tests all pass
- You have to actually run through this test!!!
- Although that is after development ends
</script></section>
<section data-markdown><script type="text/template">
## Usability test
- This is a *subjective* test, not an *objective* test
- This means that you will have to get people's *opinions* as to how usable the system is!
- Who to ask?
- Certainly the customer and the members of your group
- Members of other groups (you will likely have to reciprocate)
- Friends / relatives / significant other(s)
- What you want from them is *opinions* of the usability of the system
- We want honest opinions and feedback, not just "it looks great"
- They will only need to spend 5 (or so) minutes playing with the system; you can help guide them through
</script></section>
<section data-markdown><script type="text/template">
## Usability test plan
- It should include:
- What parts of the system people are going to use
- What questions you will ask them, and what opinions you want to get from them
- Once you get those, you should put them in the plan (please anonymize)
- And then use those opinions to improve your system!
- We are looking for at least 4 people *outside* the group, and at least 2 of those who are not the customer or in this class
- This is in addition to the entire group and the customer
- And this can be long-hand opinions; we aren't looking for formal human subject experiements with Likert scales here
</script></section>
<section data-markdown><script type="text/template">
## Security testing, part 1
- Questions to answer:
- What prevents people from doing actions they should not be allowed to do (deleting things)?
- What about bot-based password login attempts?
- [SQL injection](https://en.wikipedia.org/wiki/SQL_injection) and/or [XSS](https://en.wikipedia.org/wiki/Cross-site_scripting) attacks?
- What about password choice security (i.e., not 'password' as your password)?
- Malformed input testing, described [earlier](#/malformed)
- And lots of others...
</script></section>
<section data-markdown><script type="text/template">
## Security testing, part 2
- These can be a series of unit tests!
- If so, just briefly state which ones they are -- and why they accomplish this goal
- Keep in mind that your framework will handle much of the security for you
- You can also describe a test where people try to log in (or whatever), and ensure the result is the desired one
</script></section>
<section data-markdown><script type="text/template">
## Requirements testing
- Does the system meet all the requirements?
- As described [before](#/reqtest)
- This will likely be a outline as to how you will go through each requirement and ensure that it is properly implemented in the system
- One paragraph is propbably sufficient
- You have to actually run through this test!!!
- Although that is after development ends
</script></section>
<section data-markdown id="compatabilitytesting"><script type="text/template">
## Compatability Testing
- Here is the entire compatability test plan (it assumes that you have completed the requirements test from the previous slide):
> We will execute the requirements test on the most recent version of the major web browsers (Firefox, Chrome, IE, and Safari). If -- and only if -- the requirements test passes on all of those browsers, then the compability test will be considered to have passed.
- You can just cut-and-paste this directly
</script></section>
<section data-markdown><script type="text/template">
## Other test types?
- The specifics of your system may require it
- Load testing
- Recovery testing
- Perhaps others...
</script></section>
</section>
<section>
<section id="codecoverage" data-markdown><script type="text/template">
# Main Code Coverage
</script></section>
<section data-markdown><script type="text/template">
## "Main" Code Coverage
- In this slide set, we will be discussing two types of code coverage:
- The code coverage of your PHP / Python / Ruby code
- So named "main" code coverage, which is this section
- The code coverage of your Javascript client-side code
- So named "Javascript" code coverage, which is the [next section](#/jscodecoverage)
</script></section>
<section data-markdown><script type="text/template">
## Problems with code coverage
Consider the following pseudo-code:
```
if ( foo() && bar() && baz() ) {
// do something...
}
```
- If `foo()` returns false, then by [short circuit evaluation](http://en.wikipedia.org/wiki/Short-circuit_evaluation), `bar()` and `baz()` will not be evaluated
- PHP, Ruby, and Python all perform short-circuit evaluation (as do most programming languages)
</script></section>
<section data-markdown><script type="text/template">
## Code coverage is not absolute!
- Code coverage is just one metric (of many)
- Consider the following (purposely terrible) PHP code:
```
class MyClass {
public function __construct(Array $properties=array()){
foreach($properties as $key => $value)
$this->{$key} = $value;
}
}
function qux($x) {
$a = array( 0 => array('field'=>0),
1 => array('field'=>1),
3 => null);
$var = new MyClass($a[$x]);
echo print_r($var,true) . "\n";
echo $var->field . "\n";
}
```
- What does it do?
