Project 2 is now on line

Discussion:

(too old to reply)

Andrew Kowpak

2006-03-09 17:32:38 UTC

The project document says that test effor estimation was going to be
discussed in class. Was this done yesterday? I can't seem to remember
hearing about it nor can anyone I've asked. Did I just miss something or is
this still coming?

Thanks,
Andrew

Hi all,
The description for Project 2 is now online under Section IV (Testing).
We can discuss it in the next class on Wednesday.
Greetings,
Kostas Kontogiannis

Andrew L Roth

2006-03-09 17:44:35 UTC

Permalink

Hi all,

After reading through Project2-description-v2.doc, I found it overall
pretty clear, but I have a few questions.

I will use PDD for Project2-description-v2.doc, TPD for Test Plan
Document and TED for Test Execution Document.

1) TPD: System Test Cases (Bottom of pg 1 of PDD)

- What test cases are we using? Are we supposed to come up with them,
or should we use the 5 listed in 3.2 of PDD?

- It says to see the scenario template. Should we use that template?
If so, how is this different from the TED section System Testing section,
which also calls for use of that template?

2) TPD: Traceability Matrix (pg1 PDD)

- Where can I find more information about this? What actually goes in
the matrix?

3) In the template given on pg 4 of the PDD, the third row, "Required ID"
has "add related requirement ID from SRS". Should SRS be TPD? If not,
can you show me where in the SRS we had requirement IDs?

4) TED: System testing, test 4. (pg3 PDD)

"Adminsitrator revokes receive privileges of the forwardee."

- Our group intentionally never stop incoming calls. Even if a phone is
suspended, it can still receive calls. The rationale is that the caller
is charged. (A suspended phone can't call out). Should we make
suspended phones unable to receive calls?

- I suggest an alternative test: "Administrator revokes calling
privileges of the forwarder".

5) TED: System testing, test 5. (pg3 PDD)

"Exercise the correct operation of double forwarding B forwards to C and
C forwards to D, and A calls B."

- Just to be clear, this means that when A calls B, it forwards to C
(and no more forwarding!). Our forward systems are single-hop, so "double
forwarding" seems kind of misleading.

-Andrew

Hi all,
The description for Project 2 is now online under Section IV (Testing).
We can discuss it in the next class on Wednesday.
Greetings,
Kostas Kontogiannis

Kostas Kontogiannis

2006-03-10 16:38:22 UTC

Permalink

Hi Andrew,

Please fine below some clarifications on your questions.

You have to come up with them. However 5 of those are already given as
you have correctly noted on section 3.2. The total number of test cases
should be between 12-15 depending on the specifics of your architecture.

- It says to see the scenario template. Should we use that template?
If so, how is this different from the TED section System Testing
section, which also calls for use of that template?

It will be the same template both in the TPD and the TED.

2) TPD: Traceability Matrix (pg1 PDD)
- Where can I find more information about this? What actually goes in
the matrix?

I will give examples in the class on Monday. Essentially is a matrix
TC1 TC2 TC3 ..... TCn
RR1 X
FR2 X

where TC is a tets case and FR is the Functional Requirement the TC
exercises or tests. X marks denote that the specific test case exercises
a specific functional requirement.

3) In the template given on pg 4 of the PDD, the third row, "Required
ID" has "add related requirement ID from SRS". Should SRS be TPD? If
not, can you show me where in the SRS we had requirement IDs?

Wiil be the same as in the TPD. However, in the TPD has to come form
somewhere. This can be the Functional Requirement ID you had in the SRS
for the extensions. If you did not use IDs for the FRs in the extension
SRS document then you can introduce them here so you can fill this entry.

4) TED: System testing, test 4. (pg3 PDD)
"Adminsitrator revokes receive privileges of the forwardee."
- Our group intentionally never stop incoming calls. Even if a phone
is suspended, it can still receive calls. The rationale is that the
caller is charged. (A suspended phone can't call out). Should we make
suspended phones unable to receive calls?

You can fill the assumptions section, to denote this issue. For CF
though it is implicit that the received call will be an outgoing call. I
guess it is a good idea not to allow setting forwarding at all to this
phone or suspend the FC when an incoming call arrives.

- I suggest an alternative test: "Administrator revokes calling
privileges of the forwarder".

Hmmm. This may be a good one. I do not though like to change the
description now. You can try it though!

5) TED: System testing, test 5. (pg3 PDD)
"Exercise the correct operation of double forwarding B forwards to C
and C forwards to D, and A calls B."
- Just to be clear, this means that when A calls B, it forwards to C
(and no more forwarding!). Our forward systems are single-hop, so
"double forwarding" seems kind of misleading.

