Thread: C Programming code help

Use fgets() to read data from the user as a string of characters. Then check if the content of those characters meet the requirements. Keep doing that until a string is received that meets requirements.

It doesn't necessarily have to be a while loop - any loop construct will do. A do-while loop may be appropriate, as it is will be necessary to read data and check it at least once. For example, you might check for the presence of digits,decimal points, sign ('+' or '-'), and the number of digits after the decimal point. You might also reject input that contains unwanted characters (letters, etc).

Personally, I wouldn't use scanf() for this task at all. And don't make the mistake of trying to use floating point for this.

Originally Posted by grumpy

Use fgets() to read data from the user as a string of characters. Then check if the content of those characters meet the requirements. Keep doing that until a string is received that meets requirements.

It doesn't necessarily have to be a while loop - any loop construct will do. A do-while loop may be appropriate, as it is will be necessary to read data and check it at least once. For example, you might check for the presence of digits,decimal points, sign ('+' or '-'), and the number of digits after the decimal point. You might also reject input that contains unwanted characters (letters, etc).

Personally, I wouldn't use scanf() for this task at all. And don't make the mistake of trying to use floating point for this.

Hi grumpy I'm new to the forum and to C, thanks for your advice. I'll need to do some research on fgets() :] I've never really used the do-while loop so I'll look into that also. Once I get a valid character string input I'll need to do some calculations with it as a float. Any idea what the best way would be to convert the string?

A friend of mine just mentioned something similar to what you said - that I should take the input as a string, search for a "." character, check that the substring preceding the decimal point consists of any number of digits and the substring following the decimal point is exactly four digit characters. And also that any numerical method won't work because for example, 4.2 and 4.2000 are identical, and because computers use binary numbers, not decimal.

OK; what I meant is not to try converting to float BEFORE you've checked the conditions are met. Converting to a float will not enforce your requirements, as your friend has said. Also, a floating point variable cannot accurately represent the value 0.1 either.

Once you've checked the string is valid, then convert it to a floating point if needed.

There are various ways - sscanf(), atof(), strtof, strtod() are all standard functions that you might consider. All have different advantages and disadvantages, but all will work.