Convert mixed JSON-Number-Array to int, uint, float using lib rapidjson

Question

As I understood this char* is a valid json-string.

const char* json = { "array":[14, -15, 3.17], "array_type": ["uint", "int", "float"] }

All numbers in the array shall be 4 bytes.

How could one loop through the array using rapidjson?

This is my code so far:

#include "rapidjson/document.h"

using namespace rapidjson;

int main(void) 
{

int i_val; unsigned int ui_val; float f_val;
const char* json = "{ \"array\":[14, -15, 3.17], \"array_type\" : [\"uint\", \"int\", \"float\"] }";

Document d;

d.Parse<0>(json);
Value& a = d["array"]; 
Value& at = d["array_type"];

for (SizeType i = 0; i<a.Size(); i++)
{
    if (a[i].IsInt() && (std::string)at[i].GetString() == "int")
    {
        i_val = a[i].GetInt();
        //Do anything
    }
    if (a[i].IsUint() && (std::string)at[i].GetString() == "uint")
    {
        ui_val = a[i].GetUint();
        //Do anything
    }
    if (a[i].IsDouble() && (std::string)at[i].GetString() == "float")
    {
        f_val = (float)a[i].GetDouble();
        //Do anything
    }
}//end for

    return 0;
}

ERROR:

TestApp: /workspace/TestApp/src/include/rapidjson/document.h:1277: rapidjson::GenericValue<Encoding, Allocator>& rapidjson::GenericValue<Encoding, Allocator>::operator[](rapidjson::SizeType) [with Encoding = rapidjson::UTF8<>; Allocator = rapidjson::MemoryPoolAllocator<>; rapidjson::SizeType = unsigned int]: Assertion `index < data_.a.size' failed.

The code crashes at the function a.Size(), when it is executed after GetDouble. How could one solve this?

I know the last "if" is wrong and thats probably why the program crashes. Double is 8 bytes and the SizeType is 4 bytes by default.

Is there a solution to loop though the array?? If not I also would be good with other libs.. I need to transport these three different types of values by json. By the way the length of the array could be 1 to 500.

(There is no GetFloat() function.)

Thanks for any kind of help. Regards

score 0 · Accepted Answer · answered Jun 30 '15 at 18:07

0

You can use JSONLint to validate json string. Your example fixed:

int i_val; unsigned int ui_val; float f_val;
const char* json = "{ \"array\":[14, -15, 3.17], \"array_type\" : [\"uint\", \"int\", \"float\"] }";
rapidjson::Document d;

d.Parse<0>(json);
rapidjson::Value& a = d["array"]; 
rapidjson::Value& at = d["array_type"];

for (rapidjson::SizeType i = 0; i<a.Size(); i++)
{
    if (a[i].IsInt() && (std::string)at[i].GetString() == "int")
    {
        i_val = a[i].GetInt();
        //Do anything
    }
    if (a[i].IsUint() && (std::string)at[i].GetString() == "uint")
    {
        ui_val = a[i].GetUint();
        //Do anything
    }
    if (a[i].IsDouble() && (std::string)at[i].GetString() == "float")
    {
        f_val = (float)a[i].GetDouble();
        //Do anything
    }
}

answered Jun 30 '15 at 18:07

doqtor

8,414
2
20
36

Thanks for the answer.. But you havent got the question I guess. – frank schmidt Jun 30 '15 at 19:19
The code is messed up, because of SizeType i. I is 4 byte, but when I use GetDouble the Size is corrupted. – frank schmidt Jun 30 '15 at 19:21
In that case prepare Short, Self Contained, Correct (Compilable), Example, see [link](http://sscce.org/) for details. – doqtor Jun 30 '15 at 19:58
please look at the answer below – frank schmidt Jul 01 '15 at 08:08
@frankschmidt, this is not an answer, move it to your question (you can edit your post and add it beneath) – doqtor Jul 01 '15 at 08:32
@frankschmidt my code and your updated code is working fine in MSVC2013 and gcc 4.8.3. Do you use recent rapidjson? – doqtor Jul 01 '15 at 12:12
Thank you for your help. I solved the problem. The const char* was not json-type. – frank schmidt Jul 02 '15 at 07:34

Convert mixed JSON-Number-Array to int, uint, float using lib rapidjson

1 Answers1