Musa Al-hassy https://github.

com/alhassy/CCheatSheet July 14, 2019 Conditionals & Assertion-Based Testing

C Language CheatSheet In C, true ≈ non-zero. Form of the conditional:

if (condition)
statementBlock
Adminstriva // The rest is optional.
else
statementBlock
 condition is any expression that returns a numeric value: All numbers are treated
 All C programs consist of a series of functions, which have return values!
as ‘true’, except 0 which is considered ‘false’.
 E.g., Assignment x = v is the function that updates the value of name x with the Asserts are essentially compile-checked comments of user intentions!
value of v then terminates yielding v as a return value.
 Semicolons act as statement terminators —in contrast to English, wherein they assert(e) does nothing when expression e is true; otherwise it gives a message showing
are separators. the filename, function name, line number, and the condition e that failed to be true.
 C is whitespace agnostic: Newlines and arbitrary spaces don’t matter, for the most #include <stdio.h>
part.
 Compound statements are formed from primitive statements by enclosing them in // Disable assertions at compile time by enabling NODEBUG.
{ curly braces }. // #define NDEBUG
◦ Compounds may appear where-ever a single statement is allowed. #include <assert.h>
 All keywords are in lowercase. // assert(n) ≈ if (n) {} else hhTerminate and display error messageii
 C comments do not nest.
int sum(int n)
 Include files personal files by enclosing them in "quotes", use <brackets> for stan- {
dard library files. int total = 0, i = 0;
 All statements must terminate with a semicolon. while (i != n + 1)
total += i, i++;
 Everything must be declared before it can be used —sequential.
 Characters are in-fact aliases for ASCII numerals. return total;
}
printf("’A’ ≈ %d ≡ %s", ’A’, ’A’ == 65? "true" : "false");
printf("\n’Z’ 23 ≈ %c ≡ %d", ’Z’ - 23, ’Z’ - 23); int main ()
{
’A’ ≈ 65 ≡ true
// print-based testing
’Z’ 23 ≈ C ≡ 67
if (1) printf("here"); else printf("there");
 C functions don’t have to return anything if it’s not appropriate. printf("Sum of 0 + 1 + · · · + 99 + 100 = %d", sum(100));
◦ E.g., exit(n) is the function that returns control to the operating system;
passing it an argument n, usually 0 if everything has gone smoothly and 1 // assertion-based testing
if it’s an error exit. Yet this function obviously’ can’t return a value. assert( sum(100) == 5050 );
assert( (1 ? "here" : "there") == "here" );
 You must specify the type of a variable before you can use it.
 A variable name is only meaningful in the curly brackets that define it, and is // Is completely ignored if the #define is enabled.
otherwise meaningless. This is its scope. // assert(0); // Otherwise, this causes a crash.
◦ Whence, the same names can occur in different places to mean different return 0;
things. }
◦ To transfer data between functions one thus uses parameter lists and return
calls. hereSum of 0 + 1 + · · · + 99 + 100 = 5050

 printf, “print formatted”, is a dependently-typed function: The number and type Enforce a particular precedence order by enclosing expressions in parentheses.
of its arguments depends on its first argument, a string. == equals != differs from ! not
◦ The number of occurrences of ‘%’ in the string argument is the number of >= at least <= at most && and
additional arguments the printf takes. > greater than < less than || or

1
 Assignments  That is, numerically % yields remainders; but logically, in C, it expresses the is-
multiple-of relationship.
When x is a number, the shift operations correspond to multiplication and division by
/* Abbreviations */ 2n , respectively.
x ⊕= y ≈ x = x ⊕ y Left Shift x « n append the bit representation of x with n-many 0s
x++ ≈ x += 1 Right Shift x » n throw away n bits from the end of the bit representation of x
--x ≈ x -= 1
The bitwise operators and &, or |, not !, and xor ^ operate at the bit representation of
The increment and decrement, ++/--, operators may precede or follow a name: an item. For example, the ASCII code of a character consists of 7 bits where
 If they follow a name, then their behaviour is executed after the smallest context Bit Function
—e.g., braces or conditional parentheses— in which they occur. 7 0 digit, 1 letter
 When they precede a name, their behaviour is executed before the context in which 6 0 upper case, 1 digit or lower case
they appear. 5 0 for a-o, 1 for digit or p-z
 The order of evaluation is not specified inside a function call and so behaviour Whence, to convert a character to uppercase it suffices to change bit 5 to be a 0 and
varies between compilers. leave the other bits alone. That is, to perform a bitwise and with the binary number
11011111, which corresponds to the decimal number 223.
Avoid using these in complex expressions, unless you know what you’re doing.
// Mask, or hide, bit 6 to be a ‘1’.
#define toLower(c) (c | (1 << 6))
Loops
#define toUpper(c) (c & 223)
Here’s the general form.
#define times10(x) ( (x << 1) + (x << 3)) // Parens matter!
while (condition) // x ⇒ 2·x + 8·x ⇒ 10 · x
statementBlock
How did we know is was 223?
/* Abbreviations */ 0. Ninth bit is on 100000000 1 « 8
/* for loop */ for(A; B; C;) S ≈ A; while(B) S 1. Negate it: Eight ones 11111111 ~(1 « 8)
/* do-while loop */ do S while B ≈ S; while(B) S 2. Sixth bit is on 100000 1 « 5
3. Xor them 11011111 ··· ^ ···
do/while: The conditional is evaluated after the statement has been executed and so the 4. See it as a decimal 223
statement is obeyed at least once, regardless of the truth or falsity of the condition. This
is useful for do once, and possible more operations. Ironically, C has no primitive binary printing utility.

