LinuxQuestions.org - New to C++ equivalent Visual Express error detection in g++/gdb ?

- Programming (https://www.linuxquestions.org/questions/programming-9/)

- - New to C++ equivalent Visual Express error detection in g++/gdb ? (https://www.linuxquestions.org/questions/programming-9/new-to-c-equivalent-visual-express-error-detection-in-g-gdb-4175452459/)

New to C++ equivalent Visual Express error detection in g++/gdb ?

MS Visual Express C++ catches goofs such as:

int CLASSA = 80.00; //Cost per A Class ticket.

warning: initalizing: conversion from double to int, possible loss of data

I am looking to have the same thing using g++ but have not seen it in man.

Couldn't test it at the moment, but here's some compiler flags you can try: -W -Wall -Wextra -pedantic

You need -Wconversion, although it doesn't warn for 80.00 because there is no possible loss of precision there. It does for warn for 80.01.

I had been trying -Werror -Wall -Wextra

Thank you.
-Wconversion is what I was looking for. Just too bad it waits until 80.01 instead of 80.00.

I will have to get better at proofreading.

Quote:

Originally Posted by dander062 (Post 4903766)

I had been trying -Werror -Wall -Wextra

Thank you.
-Wconversion is what I was looking for. Just too bad it waits until 80.01 instead of 80.00.

I will have to get better at proofreading.

I am working (was aided by this forum) on what I call strict C++ types. At the moment:

Code:

sergei@amdam2:~/junk/c++_work/strict_types> cat -n main.cxx

    1  #include "Strict_types.hxx"

    2  #include <iostream>

    3

    4

    5  int main()

    6    {

    7    Strict_unsigned uu = 80.0;

    8    // Strict_unsigned u = 1; // error: ?Strict_unsigned::Strict_unsigned(int)? is private

    9    Strict_unsigned u1 = 1u;

    10

    11    std::cerr << "sizeof(Strict_unsigned)=" << sizeof(Strict_unsigned) << "\n";

    12

    13    std::cerr << "u1=" << u1.get() << "\n";

    14

    15    Strict_unsigned u2 = 2u;

    16

    17    u1 = u1 << u2; std::cerr << "u1=" << u1.get() << "\n";

    18

    19    u1 = 16u; std::cerr << "u1=" << u1.get() << "\n";

    20

    21    u1 >>= 2u; std::cerr << "u1=" << u1.get() << "\n";

    22

    23    u1 = u1 << 2u; std::cerr << "u1=" << u1.get() << "\n";

    24

    25    u1 *= (Strict_unsigned)2u; std::cerr << "u1=" << u1.get() << "\n";

    26

    27    return 0;

    28    }

sergei@amdam2:~/junk/c++_work/strict_types> ~/AFSWD/install/gcc-4.5.4/binsh/g++ -Wall -Wextra main.cxx -o main

Strict_types.hxx: In function ‘int main()’:

Strict_types.hxx:2882:5: error: ‘Strict_unsigned::Strict_unsigned(double)’ is private

main.cxx:7:24: error: within this context

sergei@amdam2:~/junk/c++_work/strict_types>

.

I.e. my goal is to be able to avoid exactly errors of such kind.