- What bugs can we find?
</script></section>
<section data-markdown><script type="text/template">
## Code coverage is not absolute!
```
function qux($x) {
$a = array( 0 => array('field'=>0),
1 => array('field'=>1),
3 => null);
$var = new MyClass($a[$x]);
echo print_r($var,true) . "\n";
echo $var->field . "\n";
}
```
- For full code coverage, we can just run:
```
qux(0);
qux(1);
```
- But there are *multiple* test cases that we don't test, but will fail (for different reasons):
```
qux(-1);
qux(2);
qux(3);
qux(4);
```
</script></section>
<section data-markdown><script type="text/template">
## Code coverage is not absolute!
- Two test cases work, most do not
- But only one of the working ones will generate 100% code coverage (for that function)
- The problem is that we did not put any *thought* into the unit tests
- We did not test array out of bounds (`qux(-1)` and `qux(4)`)
- We did not test null pointer exceptions (`qux(3)`)
- We did not test "holes" in the array (`qux(2)`)
- We did not test boundary cases (`qux(3)` and `qux(4)`)
- The purpose of writing unit tests is to test for all of these conditions!
- Put thought into the unit test design!
</script></section>
<section data-markdown><script type="text/template">
## Code coverage requirements
- Everybody will need to run a code coverage tool
- Suggestions follow for each platform; you can also choose your own
- The output of said code coverage tool must be included in your repo in a `docs/coverage` directory
- And updated regularly!
- You will have code coverage percentage requirements!
- Eventually reaching 100%
- That's 100% of *your* code, not the framework code
- Each of the platform tools follow, with an online example
</script></section>
<section data-markdown id="codecoverage-python"><script type="text/template">
## Python / Django instructions
- This assumes that `python manage.py test` works
- Install [coverage.py](https://pypi.python.org/pypi/coverage) (already done on the course server):
```
sudo apt-get install python-coverage
```
- Note that the binary is `python-coverage`
- The tests must be run *prior* to generating the reports:
```
python-coverage run --source='.' manage.py test
```
- Commands generate the report:
```
python-coverage report
python-coverage report -m
python-coverage html -d coverage
```
- The first is a text-based report; the second is text-based and includes line numbers; the third is an HTML-based report (in a coverage/ sub-directory)
</script></section>
<section data-markdown id="codecoverage-php"><script type="text/template">
## PHP / CakePHP instructions
- This assumes that `phpunit` works
- No additional installs are necessary
- The tests are run *while* generating the reports, not before
- Commands to generate the report:
```
phpunit --coverage-text
phpunit --coverage-text=report.txt
phpunit --coverage-html coverage
```
- The first generates a text-based report to standard output; the second generates a text-based report to `report.txt`; the third creates an HTML-based report in the `coverage/` sub-directory
</script></section>
<section data-markdown id="codecoverage-rails"><script type="text/template">
## Ruby / Rails
- This assumes that `bundle exec rake test` works
- Installation (instructions are duplicated from [here](https://github.com/colszowka/simplecov#getting-started)):
- Install the [simplecov](https://github.com/colszowka/simplecov) gem
- Put in the following into the Gemfile:
```
gem 'simplecov', :require => false, :group => :test
```
- Run `bundle install --path=vendor/bundle`
- Add a few lines to the top of `test/test_helper.rb`:
```
if ENV['RAILS_ENV'] == 'test'
require 'simplecov'
SimpleCov.start 'rails'
puts "required simplecov"
end
```
- Now when you run `bundle exec rake test`, it will generate the HTML report in the `coverage/` sub-directory
</script></section>
<section data-markdown><script type="text/template">
## Links
- Django
- [coverage.py](https://pypi.python.org/pypi/coverage) and the [documentation](http://nedbatchelder.com/code/coverage/)
- [Django testing tutorial for version 1.6](https://docs.djangoproject.com/en/1.6/topics/testing/)
- [The part of that page on code coverage](https://docs.djangoproject.com/en/1.6/topics/testing/advanced/#integration-with-coverage-py)
- CakePHP
- [phpunit](https://phpunit.de/) and the [documentation](https://phpunit.de/manual/current/en/index.html)
- [CakePHP's testing page](http://book.cakephp.org/3.0/en/development/testing.html)
- [The part of that page on code coverage](http://book.cakephp.org/3.0/en/development/testing.html#generating-code-coverage)
- Rails
- [Rails testing page](http://guides.rubyonrails.org/testing.html)
- [simplecov](https://github.com/colszowka/simplecov)
- [The part of that page on installation](https://github.com/colszowka/simplecov#getting-started)
</script></section>
<section data-markdown><script type="text/template">
## Using other tools
- The tools listed here were chosen because they can run on the course server and on your VirtualBox image, and do not require external accounts
- You are welcome to use a different code coverage tool, with the following requirements:
- You need to run it by me first
- It needs to generate the same type of data:
- Percent covered for the entire system
- Percent covered for each individual "part" (file, class, etc.)