int i = 0;
do printf("%d \n", i++); Floats & Other Types
while (i != 10); //Note the ending semicolon.

 float: Representing huge numbers to tiny fractions.

Arithmetic and Logic  double: Like float, but greater precision.
 int: The integers; it has 2 subtypes; namely short, long, and unsigned.
◦ Short and long allocate less (half) or more (twice) memory, respectively.
. . . and then the different branches of arithmetic —Ambition, Distraction,
Uglification, and Derision. ◦ Unsigned means we omit negative numbers, and therefore have twice as many
non-negative numbers since the sign is no longer a necessary place-holder.
—Alice’s Adventures in Wonderland ? Also useful when negative integers are unreasonable for the current
task.
The modulus operator % gives the remainder of a division.
Type aliases: typedef existingType newName;
printf("rem = 10 %% 3 = %d ⇒ ∃ quot • 10 = 3 × quot + rem ∧
quot = %d", 10 % 3, 10 / 3);
// A named product: String × Z × R having names name, age, height.
rem = 10 % 3 = 1 ⇒ ∃ quot • 10 = 3 × quot + rem ∧ quot = 3 struct entry
 In conditionals, one may see n % d to mean that n % d is true, i.e., is non-zero. {
This expresses that n is a multiple of d. char name[30];

2
 short age; Variables may be used without having values declared! This is akin to buying the land
float height; —computer memory— but not building the house —assigning a value— thereby leaving
}; // Note that this is a statement us with a vacant lot that contains trees, garbage, and whatever was there before you get
there —as is the case in C.
// Alias for readability.
typedef struct entry Person; char c;
int i;
int main() long x;
{ // Exponent form for numbers. float f;
struct entry qasim = {"qasim", 0, 613e-2}; // 613e-2 ⇒ 613 × 10− ² ⇒ 6.13double d;
Person q = qasim;
q.age = 23; // Uninitalised ⇒ have random values.

printf("Hello! My name is %s and I’m %d years old, being %f ft tall", printf("char %c\n",c);
q.name, q.age, q.height); printf("int %i\n",i);
return 0; printf("long %l\n",x);
} printf("float %f\n",f);
printf("double %d\n",d);
Hello! My name is qasim and I’m 23 years old being 6.130000 ft tall
#RESULTS:
char
Forming Numbers using Octal & Hexadecimal
int 32766
long
Numbers that begin with a 0 are interpreted as octal and those beginning with 0x or
float 0.0
0X are hexadecimal. These forms are obtained from binary by grouping using 3 bits and
double 73896
4 bits, respectively. E.g., 129 ≈ 0201 ≈ 0x81 since:
 129 →hbinaryi 10 000 001 →hoctali 2 0 1 Rather than being set to 0, variables automatically obtain the random values in memory
 129 →hbinaryi 1000 0001 →hhexadecimali 8 1 and one should always initialise variables upon declaration.
 The amount of space a variable needs is its size in memory.
Indeed:
 The actual location of the variable in memory is called its address —just like the
int x = 129, y = 0201, z = 0x81, zz = 0X81, b = (x == y) && (x == zz); address on your house or on an ugly vacant lot.
printf("%d ≈ %d ≈ %d; %d", x, y, z, b);  In your computer, memory space is measured in bytes.
129 ≈ 129 ≈ 129; 1 ◦ Not every computer uses the same storage space for its variables.
◦ The sizeof keyword tells us how many bytes a variable, or structure, uses
up.
The Preprocessor ? Both sizeof(x) and sizeof(int) are valid calls.
The preprocessor performs alterations to a program before it is compiled; e.g., the ◦ This keyword is used primarily for manual allocation of memory for new
following ensures the replacement of every occurrence of this(· · · , · · · ) after it with variables we create.
whatever that is. char c;
#define this(arg1, arg2) that int i;
long x;
 All preprocessor commands are preceded by the # symbol. float f;
 One generally defines useful constants or abbreviations this way, and if they have double d;
a collection of such parameters or utilities in some file, then they can copy-paste
them into the current program file by using #include as mentioned before. // Uninitalised ⇒ have random values.

