-
1. Mark all correct statements
Recall not only theory, but also your experience.
-
A. An operating system
- [T]
is a program that controls execution of other programs
- [F]
is one of the most important hardware components in a computer
- [T]
allocates physical resources (CPU time, memory, communication ports, etc.) to
running programs
- [T, usually]
doesn't communicate with users directly
- [F]
... and therefore doesn't need to be foolproof
(i.e. protected from wrong or malicious user's action)
{OS must be foolproof to the fullest, not only against a user's
improper keybord input, but also against hardware failures,
competing and conflicting resource requests,
malicious programs (viruses), network attacks, etc.}
- B.
Which statements comparing C and C++ are correct:
- [F]
A C program must be contained in one file, while
a C++ program may extend to several files.
- [T]
A C++ compiler will compile every compilable
C program.
- [F]
In C++, it is not possible to use the stdio.h
library. A programmer must use
the input/output library iostream.h instead.
- [T]
There is no keyword class in C.
- [F]
C++ doesn't use pointers
- [F]
Programming in C++ is more difficult because a
programmer must use classes. {Using classes is not mandatory in C++.
Also, after some practice, using classes makes programming easier.}
- C.
ASCII code is
- [T]
a specific byte value assigned to each character
- [F]
a specific number of bits assigned to each character
- [F]
a table of escape sequences
- [F]
a special character such as endline (\n), NULL
(\0),
and some others.
- D.
Suppose the variable x is double and
the variable i is int.
- [F]
Both assignments i=x; and x=i;
are valid and don't cause any problems.
- [F]
At least one of them is a syntax error.
- [F]
The assignment i=x; doesn't cause a problem,
while x=i; may lead to a loss of precision due to round-off.
- [T]
The assignment x=i; doesn't cause a problem,
while i=x; may lead to a loss of precision due to round-off.
- [F]
Using explicit casts: x=(double)i; and
i=(int)x; ,
a programmer avoids potential round-off problems.
- E.
How many constructors and destructors can a class have?
- [F]
Any number of constructors and any number of destructors.
- [F]
Same (any) number of constructors and destructors.
- [T]
Any number of constructors but at most one destructor.
- [F]
At least one constructor and exactly one destructor.
- F.
The following lines in a C++ program
ofstream ofile;
ofile.open(argv[0]);
- [F]
make possible input from a file specified in the command line
- [F]
make possible output to a file specified in the command line
{would be True for argv[i] with i>0.}
- [T]
is an attempt to erase the running program's file
{whose name is argv[0]}
- [F (with reservations)]
make possible for the program to alter its own executable code
"on the fly" {Depends on the operating system.
A good OS fails the open operation and enforces the value
TRUE of
ofile.fail() after it.
- G.
What of the following will likely cause "Memory Fault"
error?
- [F]
A variable declared but never used { => compiler's warning}
- [F]
Reference to a variable that wasn't declared { => syntax error}
- [T]
Reference to an array element with negative index
- [T]
Improperly defined size of array
- [F]
Attempt to divide by zero { => Overflow error}
- [F]
Passing a negative value to function sqrt
{ => NAN (not a number) error}
-
2. Evaluate (the answer must be a number):
-
a) 4|2 Ans: 6 {100 | 010 = 110}
b) 5&3 Ans: 1 {101 & 011 = 001}
c) 7%1 Ans: 0
d) 11%4 Ans: 3
e) (x*x==-1) Ans: 0 {false}
f) !(x&&0) Ans: 1 {since
for any x, x&&0=0}
g) 11/4.0 Ans: 2.75
h) 11/4 Ans: 2
(in theory. In practice, the test program
revealed a strange effect, which I can't explain but suspect
it is a deviation from language standards in implementation of
printf.).
-
3. Each of the following lines of code contains some
uncommon (suspicious) feature that may have been overlooked by a programmer.
Identify such a feature.
Match the codes with descriptions of effect. Assume i
is of type int.
-
A
for (i=0; i!=5; ); {i++;}
/* semicolon before '{': empty loop. The
value of i isn't changing, because {i++;} is not a part of the loop. */
B
for (i=i; i!=5; i++) { }
/* initialization i=i has no effect.
Value of i remains uninitialized. */
C
for (i=0; i!=5) {i++;}
/* missing "update" part of for-loop */
D
for (i=0; i!=5; i++) {i++; }
/* i incremented in two places in the loop. It never equals 5 exactly. */
-
Descriptions / answers:
-
[B]
Unpredictable number of iterations
- [A, D]
Infinite loop
- [none]
Loop exits in less than 5 iterations
- [C]
Syntax error
-
4. Trace execution and determine the output of the following
C code.
12 int p,q,a,b,i,j;
13 char* hint="i eat shy"; /* note: spaces are characters too */
14 p=737;
15 q=236;
16 for(i=0; i<4; i++)
17 {
18 a=p/q;
19 b=q;
20 q=p%q;
21 p=b;
22 j=i+(i!=2);
23 printf("%c%c", hint[a-j], hint[a]);
24 }
25 printf("!\n");
-
Output: easy hit!
Test program
line | p | q | a | b | i | j | output | comment
------------------------------------------------------------
13i hint=[i eat shy\0] hint[0]=i, hint[1]=<space> etc.
012345678 9
14,15 | 737 | 236 | ? ? ? ?
