SPONSORS:






User Tag List

Thanks Thanks:  0
Likes Likes:  0
Dislikes Dislikes:  0
Results 1 to 9 of 9
  1. #1
    Junior Member
    Join Date
    Nov 2002
    Location
    Bangalore, Karnataka, India
    Posts
    3
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    How can i design test cases?

    Hi All,

    This must be a repeated question. I checked the answers for this question. But need some more clarifications.

    I read some of the answers for this question. It says, first write down the test condition, out of that prepare test cases. The following test case design techniques can be used to desing a test case.

    1. Equivalence Class
    2. Boundary Value Analysis
    3. Cause Effect Graph
    4. Error Guessing

    Can any body please tell me, if i get a test condition
    1. How many test cases i can write on that? Is there any particular calculation rule for that?

    2. Taking an example can anybody tell me how the above test case desing techniques can be implemented?

    3. How can i ensure that the test cases are good enough to find all the bugs?

    Expecting your thoughts regarding this and thanks in advance.

    Regards
    Bobby


    ------------------

  2. #2
    Senior Member
    Join Date
    May 2001
    Location
    Ankh-Morpork
    Posts
    2,882
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: How can i design test cases?

    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Can any body please tell me, if i get a test condition
    1. How many test cases i can write on that? Is there any particular calculation rule for that?<HR></BLOCKQUOTE>
    1 <= nbr_test_cases <= infinity
    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>2. Taking an example can anybody tell me how the above test case desing techniques can be implemented?<HR></BLOCKQUOTE>
    Not without taking a lot more time than I have available right now. I'd suggest going to www.QABooks.com and looking for a basic software testing book.
    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>3. How can i ensure that the test cases are good enough to find all the bugs?<HR></BLOCKQUOTE>
    Essentially, you cannot. There have been mathematical studies done which claim the number of test cases required to validate that any program other than the simplest (E.g.: a "Hello, world" program) are very large numbers ("astronomical" comes to mind).

    Equivalence class, boundary test, pairwise combinatorial sets, etc. are all techniques designed to give you the best chance of detecting bugs while keeping the number of test cases to a reasonable minimum.

    ------------------
    Charles Reace
    charlesDOTreaceATverizonDOTnet

    "Any sufficiently advanced bug is indistinguishable from a feature." -Bruce Brown
    web site | [url=http://www.ebookworm.us/[/url]

    [i]...Sound trumpets! Every trumpet in the host! / Sixty thousand, on these words, sound, so high the mountains sound, and the valleys resound.&lt;/i] (The Song of Roland)

  3. #3
    Junior Member
    Join Date
    Oct 2002
    Location
    Rochester hills, MI, US
    Posts
    10
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: How can i design test cases?

    The bottom line for testing is to break the code in all possible ways.
    To break an application understand thoroughly what it is supposed to do and perform both Positive and Negative test case scenarios.
    Coming to the number of test cases, there is no limit. It only depends on the scope of the functionality.
    Idealy if you try to maintain "One Objective - One Testcase" it will you give you a quick idea of what you all you have covered and defect tracking will also be simplified.
    Hope this helps you.


    ------------------


    [This message has been edited by siar (edited 11-18-2002).]

  4. #4
    Junior Member
    Join Date
    Nov 2002
    Location
    Bangalore, Karnataka, India
    Posts
    3
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: How can i design test cases?

    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by Charles Reace:
    Can any body please tell me, if i get a test condition
    1. How many test cases i can write on that? Is there any particular calculation rule for that?<HR></BLOCKQUOTE>
    1 <= nbr_test_cases <= infinity
    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>2. Taking an example can anybody tell me how the above test case desing techniques can be implemented?<HR></BLOCKQUOTE>
    Not without taking a lot more time than I have available right now. I'd suggest going to www.QABooks.com and looking for a basic software testing book.
    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>3. How can i ensure that the test cases are good enough to find all the bugs?<HR></BLOCKQUOTE>
    Essentially, you cannot. There have been mathematical studies done which claim the number of test cases required to validate that any program other than the simplest (E.g.: a "Hello, world" program) are very large numbers ("astronomical" comes to mind).