printf("char %i\n",sizeof(c));
Pointers printf("int %i\n",sizeof(i));
printf("Z %i\n",sizeof(int));
printf("long %i\n",sizeof(x));
 A fly is a bug that flies; so a fly flies. printf("float %i\n",sizeof(f));
 Likewise, a pointer points. printf("double %i\n",sizeof(d));

3
 assert( &arr == arr );
#define Length 123 assert( &arr[0] == arr );
#define Type char 1. Here’s some useful conversion laws
Type arr[Length];
*&-inverses p = &x ≡ *p = x
assert( sizeof(arr) == sizeof(Type) * Length );  The lhs lives in the land of addresses and locations, whereas the rhs
 Memory is set aside for variables, even if they’re not used. lives in the land of values.
◦ To avoid hogging up excess memory, programs that need a lot of memory []&-array arr + i = &arr[i]
usually request it a little at a time using malloc —‘m’emory ‘alloc’ation.
 C is a mid-level language that has the capability to examine memory and the From these we obtain:
variables stored there. []*-array *(arr + i) = arr[i] —immediate from the above two laws.
 Why care about where variables are located in memory? *-array *arr = arr[0] —immediate from the above two laws.
Consider sorting an array of complex data, rather than moving the data itself *-array **arr = arr[0][0] —the previous law iterated twice.
around, it is much cheaper to simply sort an array corresponding to their numeric
locations in memory. Tips:
◦ The variables that hold addresses are called pointers.
X Pointers must be declared with the asterisk
 A variable consists of 4 pieces: Name, type, size, and location in memory. X When a pointer is used without its asterisk, it refers to a memory location
◦ The first two we know, since we write them down. X When a pointer is used with its asterisk, it indirectly refers to a value
◦ The third is obtained via the sizeof operator. X To assign a pointer a memory location, prefix the other variable with &
◦ The fourth is obtained by using the “address of” operator &, resulting in a X Pointers are memory addresses and can be displayed as unsigned integers using %u
pointer. ( One may think of ‘&’ as simply projecting the fourth component
from a variable structure. ) int a[30], *p;

A pointer is a variable that holds a memory address. // Make p point to the beginning of the array a, since a is itself a pointer to the
◦ Pointers are represented as numeric locations, but they are not number val- // beginning of the array.
ues. p = a;
 T* is denotes pointers of type T.
// Determine the actual machine address of a variable by predicing it with &.
◦ Pointers are declared like any other variable. int fred;
◦ Conventionally a declaration looks like T *t; —the whitespace around the p = &fred;
‘*’ is completely irrelevant— as a reminder that the expression *t denotes a
value of type T. Huh? // If we need to know exactly where in the machine C had decidede to allocate space
? Without ‘*’, pointers consume and represent memory locations. // to fred, we need only to print out p; whence & is read “address of”.
? With ‘*’, they denote a value of type T. printf("%p", p);

Unless you’re working with pointers to pointers, you do not want to write 0x7ffee36348ac
down *p = &x, which sets the value at the location of p to refer to an address.
Since the name of an array is itself a pointer to the beginning of an array, we have
// Valid declarations arr ≈ &arr[0]
int*p;
int* q; // Find furthest points of ‘int a[N-1]’ that are identical.
int *r; #define N 12
int * s; int a[] = {9,8,1,2,3,4,4,3,2,1,7,6};

printf("p is location %u \n", p); int * p = a, *q = &a[N-1];

while(*p != *q) p++, q--;
char arr[] = "hello";
p = &arr; p = &arr[0]; p = arr; assert(*p == a[2] && *q == a[9]);
printf("p is location %u \n", p);
Before C99, function return types could be omitted with the understanding that the de-
// For arrays, the array name and the address of the array both refer to the address
fault is int. However, main still needs its return type. The following works with many
// of the zeroth element. warnings.

4
f() {return 9; } char surprise [] = "surprise! No more, my friend";
g() {return ’c’; } // Remember that chars are actually numbers. assert(strlen(surprise) == 28);

int main() { printf("%d", f() + g()); return 0; } surprise[8] = ’\0’;