16a | | | 0 enter loop
16b | | | (0<4) True
18,19 | 737 | 236 | 3 236
20 | | 29 | 737=236*3+29
21 | 236 | |
22a | | | (i!=2) True=1
22b | 236 | 29 | 3 236 0 1 j=0+1=1
23 | (a-j=2, a=3) ea
---------------------------------------------------------------
16a | | | 1 i=0+1
16b | | | (1<4) True
18,19 | 236 | 29 | 8 29
20 | | 4 | 236=29*8+4
21 | 29 | |
22 | 29 | 4 | 8 29 1 2 j=1+(T)=2
23 | (a-j=6, a=8) sy
---------------------------------------------------------------
16 | | | 2 i=1+1, (2<4) T
18,19 | 29 | 4 | 7 4
20 | | 1 | 29=7*4+1
21 | 4 | |
22 | 4 | 1 | 7 4 2 2 j=2+(F)=2
23 | (a-j=5, a=7) _h
---------------------------------------------------------------
16 | | | 3 i=2+1, (3<4) T
18,19 | 4 | 1 | 4 1
20 | | 0 | 4=4*1+0
21 | 1 | |
22 | 1 | 0 | 4 1 3 4 j=3+(T)=4
23 | (a-j=0, a=4) it
---------------------------------------------------------------
16 | | | 4 i=2+1, (4<4) F
---------------------------------------------------------------
25 | 1 | 0 | 4 1 4 4 !<\n> loop exited
-
5. Determine the output of the following C++ program.
(Full trace is not required.)
1 #include <stdio.h>
2 class MyClass
3 {
4 private:
5 int y,m,d;
6
7 public:
8 MyClass(int a, int b, int c);
9 ~MyClass();
10 int count();
11 };
12
13 int main()
14 {
15 MyClass x(2004,12,11);
16 while (x.count()){}
17 return(0);
18 }
19
20 MyClass::MyClass(int a, int b, int c)
21 {
22 y=a;
23 m=b;
24 printf("Start: %d/%d/%d!\n",y,m,d);
25 d=c;
26 }
27
28 MyClass::~MyClass()
29 {
30 y++;
31 m = m%12+1;
32 d=1;
33 printf("End: %d/%d/%d!\n",y,m,d);
34 }
35
36 int MyClass::count()
37 {
38 printf("%d,",++d);
39 return(31-d);
40 }
-
-
Output: Start: 2004/12/???!
12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,End: 2005/1/1!
(here ??? is a "random" (unitialized) value.)
Test program
-
Argument (instead of full trace).
- Execution begins in line 15,
where the constructor with arguments is invoked.
We jump to line 20, where the values a,b,c become 2004,12,11
respectively. These values are then assigned to class member variables
in lines 22,23,25. The output in line 24
refers to already known values of y, m and to a not
yet initialized value of d. Thus we obtain "Start: 2004/12/???!"
Upon returning from constructor to main(), we proceed to line
16, which is a loop. Every time the member function
count() is invoked and the loop continues while the value received
from count() is nonzero.
Analyse the code of count() in lines 38-39.
The returned value is zero when d=31. How does d change?
It gets incremented in line 38 by 1 each time. The
prefix form i++ of the increment operator is used, therefore the
value of i is first incremented, then printed (followed by comma).
The initial value (from line 25) was d=11, therefore the first printed value
will be 12. The last printed value will be 31.
After the loop in line 16terminates, the program must exit, but
the destructor is automatically invoked in line 17.
We proceed to lines 28-34, where the values become
y=2004+1=2005, m=12%12+1=1, d=1. These are printed. Destructor
returns to main() and the program ends.
-
6. Programming project:
electrostatic potential of a system
of point charges.
[The word electrostatic was misspelled in
exam sheets.]
N electric charges of given magnitudes Q1,..., QN
are located at
points P1,...,PN in space.
Each point is characterized by three given
coordinates (xj, yj, zj), j=1,...,N.
Write a program to find the electrostatic potential
phi(S) at the observation point S with given coordinates (X,Y,Z).
Mathematical definition.
phi(S)=A (Q1 / R1+...
+QN / R N),
where:
A is a physical constant,
Qj is the magnitude of the j-th charge,
Rj=|Pj S|=
Sqrt [ (xj-X)2+
(yj-Y)2+
(zj-Z)2 ]
is the distance from the observation point to the j-th charge.
The potential is undefined (is infinite) when the observation point
coincides with one of the charges.
Physical units.
If the unit of length is meter, the unit of charge is Coulomb,
and the unit of electric potential is Volt, then the numeric value
of the constant A in vacuum is
A=1/(4*Pi*epsilon0)=8.988 x 109
(Volt m)/Coulomb.
You may (if you wish) assume for simplicity that the units are
such that A=1.
Programming notes.
-
You may (if you wish) assume that the physical system is planar,
i.e. all the charges and the observation point lie in the plane z=0.
In that case, the z component of coordinates can be ignored throughout.
-
Users of your program will likely want to evaluate the
potential at various different observation points for the
same system
of charges. Have this in mind when you design the input module of
your program.
-
It is more important that your program be mathematically correct
than for it to have a sophisticated user interface.
Verify your program by a sample calculation.
- Solution:
A program
and a sample input file
kindy provided by Shannon Sullivan
(BTW, it is an example
of a C++ program without classes
-- cf. last item in Q. 1B.)
Here is
a "minimal" passing-guaranteed program