    Equivalence class, boundary test, pairwise combinatorial sets, etc. are all techniques designed to give you the best chance of detecting bugs while keeping the number of test cases to a reasonable minimum.
    Thanks for the answer Charles Reace. Let me put down my question in other way.

    Recently i had a client interview. Through the course of the interview, the interviewer asked me to tell all the possible test cases for a simple scenario.

    Scenario: There is door for a room and there is senson on top of the door. Whenever a person stands infront of the door, it should open automatically. Once he moves away from the door it should close.

    1. Using all the test case design techniques, how many test cases we can write?
    I don't think that we can write infinite test cases for the above scenario.

    2. Broadly we can say that, we can test it with all +ve and -ve test data. But i want to know what are the +ve and -ve test data?

    I found his question a quite relevent one to test a person to find his analytical and thinking capabilities.

    Hope some of you can help me in this regard.

    Thanks
    Bobby



    ------------------

  5. #5
    Junior Member
    Join Date
    Oct 2002
    Location
    Rochester hills, MI, US
    Posts
    10
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: How can i design test cases?

    Top of my head thise are the follwoing tests i can think of for the Automatic door system.
    1. One Person Standng in front of the door (+ve)
    2. One person passing parallel to the door (+ve)
    3. One person standing right at the point where he is in front of the door but out of range for the sensor, door should not open-boundary testing
    4. Person stading On the other side of the door, door should not open -ve
    5. 2 persons standing on either side of the door, but the door should open +ve

    Doing something the functionality is supposed to do is +ve testing, doing the other way to break it will be -ve testing.


  6. #6
    Senior Member
    Join Date
    May 2001
    Location
    Ankh-Morpork
    Posts
    2,882
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: How can i design test cases?

    <BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by siar:
    Top of my head thise are the follwoing tests i can think of for the Automatic door system.
    1. One Person Standng in front of the door (+ve)
    2. One person passing parallel to the door (+ve)
    3. One person standing right at the point where he is in front of the door but out of range for the sensor, door should not open-boundary testing
    4. Person stading On the other side of the door, door should not open -ve
    5. 2 persons standing on either side of the door, but the door should open +ve

    Doing something the functionality is supposed to do is +ve testing, doing the other way to break it will be -ve testing.

    <HR></BLOCKQUOTE>
    Let's assume the specification is that the door should open once a person is standing withing 3 feet of the front of the door and is within that range for at least 2 seconds (so that it doesn't open for every casual passer-by). Ignoring multiple people, the other side of the door, etc.; we could still have an essentially infinite number of test cases: 2.9' away for 1.1 seconds, 2.8' away for 0.5 seconds, 3.0001' away for 1.999999 seconds, etc....

    The art/science of testing is determining what are the "interesting" test cases, i.e. the ones most likely to find errors (let's not worry about 1 person 5 miles away for 93.2 seconds), while making the total number of test cases a number which can be tested within a reasonable amount of time and other resources. For the one-person test cases with an assumption that the sensor will be accurate within 0.05 feet, boundary testing might suggest we want:

    - one person at 2.9 feet,
    - one person at 3.0 feet,
    - one person at 3.1 feet.

    Assuming the timing is +/- .1 second, we might have this set of boundary tests:

    - one person within range for 1.8 seconds,
    - one person within range for 2 seconds,
    - one person within range for 2.2.seconds.

    Being a pessimist (a.k.a.: a tester), I might add a couple more cases looking for other boundaries, timing issues, or anything else that seems suspicious:

    - one person as close to the door as possible
    - one person within sensor range for 0.5 seconds
    - one person within range for 10 seconds (does the door open and shut several times?)