Equal(surprise, "surprise")
108
 string.h contains the strX functions.
 Multi-dimensional arrays: T arr[d1][d2]...[dN].
Arrays and Strings and Things ◦ Just as chars are numbers, multi-dimensional arrays are plain one-
dimensional arrays: The multiple-dimension declaration syntax simply tells
the compiler how to organise the array into (nested) groups and the indexing
 C arrays are homogeneous —all elements are of the same type— and their size is tells it which (nested) group we’re interested in working with.
static —one cannot readily change how many elements they contain.
 T name[N] ⇒ Declares name to be an array of type T consisting of N elements. #define D0 4
#define D1 3
◦ In an initialisation, N may be omitted and is then inferred. #define D2 2
? E.g., int a[] = {2, 4, 6};
? E.g., T a[N] = {x0, ..., xM} for M < N initialises the first M elements int arr [D0][D1][D2] = // 4 groups consisting of 3 rows with 2 items in each row
and the remaining N - M are uninitialised, and may consist of random {
junk. 0, 1, // Items of arr[0][0][]
2, 3, // Items of arr[0][1][]
 Array access and modification is done with square brackets: arr[int_expression]
4, 5, // Items of arr[0][2][]
may be treated as a normal variable wherever it occurs.
 Strings are just single-character arrays. // Items of arr[1][][]
6, 7,
char hello [] = "hello";
8, 9,
char hello1 [] = {’h’, ’e’, ’l’, ’l’, ’o’, ’\0’};
10, 11,
char hello2 [6] = {’h’, ’e’, ’l’, ’l’, ’o’}; // Null byte included automatically.
// Items of arr[2][][]
// The null byte is not considered part of the string,
12, 13,
// since it’s only a terminating marker. 14, 15,
// As such, it’s not counted in the length of the string.
16, 17,
// Strings end in the null byte, 0x00 aka \0; // Items of arr[3][][]
// that’s how C knows where the end of a string is;
18, 19,
// otherwise it’d keep looking for the null byte,
20, 21,
// scanning memory then crashing. 22, 23
};
char nope [] = {’h’, ’e’, ’l’, ’l’, ’o’}; // Just an array of chars, neither size nor \0 at the end to indicate that this is a string.
// Third group’s second row, second item is 15.
// The array name suffices to refer to the whole string,
assert(arr[2][1][1] == 15); // Note: 2*D0 + 1*D1 + 1*D2 ≈ (+ 8 3 2)
// no need to mention the brackets.
assert(arr[0][1][0] == 2);
// Strcmp succeeds when its arguments are different;
// and fails (returning 0) if they’re the same. int nope[2][1] = {2, 4};
#define Equal(s,t) assert(strcmp(s, t) == 0);
assert( *(nope + 0) == 2);
// assert( *(arr + (3*D0 + 2*D1 + 2*D2)) == 15);
Equal(hello, "hello")
Equal(hello1, "hello")
// Anomaly:
Equal(hello2, "hello")
// char words [a][b][c]
// ⇒ has a lines, consisting of b words, where each consists of c many characters.
assert(strcmp(nope, "hello")); // different args.
 The uppercase letters start at ASCII code 65 and lowercase start at code 97.
// C stops looking when the null byte is found!  The compiler sees all characters as numbers.
 The idea of a pointer is central to the C programming philosophy.

5
 ◦ It is pointers to strings, rather than strings themselves, that’re passed around Input –getting things into the machine
in a C program.
 C strings like s = "this" actually, under the hood, are null-terminated arrays
of characters: The name s refers to the address of the first character, the ’t’,  getchar returns the next character input from the keyboard.
with the array being ’t’ → ’h’ → ’i’ → ’s’ → 0, where 0 is not ASCII zero
—whose value is 48— but ASCII null —i.e., all bits set to 0. #include <stdio.h>
◦ Note that null 0 is denoted by ’\0’ and has value 0, i.e., false, so conditions
of the form p != ’\0’ are the same as p. int main()
{
 An array name is a pointer to the beginning of the array.
printf("Please enter a character: ");
◦ Yet, an array name is a constant and you can’t do arithmetic with it. char c = getchar();
int length(char c[]) // A string is a character array printf("You entered: %c", c);
{ printf("\nBye\n");
// While c[l] is not an ASCII null, keep counting until.
int l = 0; return 0;
while( c[l] ) l++; }
return l;
Let’s extend getchar to work on buffers of length, say, 20:
}
#include <stdio.h>
int main()
{ #define BUFFER_LENGTH 10
char str[] = "hello world 0123";
printf("length(“%s”) = %d", str, length(str)); int main()
return 0; {
} printf("Please enter a string, only first %d chars or newline will be read: \n\t\t
length(“hello world 0123”) = 16 BUFFER_LENGTH);
 T *p; ⇒ declare p to be a pointer to elements of type T. char cs[BUFFER_LENGTH];
int keep_reading = 1;
 *p = v ⇒ “put the value of v in the location which p points to” char * p = cs;
We can now rewrite the length function with even less square brackets. while(keep_reading && p < cs + BUFFER_LENGTH)
{*p = getchar(); if (*p == ’\n’) keep_reading = 0; else p++;}
int length(char* c)
{ *p = ’\0’; // Strings are a null-terminated arrays of chars.
char* start = c;
while( *c ) c++; // Local copy of c is affected. printf("\nYou entered: %s", cs);
return c - start; printf("\nBye\n");
}
return 0;
assert(length("hello world") == 11); }