change string file? - output

Question

I have an output of a string that is:
I need to print out only the parts that are between"tele": and then the ",". for example the first one i need to print out is hel. In the above string output. "text":"hel","source": is the first part. Every where there is a "tele": and the "," i need to be printed.
the code i have written so far prints out only the first one and is shown below
int comm_poss;
string econ;
tax= mes.find("tele");
econ=mes.substr(tax,15);
cout<<"msg test : "<<econ;
This prints out tele":"hel"," but not the other part that is between the "tele" and ","
how do i print only whats between those two and make the code do it multiple times?
thanks
EDIT: the other part in the output is "tele":"FIRS","code":" found near the end

Answer 1

Save yourself some trouble and use a JSON parser. For example using JsonCpp library and reading the JSON file from stdin (untested):
Json::Value root;
std::cin >> root;
for (int i = 0; i < root.size(); i++) {
std::string text = root[i]["text"].asString();
std::cout << text << std::endl;
}

Answer 2

Could something like this help.
string txt_test;
size_t poss_text;
poss_text= replyMsg.find("text:\":\""); // Search for text":"
replyMsg=replyMsg.substr(poss_text + 8); // Get rest of string
// Add 8 to get beyond text":"
poss_text= replyMsg.find("\","); // Search for ",
txt_test = replyMsg.substr(0, poss_text); // Get from start to ",
cout<< "msg test : " << txt_test;
You should however always check whether your string were found:
if (poss_text != std::string::npos)
{
// Found - so go on
}
else
{
// Not found - stop now
}
Now you can add the loop:
int main()
{
string replyMsg = "HEJtext:\":\"first\",\"hhhtext:\":\"second\",";
string txt_test;
size_t poss_text = 1;
while(poss_text != std::string::npos) // Loop while there is still something in replyMsg
{
poss_text= replyMsg.find("text:\":\""); // Search for text":"
if (poss_text != std::string::npos)
{
replyMsg=replyMsg.substr(poss_text + 8); // Get rest of string
// Add 8 to get beyond text":"
poss_text= replyMsg.find("\","); // Search for ",
if (poss_text != std::string::npos)
{
txt_test = replyMsg.substr(0, poss_text); // Get from start to ",
cout<< "msg test : " << txt_test << std::endl ;
replyMsg=replyMsg.substr(poss_text); // Prepare for next search
}
}
}
return 0;
}

Answer 3

This is a working solution.
I will not use any library since you probably want to know how to do it from scratch.
1. Using Strtok() with delim(':') and checking if the 6 characters to
the left is equal to "text", If TRUE, then pickup all characters to the right
until you hit ",".
2. When you're picking up characters to the right, to avoid reading past your domain, modify the content string by adding a '$' symbol to the end.
#include <iostream>
void get_text(char * token);
int main(){
char =content[]="586162004425048066\",\"text\":\"hel\",\"source\":...\"text\":\"yo\";$";
char * token;
token=strtok(content,":");
get_text(token);
while(token=strtok(NULL,":")){
get_text(token);
}
}
void get_text(char * token){
unsigned int len=strlen(token);
std::string temp;
if (len>=6){ // Reading the 6 characters to the left and storing it in temp
for (int i=-6; i<0; i++){
temp=temp+token[len+i];
}
}
if (temp=="\"text\""){
temp="";
unsigned i=len;
while(token[i]!=',' and token[i]!='$'){ // retrieve characters to the right until you hit a ',' or '$' (which indicates end of line as a boundary condition)
temp=temp+token[i];
i++;
}
std::cout<<temp;
}
}
OUTPUT
"hey" "yo"