Printable View
Why I can not read the sentence "The No. is " in the output. Thank you very much.Code:main()
{
int i=0;
char c;
while(1)
{
c='\0';
while(c!=13&&c!=27)
{
c=getch();
printf("%c\n", c);
}
if(c==27)
break;
i++;
printf("The No. is %d\n", i);
}
printf("The end");
}
Because you can. No really: I got
where the funny symbol is ASCII 27, I guess (it showed up as a left-pointing arrow in my command prompt window).Code:T
h
i
s
i
s
a
t
e
s
t
The No. is 1
The end
http://cpwiki.sourceforge.net/Implicit_main
And try avoiding magic numbers. In this code, it's not very easy, but in the future, if you wish representable characters such as 'a' or 'z', just type 'a' or 'z' instead of some number.
I used to think like that, but compilers are now translating character codes like 'a' by using the *current* system encoding which may vary depending on the machine you're using (or you can also switch of encoding on a same machine fortunately). So if you really need to work with the ASCII encoding, it seems reasonable to use the corresponding code directly. I don't think it can be considered as 'magic numbers' as those codes are known and fixed (as for any encoding anyway). Obviously if you are working on a program which is independent of any encoding, I agree it's better to use character litterals instead of raw codes.Quote:
And try avoiding magic numbers. In this code, it's not very easy, but in the future, if you wish representable characters such as 'a' or 'z', just type 'a' or 'z' instead of some number.
Well, this is somewhat true, but not all that much that you're hyping it to be.Quote:
Originally Posted by root4
The whole reasoning behind "magic numbers" aren't that they are dynamic or static but that it makes your code hard to read.Quote:
So if you really need to work with the ASCII encoding, it seems reasonable to use the corresponding code directly. I don't think it can be considered as 'magic numbers' as those codes are known and fixed (as for any encoding anyway). Obviously if you are working on a program which is independent of any encoding, I agree it's better to use character litterals instead of raw codes.
What ascii character is number 63, for example?
And if you consistently use correct variable type and character type, then you won't get undefined behavior.
In C and C++ - there are two types, char and wchar_t, respectively. What they might be on the target machine doesn't matter to the language.
At the very least, put a comment explaining what you're doing with those "magic numbers". This would help other people reading your code, and YOU, months from now when you have to look at the code again.Quote:
Originally Posted by Elysia
Thank you very much.
This is indeed one possible definition, I would rather consider magic numbers as undocumented/arbitrary values used in a program.Quote:
The whole reasoning behind "magic numbers" aren't that they are dynamic or static but that it makes your code hard to read.
Why undefined behavior? this is not undefined, the standard is explicit on this point. Types don't change anything as this is a semantics problem. Whatever type you use "c='a'" is not guaranteed to result in the same value on a different machine (or on the same machine with a different configuration). So if the program expects a precise encoding, I simply say this is not a good solution. But you're right for the readability issue, using raw codes is not a good idea, constants (or equivalent) are better in this case.Quote:
And if you consistently use correct variable type and character type, then you won't get undefined behavior.
Yes, a likely definition.Quote:
Originally Posted by root4
But the point is, that if we want a 'c', we want a 'c'. Using magic numbers, or static numbers will provide hard-to-read code and may result in undefined behavior on systems where 'c' isn't the same as on the other system you hard-coded the value on.Quote:
Why undefined behavior? this is not undefined, the standard is explicit on this point. Types don't change anything as this is a semantics problem. Whatever type you use "c='a'" is not guaranteed to result in the same value on a different machine (or on the same machine with a different configuration). So if the program expects a precise encoding, I simply say this is not a good solution. But you're right for the readability issue, using raw codes is not a good idea, constants (or equivalent) are better in this case.
So by forcing the use of 'c', we will get 'c' wherever the program runs.
If we're talking about something else - something not about displaying data, then yes, you should probably use "magic numbers." But not when displaying or working with data representable to the user.
Ok, that concludes the discussion.Quote:
If we're talking about something else - something not about displaying data, then yes, you should probably use "magic numbers." But not when displaying or working with data representable to the user.