- Specific parts not covered (displayed code with line numbers, most likely)
- All this in a nicely formatted HTML-based report (with hyperlinks and what-not)
</script></section>
<section data-markdown><script type="text/template">
## Submission requirements
- Each of the commands listed previous create a `coverage/` sub-directory
- If you use another tool, then an HTML-based copy of the report should likewise be put into a `coverage/` sub-directory
- That `coverage/` sub-directory must go in the `docs/` directory in your repo (i.e., `docs/coverage/`)
- The assumption is that there is a `docs/coverage/index.html` file used to view the report
- And viewing that file will properly view the other HTML-based assets (images, CSS, etc.)
</script></section>
<section data-markdown><script type="text/template">
## Code Coverage Amount
- But the time development ends at the end of iteration 12 (one week into the transition phase), your unit tests must have 100% coverage of your code!
- By the end of iteration 11 (when development ends), we are looking for 80% coverage
- 70% coverage by the end of iteration 10
- This is why we are having you write all those unit tests!
- All of these are *overall* coverage; coverage for individual files may vary (except for the 100% one, of course)
</script></section>
<section data-markdown id="codecoverageslacking"><script type="text/template">
## If <100% code coverage...
- It may be that there is a part of your code that you cannot test, such as from a sanity check
```
if ( the sun explodes )
print "sorry, I cannot continue"
```
- If there is such code in your system, it will prohibit you from hitting 100% code coverage
- In this case, create a docs/code-coverage-excuses.md file that explains for ***each*** code snippet that is not covered:
- What code it is that is not covered
- Why it is not covered
- Why you think it is reasonable that you code will still work
- This is only for code that cannot be tested!
</script></section>
</section>
<section>
<section id="jscodecoverage" data-markdown><script type="text/template">
# Javascript Code Coverage
</script></section>
<section data-markdown><script type="text/template">
## Javascript Code Coverage
- You typically have to run it in a browser, which is more difficult than it should be
- A full tutorial of this setup procedure is in the [uva-slp/tutorials](https://github.com/uva-slp/tutorials) repo, specifically [here](https://github.com/uva-slp/tutorials/blob/master/other/qunit-blanket-phantomjs/index.md)
- There are three parts:
- [QUnit](http://qunitjs.com), which is the unit testing framework
- [blanket.js](http://blanketjs.org/), which is the code coverage tool
- [PhantomJS](http://phantomjs.org/), which is a headless browser that can automate the tests
- The result of this will be a jscoverage.pdf file, which goes in the docs/ directory in your repo
</script></section>
<section data-markdown><script type="text/template">
## Part 1: Set up unit testing
- Your Javascript code should all be in separate .js files
- And not embedded in your HTML pages
- You are going to want to use [QUnit](http://qunitjs.com)
- It works really well with Javascript
- You will create a qunit.html file, based on the example at [QUnit](http://qunitjs.com), which will include all your system's Javascript files
- A sample unit test:
```
QUnit.test( "hello test", function( assert ) {
assert.ok( 1 == "1", "Passed!" );
});
```
- Then write some (real) unit tests!
</script></section>
<section data-markdown><script type="text/template">
## Part 2: Set up the code coverage tool
- We are using [blanket.js](http://blanketjs.org/)
- This is recommended by QUnit, and integrates well with it
- You have to download the blanket.min.js file, and put that in a `<script>` line in your qunit.html file
- But AFTER you include the qunit.js file!