    Now you start combining the above time and distance cases, perhaps using pairwise combinatorial techniques so that you don't have to test every possible combination, but you will still come up with quite a few test cases if you want to be really thorough. This, of course, begs the question which would need to be answered in your original test plan: just how thorough do you need to be? Are there safety considerations, legal aspects, etc. which indicate you better be as close to 100% sure as possible that nobody will break their nose against a malfunctioning door, or does the door have an alternative opening mechanism making it less crucial that it operate perfectly every time, etc.?



    ------------------
    Charles Reace
    charlesDOTreaceATverizonDOTnet

    "Any sufficiently advanced bug is indistinguishable from a feature." -Bruce Brown
    web site | [url=http://www.ebookworm.us/[/url]

    [i]...Sound trumpets! Every trumpet in the host! / Sixty thousand, on these words, sound, so high the mountains sound, and the valleys resound.&lt;/i] (The Song of Roland)

  7. #7
    Senior Member
    Join Date
    Feb 2001
    Location
    Colorado Springs, CO, USA
    Posts
    864
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: How can i design test cases?

    How does the trigger work? If it is by movement, will it work if the person moves slowly? Will it open fast enough if someone is walking quickly? Will a strong gust of wind be considered movement? What about a bird flying in front of the door?

    How high off the ground is the sensor? Will it trigger for shorter people, like children? Will it notice people in wheelchairs?

    If the door is triggered by weight, what is the limit? Is it so sensitive that rain or hail might open the door? What about a large amount of snow? A cat? A child?

    Will the sensor still open the door after the 100,000th visitor? The 1,000,000th visitor?

    What happens if the door hits something while closing? What happens if the door is triggered while it is closing? What happens if it is triggered while fully open?

    What happens if there is a power outage? What happens if the store is locked down because of a robbery or child abduction? How is the sensor turned off at night, when the store is closed? How would the system know the difference between being locked at night because the store is closed and losing power during the day where people still need to get out?
    Thanks,
    Tim Van Tongeren

  8. #8
    Junior Member
    Join Date
    Nov 2002
    Location
    Bangalore, Karnataka, India
    Posts
    3
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: How can i design test cases?

    Thanks everybody. Now i got an idea of designing test cases.

    ------------------

  9. #9
    SQA Knight
    Join Date
    Jan 2002
    Location
    Highlands Ranch, CO, USA
    Posts
    2,860
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: How can i design test cases?

    The last couple of responses show how adequately detailed specifications can help drive test case development.

    Not that you will always get user requirements / design specifications in this level of detail, but that you need to think beyond the norm and postulate additional information and then create tests to prove the theorums. In other words if needed ask for more details, go talk to the developer or whoever and make sure you have a good understanding of what it is your testing and what your testing for.

    From a boundry value testing scheme I look at it like this:
    7 basic tests will give good coverage.
    1 - Just out of range of sensor
    2 - Right at range of sensor
    3 - Within range of sensor
    4 - Stand right at doorway and do not cross through
    5 - Stand in doorway
    6 - Stand right after doorway and do not continue
    (In the previous 3 let the door smack you in the face, which hopefully it will not, but just in case wear a helmet for this one)
    7 - Normal procession through doorway.

    Good example guys. That is my take on it.

    Jim


    ------------------
    Jim
    -------------------------------------------
    For all the general stuff to know about QA/Test go here http://www.softwareqatest.com/

 

 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Search Engine Optimisation provided by DragonByte SEO v2.0.36 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 10.00%
vBulletin Optimisation provided by vB Optimise v2.6.4 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
User Alert System provided by Advanced User Tagging v3.2.8 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
vBNominate (Lite) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Feedback Buttons provided by Advanced Post Thanks / Like (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Username Changing provided by Username Change (Free) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
BetaSoft Inc.
Digital Point modules: Sphinx-based search
All times are GMT -8. The time now is 11:48 AM.

Copyright BetaSoft Inc.