That's correct. The spec has one ho only, so you test that there will be
only one hop. There may be groups though that have experimented with
more than one hop so for them may make sense.

-Andrew

I hope these help.

Greetings,

Kostas Kontogiannis

Hi all,
The description for Project 2 is now online under Section IV (Testing).
We can discuss it in the next class on Wednesday.
Greetings,
Kostas Kontogiannis

Chris Kam Ming Chui

2006-03-10 22:41:32 UTC

Permalink

Hi,

Post by Kostas Kontogiannis

Post by Andrew L Roth
1) TPD: System Test Cases (Bottom of pg 1 of PDD)
- What test cases are we using? Are we supposed to come up with them,
or should we use the 5 listed in 3.2 of PDD?

You have to come up with them. However 5 of those are already given as
you have correctly noted on section 3.2. The total number of test
cases should be between 12-15 depending on the specifics of your
architecture.

The question is regarding System Test section, right? Why is there 12-15
test cases? I thought we only need 5 test cases in total? According to
the beginning of Section 3.2 in the Project Desc:

"For this part of the project you will perform system testing for the
following five scenarios"

And near the end of the section:

"For the purpose of this project, you are required to implement one system
test per scenario, although in general, the mapping between a scenario and
a test case is one-to-many."

I suppose that means we only need 5 test cases in total to exercise one
test case per scenario? Or is the prof suggesting that we will probably
need more than the minimal 5 cases?

Thanks,

Chris

Post by Kostas Kontogiannis
Hi Andrew,
Please fine below some clarifications on your questions.

Post by Andrew L Roth
Hi all,
After reading through Project2-description-v2.doc, I found it overall
pretty clear, but I have a few questions.
I will use PDD for Project2-description-v2.doc, TPD for Test Plan Document
and TED for Test Execution Document.
1) TPD: System Test Cases (Bottom of pg 1 of PDD)
- What test cases are we using? Are we supposed to come up with them, or
should we use the 5 listed in 3.2 of PDD?

You have to come up with them. However 5 of those are already given as you
have correctly noted on section 3.2. The total number of test cases should be
between 12-15 depending on the specifics of your architecture.

Post by Andrew L Roth
- It says to see the scenario template. Should we use that template? If
so, how is this different from the TED section System Testing section,
which also calls for use of that template?

It will be the same template both in the TPD and the TED.

Post by Andrew L Roth
2) TPD: Traceability Matrix (pg1 PDD)
- Where can I find more information about this? What actually goes in
the matrix?

I will give examples in the class on Monday. Essentially is a matrix
TC1 TC2 TC3 ..... TCn
RR1 X
FR2 X
where TC is a tets case and FR is the Functional Requirement the TC exercises
or tests. X marks denote that the specific test case exercises a specific
functional requirement.

Post by Andrew L Roth
3) In the template given on pg 4 of the PDD, the third row, "Required ID"
has "add related requirement ID from SRS". Should SRS be TPD? If not, can
you show me where in the SRS we had requirement IDs?

Wiil be the same as in the TPD. However, in the TPD has to come form
somewhere. This can be the Functional Requirement ID you had in the SRS for
the extensions. If you did not use IDs for the FRs in the extension
SRS document then you can introduce them here so you can fill this entry.

Post by Andrew L Roth
4) TED: System testing, test 4. (pg3 PDD)
"Adminsitrator revokes receive privileges of the forwardee."
- Our group intentionally never stop incoming calls. Even if a phone is
suspended, it can still receive calls. The rationale is that the caller is
charged. (A suspended phone can't call out). Should we make suspended
phones unable to receive calls?

You can fill the assumptions section, to denote this issue. For CF though
it is implicit that the received call will be an outgoing call. I guess it is
a good idea not to allow setting forwarding at all to this phone or suspend
the FC when an incoming call arrives.

Post by Andrew L Roth
- I suggest an alternative test: "Administrator revokes calling
privileges of the forwarder".

Hmmm. This may be a good one. I do not though like to change the description
now. You can try it though!

Post by Andrew L Roth
5) TED: System testing, test 5. (pg3 PDD)
"Exercise the correct operation of double forwarding B forwards to C and
C forwards to D, and A calls B."
- Just to be clear, this means that when A calls B, it forwards to C
(and no more forwarding!). Our forward systems are single-hop, so "double
forwarding" seems kind of misleading.

That's correct. The spec has one ho only, so you test that there will be
only one hop. There may be groups though that have experimented with more
than one hop so for them may make sense.

