K&R 1.6 Array. Not understanding the code [closed] - c

Question

Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 6 years ago.
Improve this question
I start reading the book K&R The C programmming ( 2nd edition). And I got stuck on the 1.6 Array; I just can't seem to figure out what the code does (even tho it says it counts digits, white spaces and others!). Here is the code:
#include <stdio.h>
/* count digits, white space, others */
main()
{
int c, i, nwhite, nother;
int ndigit[10];
nwhite = nother = 0;
for (i = 0; i < 10; ++i)
ndigit[i] = 0;
while ((c = getchar()) != EOF)
if (c >= '0' && c <= '9')
++ndigit[c-'0'];
else if (c == ' ' || c == '\n' || c == '\t')
++nwhite;
else
++nother;
printf("digits =");
for (i = 0; i < 10; ++i)
printf(" %d", ndigit[i]);
printf(", white space = %d, other = %d\n",nwhite, nother);
}
So first it defines Integers, ( c,i,nwhite,nother);
After that it creates an array of 10 digits, ( 0 -9 )
After that it sets nwhite and nother to 0.
the for loops set I to 0, i < 10 means if its lower, add i = i + 1.
ndigit[i] = 0? I dont quite understand it, isnt i already is 0?
while ((c = getchar() != EOF) means What ever the input is and isnt at the end of the file?.
After that part I kinda got lost and I'm not sure what
if (c >= '0' && c <= '9')
++ndigit[c-'0'];
Does at all.
And I don't quite understand why the for (i = 0; i < 10 ; +=i ) is repeated . I do understand English but some expensive use of words will confuse me. So if you dont mind, please keep it basic for me. I really hope there is someone out there who can help me understanding this code 100%. Because after all, who wants a programmer who cant even understand the code? :)

Answer 1

Let us step through the code and see what is happening.
main()
{
int c, i, nwhite, nother;
int ndigit[10];
nwhite = nother = 0;
In the first first line of code we are declaring[0] (to the compiler) that c, i, nwhite and nother will be integer variables. At this point, while we have declared these variables, we have not given them any value.
The next line we are declaring that ndigit will be an array of 10 integers, again no initialization is happening so we have no idea of what the value of those ten integers might be.
In the third line we are defining nwhite and nother to be zero, in other words we are initializing them to some value.
for (i = 0; i < 10; ++i)
ndigit[i] = 0;
In this loop, we are initializing the variable i to be zero, and we will increment it by one ever time through the loop, till the value become ten or larger. The body of the loop sets each element of the array to zero. This is a common c-idiom for initializing the elements of an array.
while ((c = getchar()) != EOF)
{
if (c >= '0' && c <= '9')
++ndigit[c-'0'];
else if (c == ' ' || c == '\n' || c == '\t')
++nwhite;
else
++nother;
}
The next block of code does the actual counting. While the code in K&R is syntactically correct, I prefer enclosing the bode of the while loop with curly-braces, I find it easier to read, but it is a personal thing [1].
The condition of the while loop ((c = getchar()) != EOF), can be kind of confusing. We perform the operation in parenthesis first, which is c = getchar() which has the effect of getting the next character and assigning it to the variable c. (remember that in C a character (i.e. variable of type char), is just a small integer so we can assign a character type to an integer type). The assignment statement has a return value[2], in that it returns the value on the right side of the assignment operator, so the operation in parenthesis returns the value of getchar(), which is then compared to EOF, and if it doesn't equal EOF we enter the body of the while statement.
The first if statement checks to see if the character is a number. In ASCII, number have the value of 0x30 ('0') through 0x39 ('9'), so we check to see if the character is in that range. If it is, we increment the appropriate value in the ndigit array. For example, suppose that we have read in the character '5' which has an ASCII value of 0x35. Because 0x35 is between 0x30 and 0x39 we have a digit. Performing the subtraction c - '0' is equivalent to 0x35 - 0x30 which equals 0x05. We then use this as the index into the array, and increment the appropriate value with ++ndigit[c-'0'].
The next branch of the if-block, check to see if c is a a white space, i.e. the
expression c == ' ' || c == '\n' || c == '\t' check to see if c is a space or if c is a new-line or if c is a tab. If c is one of those characters we then
increment nwhite.
Finally, the else branch is taken if we do not have a digit or white space, and we then increment nother.
printf("digits =");
for (i = 0; i < 10; ++i)
printf(" %d", ndigit[i]);
printf(", white space = %d, other = %d\n", nwhite, nother);
}
The last bit of code just prints out the results. Because we want to look at all ten elements of the ndigit array, we need to step through the array again so we use the for loop structure to look at each element of the array.
Hopefully, this clears up some stuff. Something you may want to try is to modify this code so that it counts the of letters that are appearing in the input as well. First just try and count letters, with out regard to case, and then see if you can count upper and lower case letters.
notes:
[0] Declaring a variable is just specifying the name and type of the variable, so int x; is just a declaration. We are providing just enough information to the compiler that it can check our usage of x. A definition is when we assign a value to the variable, so x=5; is a definition. Note that the declaration and definition can be combined into a single line int x = 5;. At the assembly level, a declaration causes storage to be allocated for the variable, but does not set what the storage location contains.
[1] The C grammar says that the curly-braces are not needed for a while block if
it consists of a single statement, i.e.
while(n > 10)
c--;
and
while(n > 10)
{
c--;
}
are equivalent, I just find the second easier to read. Also, the C grammar
says that curly braces are not need for the body of an if statement if the body consists of a single statement, so for example
if(n < 10)
n = n - 10;
and
if(n < 10)
{
n = n - 10;
}
are equivalent.
Finally, the else if and end all are part of the if statement so the statement
if (c >= '0' && c <= '9')
++ndigit[c-'0'];
else if (c == ' ' || c == '\n' || c == '\t')
++nwhite;
else
++nother;
is effectively a single statement, and thus why the curly braces are not needed.
Also, for readability and maintainability I tend to use curly braces with if / else if / else blocks - but again it is a personal think.
[2] The assignment statement has a return value of the left hand side, so a simple expression of a = 10; the return value is just ignored. Having a return value allows us to write something like, a = b = c = 10 which will have the effect of setting a, b and c to 10. In addition of having a return value, the assignment operator is right associative, so the above expression would be
interpreted as a = (b = (c = 10)).
-T.