|
|
| |
BLACK – BOX TESTING |
| |
Black box testing
attempt to final errors in the following categories:
1. I correct or missing Fns
2. Interface errors.
3. Errors in data structures or external data
base access.
4. Behavior or performance errors.
5. Initialization and termination errors
|
| |
|
|
| |
| I
Graph Based Testing:
First step in black box testing is to understand
the objects that are modeled in s/w and the relationships
that connect their object. Testing begins with
creating a graph of imp. Objects and their relationship
and then devising a series of tests that will
cover the graph so that each object and relationship
is exercised and errors are uncovered.
We start with creating a graph |
| |
Directed
link ----> relations bet. Objects
Node weight ----> attributes of the objects.
Nodes ----> Objects.
Graph based testing begins with the defn of all
nodes and node weights. ie, objects and attributes
are identified. The data model can be used as
a starting point.
Based on this, we can conduct the following behavioral
testing methods:
Transaction flow modelling
Finite state modeling
Data flow modeling.
Now we must go for node coverage and link coverages
to ensure that all objects and their relations
are executed.
|
| |
|
| |
II
Equivalence partitioning:
Here the ip domain is divided in to classes of
data from which test cases can be derived. An
ideal test case is one which single – handedly
uncovers a class of errors.
If the ip class defines a range, one valid and
two invalid classes are defined.
If an ip cond,n req. a spec. value, one valid
and two invalid equivalence classes are defined.
If an ip condn specifies a member of set. One
valid and one invalid equivalence class are defined.
If an ip cond’n is one valid and one invalid
class are defined.
III Boundary Value
Analysis :
Boundary value analysis leads to a selection of
teat cases that exercise bounding values. BVA
leads to the selection of test cases at the edges
of the class. BVA derives test cases from the
output domain.
If ip specifies a range of values bounded by a
& b, then test cases should be designed with
values a and b and just above and below a and
b.
Test care should be designed to create an op report
that produces the max (win) allowable no. of table
entries.
III Boundary Value
Analysis : (Back – to –
back)
And when the reliability of the s/w is very critical.
In such situations, redundant s/w and L/W are
used to minimize the possibility of error. When
redundant s/w is dev, sep. s/w engg. Teams dev.
independent versions of an applies using the same
spec. In such cases each version can be tested
with the same test data to ensure that all provide
identical output.
If the op from each version is the same, it is
assumed that all implementations are correct.
If the op is different, each of the applies is
investigated to det. if a defect is one or more
versions is responsible for the difference.
If the spec. from which all versions have been
dev. is in error. If each of there versions will
provide identical but incorrect results, condn’s
testing will fail to detect the error.
|
| |
|
|
|
|
|
|
Home