- When you re-run the tests, an "Enable coverage" check box will appear
</script></section>
<section data-markdown><script type="text/template">
## Part 3: Set up PhantomJS
- [PhantomJS](http://phantomjs.org/) is a headless browser, which allows you to run the tests, and save the output
- First, install it: `sudo apt-get install phantomjs`
- Next, download the [run-qunit.js](https://github.com/ariya/phantomjs/blob/master/examples/run-qunit.js) script
- Modify that script, as described in the [tutorial](https://github.com/uva-slp/tutorials/blob/master/other/qunit-blanket-phantomjs/index.md)
- The command to run it is:
```
phantomjs run-qunit.js file://`pwd`/qunit.html\?coverage
```
- Note the backslash!
- This will output a jscoverage.pdf file, which will contain a report of the code coverage
</script></section>
<section data-markdown><script type="text/template">
## Result
This is what the jscoverage.pdf file (and thus the qunit.html web page) looks like:
</script></section>
<section data-markdown><script type="text/template">
## Javascript Code Coverage Schedule
- Iteration 10: 25% of your Javascript code needs to be covered by unit tests
- Iteration 11: 50% of your Javascript code needs to be covered by unit tests
- Iteration 12: 75% of your Javascript code needs to be covered by unit tests
- Iteration 13: 100% of your Javascript code needs to be covered by unit tests
</script></section>
</section>
<section>
<section id="conclusion" data-markdown><script type="text/template">
# Conclusions
</script></section>
<section data-markdown><script type="text/template">
## One other requirement...
- Your unit tests, in addition to running on Travis or Circle, must *also* be able to be run on the course server!
- A formal requirement for iteration 12
</script></section>
<section data-markdown><script type="text/template">
## Acceptance testing
- This is a test (or multiple tests) performed to ensure that the system meets customer specifications
- For our purposes, the 'acceptance test' will be the successful execution of the master test plan
</script></section>
<section data-markdown><script type="text/template">
## Iteration deliverables, part 1
- This is all detailed in the [iteration goals](../uva/iteration-goals.html#/) slide set, and will be mentioned in the [daily announcements](../uva/daily-announcements.html#/)
- All of these are due on the last day (a Monday) of each iteration
- Iteration 7: deployment plan (due Jan 30th)
- Iteration 8: installation instructions (due Feb 13th)
- Iteration 9: master test plan (due Feb 27th)
- Which includes the brief summary of the specific test plans
- Iteration 10: code coverage report (due Mar 13th)
- As well as main code coverage @ 70%
- And Javascript code coverage @ 25%
</script></section>
<section data-markdown><script type="text/template">
## Iteration deliverables, part 2
- Iteration 11: the full transition plan, encompassing all of the above, all edited as necessary (due Mar 27th)
- As well as main code coverage @ 80%
- And Javascript code coverage @ 50%
- Iteration 12: the completed specific test plans (due Apr 10th)
- As well as main code coverage @ 100%
- And Javascript code coverage @ 75%
- Iteration 13: all test completed (due Apr 24th)
- Including the specific test plans completed
- Javascript code coverage @ 100%
- Iteration 14: is project wrap-up, and all testing should be completed before then
</script></section>
<section data-markdown><script type="text/template">
## Test plan standard
- [IEEE 829 (2008): Standard for Software and System Test Documentation](http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=4578383)
- Free if viewed from UVa; it's also in Collab under misc/ as well
- And on [Wikipedia](http://en.wikipedia.org/wiki/IEEE_829)
- We will be following *SOME THINGS* of this standard
- The full standard is way too bloated and just encourages fluff
- This is intended to give you a structure, not to restrict what you do...
</script></section>
<section data-markdown><script type="text/template">
## I don't want fluff!
- If you add verbiage for the sake of adding verbiage, I'll just get really cranky
- It's a waste of everybody's time
- If a section doesn't apply just put a single sentence that this doesn't apply (and perhaps why not)
- That being said, I want a well thought-out master test plan
- So you are going to have to actually read through some of that standard, various Wikipedia pages, etc.
- You can start at [Wikipedia](http://en.wikipedia.org/wiki/Portal:Software_Testing)
</script></section>
<section data-markdown><script type="text/template">
## Iteration deliverables format
- It should be in either Markdown, and kept in the docs/ directory of your github repo
- Name them the following, please:
- docs/master-test-plan.md
- docs/specific-test-plans.md
- docs/coverage/
- etc.
</script></section>
</section>
</div>
</div>
<script src="../slides/reveal.js/lib/js/head.min.js"></script>
<script src="../slides/reveal.js/js/reveal.js"></script>
<script src="../slides/js/settings.js"></script>
</body>
</html>