Post by Andrew L Roth
-Andrew

I hope these help.
Greetings,
Kostas Kontogiannis

Post by Andrew L Roth

Hi all,
The description for Project 2 is now online under Section IV (Testing).
We can discuss it in the next class on Wednesday.
Greetings,
Kostas Kontogiannis

Andrew Jacob Anthony Kowpak

2006-03-11 02:21:14 UTC

Permalink

Hi,

Can you please clarify the following response. I went back and looked
at our SRS for the extensions and I cannot find anything that corresponds
to a Functional Requirement ID. The closest I can find are Use Cases, but
the encapsulate several functional requirements. I plan to do as you
suggested and introduce the Functional Requirement IDs in this document,
but, I'm still wondering where they should be located in our SRS.

Thanks,
Andrew

Post by Kostas Kontogiannis

Wiil be the same as in the TPD. However, in the TPD has to come form
somewhere. This can be the Functional Requirement ID you had in the SRS for
the extensions. If you did not use IDs for the FRs in the extension
SRS document then you can introduce them here so you can fill this entry.

Andrew Jacob Anthony Kowpak

2006-03-18 22:26:57 UTC

Permalink

Can somebody please clarify this. I still don't understand the differance
between a functional requirement id and a feature id. Because the project
is due in just over a week, I'm going to work under the assumption that
they can be used the same way (ie. id's in section 2 can be used in the
Requirement ID entry in section 2.1). Can somebody please verify this
assumption? If the assumption is incorrect, can you please let us know
what the differance is? This is going to be a busy week for my group
members, and if we need to change the work we've done, we'd like to
clearly understand what we should be doing.

Thanks,
Andrew

Hi,
Can you please clarify the following response. I went back and looked at
our SRS for the extensions and I cannot find anything that corresponds to a
Functional Requirement ID. The closest I can find are Use Cases, but the
encapsulate several functional requirements. I plan to do as you suggested
and introduce the Functional Requirement IDs in this document, but, I'm still
wondering where they should be located in our SRS.
Thanks,
Andrew

Post by Kostas Kontogiannis

Post by Andrew L Roth
3) In the template given on pg 4 of the PDD, the third row, "Required ID"
has "add related requirement ID from SRS". Should SRS be TPD? If not,
can you show me where in the SRS we had requirement IDs?

Wiil be the same as in the TPD. However, in the TPD has to come form
somewhere. This can be the Functional Requirement ID you had in the SRS for
the extensions. If you did not use IDs for the FRs in the extension
SRS document then you can introduce them here so you can fill this entry.

Kostas Kontogiannis

2006-03-10 16:43:54 UTC

Permalink

HI Andrew,

No you have nit missed it. We had a fire alarm going off and we did not
go on that. We will cover it on Monday.

Greetings,

KK

Post by Andrew Kowpak
The project document says that test effor estimation was going to be
discussed in class. Was this done yesterday? I can't seem to remember
hearing about it nor can anyone I've asked. Did I just miss something or is
this still coming?
Thanks,
Andrew

Hi all,
The description for Project 2 is now online under Section IV (Testing).
We can discuss it in the next class on Wednesday.
Greetings,
Kostas Kontogiannis

Robert Kitts

2006-03-12 08:28:27 UTC

Permalink

Hi,

In Section 3.1.2 it says we should use "Condition Coverage" for our test
cases. I'm not sure I am understanding Condition Coverage correctly.

Consider the following simplified program. A, B, and C are all single
valued predicates, and all "stmts" have no loops or decisions:

if (A) {
//stmts 1
} else {
//stmts 2
if (B) {
//stmts 3
} else {
//stmts 4
if (C) {
//stmts 5
} else {
//stmts 6
}
//stmts 7
}
//stmts 8
}
//stmts 9

Now, using Condition Coverage, would we have only 2 test cases: A=T, B=T,
C=T; A=F, B=F, C=F? This ensures that each condition in a
compound decsion took both values (slide 18, lecture 10). Even if A was
(D || E), we could continue to just use (D=T, E=T; D=F, E=F) to ensure
that the conditions take each value at least once.

This is only 2 test cases. Is this proper use of Condition Coverage? And
if so, how do we generate more than two test cases using only this
criterion, given that we need 5?

As a further question, if we look at feasable paths, there are only 4
(1->9; 2->3->8->9; 2->4->5->7->8->9; 2->4->6->7->8->9). So how can a
method such as this, which has Cyclomatic Complexity 4, which is within
the recommended range of 3-6, generate 5 different test cases in any case?

Thank you,
Robert Kitts