The
fourth chapter of the global Exploratory testing method
1. Explore the Software
The exploratory test has several goals:
A. Understand how the application works, how his interface looks, what functions he implements
B. Forcing software to demonstrate all of its capabilities
C. Finding defects
2. Tourist analogy
(1) Commercial area test type
A. Guidelines test method (the Guidebook tour)
Travel Brochure-User manual
The Guide testing method requires testers to perform their actions by reading the user manual and strictly following the recommendations of the manual.
This test method not only verifies that the software does realize the various features described in the manual, but also verifies the accuracy of the user manual.
variants of this test method:
* Blog Test (blogger's tour), this method requires testers to follow the recommendations of third parties to test.
* Expert testing, which requires testers to create test cases based on complaints from angry commentators (forums, communities, etc.).
* Competitor Testing (Competitior's tour), a method that requires testers to follow the advice provided by these experts or blogs for competitors.
The Guide testing method (including its various variants) forces testers to test the software features in tandem with the user's use, while also requiring them to interact with each other in the way they are used.
so the flaws found are likely to be extremely important.
B. Selling point test method ( the Money Tour)
any place that tourists aspire to must have a reason to attract them.
The software is similar: users buy it naturally for a reason. If certain features attract users, it is the software's selling point, and the software can make money from that.
sales people are a great source of information for selling test methods.
variants of this test method:
* Questioning the test method (skeptical customer tour), that is, when testers execute the selling point test, imagine a customer asking questions, constantly interrupting the presentation and asking some tough questions.
The flaws found in this test are very important because they may be encountered by real users.
C. Landmark test method (the Landmark tour)
Use the compass to locate a landmark (tree, rock, cliff face, etc.) that is roughly in the direction of the destination, go there, and then determine the next landmark, so back and forth. As long as all the landmarks are in one direction, you can walk through the lush forest.
by using the Guide test method and the selling point test method, you can identify the key software features in advance, which is the landmark here.
Once you've selected the landmarks, you need to determine their order and then go from one landmark to another to explore the application until you have access to all the landmarks in the list.
in this process, you need to record which landmarks have been used and create a landmark overlay to identify the progress of the work.
D. Limit testing method (the intellectual tour)
the approach of the limit test method is to ask the software many difficult questions.
For example:
* How to maximize the software?
* Which feature will be the software to run to its design limits?
* which inputs and data will consume the most software computing power?
* which input may spoof its error detection routines?
* If the software is used to generate certain output, what input and internal data can be used to constantly challenge the ability of the software?
depending on the application under test, the questions raised are different.
variants of this test method:
* Trouble-test method (arrogant American tour)
This approach requires testers to deliberately set a variety of obstacles to see how the software should respond.
this test method and its variants can detect various types of defects, from very important flaws to simple and even foolish flaws. It depends on the ability of the exploratory tester to harness.
testers must be able to differentiate between what is purely a problem and what can make the software work correctly.
in testing, you should try to create complex documents, orders, or other data that are really useful to users, and the result is to let people know that the flaws they find really affect the end user and should be repaired in a timely manner.
E. Late-night testing method (after-hours tour)
variants of this test method:
* early morning test (Morning-commute tour)
F. Traverse test method (The garbage Collector 's tour)
for software, this is like a planned spot check.
we can decide to check the user interface, such as a screen after a screen, a dialog box and a dialog box (preferably can be like a garbage truck driver to use the shortest path to test).
The test does not pursue details to avoid affecting the speed of the test, but only to examine those obvious things.
You can also use this test to perform features and modules, or, if applicable, to test certain landmarks for that particular application.
traverse testing is done by selecting a target (for example, all menu items, all error messages, or all dialog boxes) and then using the shortest path that can be found to access all objects contained by the target.
(2) Historical area test Type
The "Historical area" in the software refers to the legacy code, or the software features that existed in previous versions, or the code used to fix known flaws.
Historical area test methods are mainly for old function and bug fix code.
A. Bad neighbor test (the Bad-neighborhad tour)
every city that is worth traveling has bad neighborhoods, and tourists are often told to avoid visiting them.
