Precision problem of cJSON_AddNumberToObject(); with double and float.

[ironengineer]

Question:
How can I add the double value only with given precision as provided 100.2 instead of 100.200000. Is it possible to generate the expected result below?

(When I gave the same number as float number = 100.2;, it generates: {"number value":100.199997} )
Example code:

double number = 100.2;
cJSON_AddNumberToObject(json, "number value", number);

Expected result:
{"number value":100.2}

Generated result:
{"number value":100.200000}

[kbranigan]

The problem you're talking about is actually a problem with floating point numbers in computers in general. Unfortunately you are actually not able to put the value 0.2 accurately into a floating point number. The only way you can store that value is with a string value. (it's just how floats represent decimal numbers in binary, you can store 0.5, 0.25, 0.125, 0.0625, 0.03125 and so on but there's no way to create precisely 0.2 by adding these values together) Take a look at https://en.wikipedia.org/wiki/Single-precision_floating-point_format or http://www.exploringbinary.com/why-0-point-1-does-not-exist-in-floating-point/

Also, I'm not the author of this repo - you should post to https://github.com/DaveGamble/cJSON if you're interested in talking with the actual author.