a practical guide to c for intermediate c++ programmer

Comparing C and C++ coding practice in different aspects.

Despite both belonging to C family, C and C++ have some difference regarding coding practice. It would save programmer a lot of time if only the difference are addressed, so that they don’t need to consult the voluminous C manual.

This guide is not intended for C beginners. Readers are assumed to have adequate knowledge and practice on basic C/C++ syntax. This guide is more like a lightweight memo addressing the difference in coding practice for those immigrants who are switching from C++ to C.

Error handling

C++ introduces exception and handling feature.

int foo() {
    int divident = 1;
    int divisor = 0;
    int result = divident / divisor;
    return result;
}

int main() {
    try {
        foo();
    } catch (Exception e) {
        // do some exception handling job.
    }
    return 0;
}

In C there are no exception throwing and catching grammar, instead programmers use if statement to catch existence of errors and do something with it, possibly logging error message to stderr.

int foo() {
    int divident = 1;
    int divisor = 0;
    if (result = divident / divisor) {
        fprintf(stderr, "Division by zero is undefined.");
        //do something to handle the error.
        exit(EXIT_FAILURE);
    }
}

Notice that EXIT_FAILURE is a constant defined in stdlib​ library.

File IO

Main() function

Print statement

Abstract Data Type

Nowadays it’s important for a general-purpose language to provide primitives and specialized constructs that support ADT (Abstract Data Type). At the time when C was first created and entered industry, the paradigm of object-oriented design was not yet pushed into public field. So C doesn’t have built-in support for abstract data type. However, people find some systematic way to simulate ADT on top of existing C language primitives.

typedef struct stackCDT* stackADT;
typedef int elementT

struct stackCDT { elementT data[100]; }
void Push(stackADT s, elementT elm);
elementT Pop(stackADT s);
int IsEmptyStack(stackADT s);

C++ has built-in class feature, that makes object-oriented design officially supported and streamlined.

class stack {
public:
    stack();
    ~stack();
    void push(elementT elm);
    elementT pop();
    isEmptyStack();
private:
    elementT data[100];
};

Primitive types

C++ has bool and string types added on top of C type set.

Boolean type

In C, programmers use int to simulate bool type, usually with 0 denoting false, and 1, or possibly any value other than 0, to denote true.

int IsPositiveNumber() { return a > 0; }

int main() {
    int a = 12;
    if (IsPositiveNumber(a)) printf("%d is a positive number.n", a);
}

String type

In C, there are strings, but they are not actual primitive type that resembles other types like int, float, or double. In fact, C strings are implemented as character array. More specifically, every string literals in C or C++ is an array storing characters, terminating with the last element