The software also has the same situation--that is, the code snippets that are rampant.
The former tries to avoid these areas, while the latter should spend as much time as possible in these areas.
defects are often haunt, so there is a lot of product defects that deserve to be tested repeatedly.
once a code region flaw is identified, it is recommended that the proximity feature be tested with a traverse test to verify that the code that fixes the known flaws does not introduce new flaws.
B. Museum Testing Act (the Museum tour)
the museum, which exhibits antiques, is popular with tourists.
the old-fashioned (legacy code) in the code is also worth the attention of testers.
flipping through the code base, program binaries, or assembly files, just look at their creation time, and it's easy to find legacy code that hasn't moved for a long time.
those old code that is either re-modified, or is put in a new environment without changes, is prone to failure.
Many source code libraries also retain code modification records.
In this test, the tester should identify the legacy code and the old executable, and make sure that they are subjected to the same experience as the new code in the test.
C. Previous version of the test method (the Prior version tour)
if the current version is an update to a previous version, it is important to run all the scenarios and test cases that were supported on the previous version.
this verifies that the functionality that the user is already familiar with is still viable on the new product and works correctly.
If some features are re-implemented or removed from the new version, testers should choose a new version definition method to enter data and use the software.
test cases that are no longer running in the new version should be examined carefully.
(3) Entertainment area test Type
The entertainment area test method helps testers test those ancillary features, rather than mainline features, and ensures that the two features can be combined in a meaningful way.
A. Supporting test method (the supporting Actor tour)
The supporting test method encourages testers to focus on specific features that, while not the main feature that we want users to use, appear on the display together with the main features.
The closer they are to the main functions, the more easily they are noticed, so we must pay enough attention to these features and cannot afford to neglect their mistakes.
B. Shenxiang test method (the Black allery tour)
refers to the most unlikely to be used or the least appealing to the user's features.
The Shenxiang test is to suggest that testers should test the usage list at the bottom of several features.
variants of this test method:* Hybrid test method (Mixed-destination Tour): Try to mix and test the most popular, drunken, and not popular features. c. Overnight test method (the all-night tour)can the application stick to the end? How long can it continue to work, processing data without crashing? testers who use the overnight test will keep the program running without shutting it down.
(4) Tourist area test typeA. Collector's Test method (the Collector ' s tour)We recommend that we collect the output of the software and collect as much as possible. The tester arrives at all those reachable places and records the observed output results. testers should ensure that any output generated by the software can be observed. B. Long-way test (the Lonely businessman Tour)try to navigate the application as much as possible before you reach your destination. c. Supermodel test method (the supermodel tour)
ask testers to care about what's on the surface. The point is not to really interact with the function or test function, but just to test the interface. d. Test-one-send-one-testing method (the Togof tour)
Test one Get one free
test the situation where multiple copies of the same application are running concurrently. E. Scottish Pub test method (the Scottish pub tour)suitable for large-scale complex applications.
(5) hotel area test TypeA. Cancellation test method (the Rained-out tour)start the operation and then stop it. find the most time-consuming operations in your application to fully implement this attack method (such as query power). Cancel button, Esc key, fallback button in program, SHIFT+F4, Close button. In this test, the majority of failures seen by testers are related to insufficient application self-cleaning capabilities. B. The lazy test method (the Couch Potato tour)
Accept all default values.
The software must accept the default value, and it must run the code that handles the blank input.
(6) Dilapidated area test typeA. Destructive testing method (the Saboteur tour)try to use every possible opportunity to sabotage the application. * Forcing the software to do some work* Master the resources that the software must use to successfully complete the operation* To varying degrees remove those resources or restrict the use of those resources B. Rebel Test Act (the Antisocial tour)requires input of the least-likely data, or known malicious input. * Reverse test method (opposite tour): Enter the most unlikely data
* Rogue test method (Crime Spree Tour): Enter some data that should not appear
* ERROR sequencing Test (wrong Turn tour): Doing things in the wrong order
c. Compulsive Disorder Test (the Obsessive-Compulsive tour)perform the same operation repeatedly.
<< Exploratory Software Testing >> fourth Global Exploratory testing method
