#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stdbool.h>

/* Basic signs over 127 and lower than 256*/
const char *umap[][2] =
{
    {"#x80","€"},
    {"#x81"," "},
    {"#x82","‚"},
    {"#x83","ƒ"},
    {"#x84","„"},
    {"#x85","…"},
    {"#x86","†"},
    {"#x87","‡"},
    {"#x88","ˆ"},
    {"#x89","‰"},
    {"#x8A","Š"},
    {"#x8B","‹"},
    {"#x8C","Œ"},
    {"#x8D"," "},
    {"#x8E","Ž"},
    {"#x8F"," "},
    {"#x90"," "},
    {"#x91","‘"},
    {"#x92","’"},
    {"#x93","“"},
    {"#x94","”"},
    {"#x95","•"},
    {"#x96","–"},
    {"#x97","—"},
    {"#x98","˜"},
    {"#x99","™"},
    {"#x9A","š"},
    {"#x9B","›"},
    {"#x9C","œ"},
    {"#x9D"," "},
    {"#x9E","ž"},
    {"#x9F","Ÿ"},
    {"nbsp",""},
    {"iexcl","¡"},
    {"cent","¢"},
    {"pound","£"},
    {"curren","¤"},
    {"yen","¥"},
    {"brvbar","¦"},
    {"sect","§"},
    {"uml","¨"},
    {"copy","©"},
    {"ordf","ª"},
    {"laquo","«"},
    {"not","¬"},
    {"shy","­"},
    {"reg","®"},
    {"macr","¯"},
    {"deg","°"},
    {"plusmn","±"},
    {"sup2","²"},
    {"sup3","³"},
    {"acute","´"},
    {"micro","µ"},
    {"para","¶"},
    {"middot","·"},
    {"cedil","¸"},
    {"sup1","¹"},
    {"ordm","º"},
    {"raquo","»"},
    {"frac14","¼"},
    {"frac12","½"},
    {"frac34","¾"},
    {"iquest","¿"},
    {"Agrave","À"},
    {"Aacute","Á"},
    {"Acirc","Â"},
    {"Atilde","Ã"},
    {"Auml","Ä"},
    {"Aring","Å"},
    {"AElig","Æ"},
    {"Ccedil","Ç"},
    {"Egrave","È"},
    {"Eacute","É"},
    {"Ecirc","Ê"},
    {"Euml","Ë"},
    {"Igrave","Ì"},
    {"Iacute","Í"},
    {"Icirc","Î"},
    {"Iuml","Ï"},
    {"ETH","Ð"},
    {"Ntilde","Ñ"},
    {"Ograve","Ò"},
    {"Oacute","Ó"},
    {"Ocirc","Ô"},
    {"Otilde","Õ"},
    {"Ouml","Ö"},
    {"times","×"},
    {"Oslash","Ø"},
    {"Ugrave","Ù"},
    {"Uacute","Ú"},
    {"Ucirc","Û"},
    {"Uuml","Ü"},
    {"Yacute","Ý"},
    {"THORN","Þ"},
    {"szlig","ß"},
    {"agrave","à"},
    {"aacute","á"},
    {"acirc","â"},
    {"atilde","ã"},
    {"auml","ä"},
    {"aring","å"},
    {"aelig","æ"},
    {"ccedil","ç"},
    {"egrave","è"},
    {"eacute","é"},
    {"ecirc","ê"},
    {"euml","ë"},
    {"igrave","ì"},
    {"iacute","í"},
    {"icirc","î"},
    {"iuml","ï"},
    {"eth","ð"},
    {"ntilde","ñ"},
    {"ograve","ò"},
    {"oacute","ó"},
    {"ocirc","ô"},
    {"otilde","õ"},
    {"ouml","ö"},
    {"divide","÷"},
    {"oslash","ø"},
    {"ugrave","ù"},
    {"uacute","ú"},
    {"ucirc","û"},
    {"uuml","ü"},
    {"yacute","ý"},
    {"thorn","þ"},
    {"yuml","ÿ"},
    {"#x2500","─"},
    {"#x2502","│"},
    {"#x250C","┌"},
    {"#x2510","┐"},
    {"#x2514","└"},
    {"#x2518","┘"},
    {"#x251C","├"},
    {"#x2524","┤"},
    {"#x252C","┬"},
    {"#x2534","┴"},
    {"#x253C","┼"},
    {"#x3B1","α"},
    {"#x3B2","β"},
    {"#x3B3","γ"},
    {"#x394","Δ"},
    {"#x2211","∑"},
    {"#x221A","√"},
    {"#x222B","∫"},
    {"#x2248","≈"},
    {"#x2260","≠"},
    {"#x2264","≤"},
    {"#x2265","≥"},
    {"#x2190","←"},
    {"#x2191","↑"},
    {"#x2192","→"},
    {"#x2193","↓"},
    {"#x2194","↔"},
    {"#x2195","↕"},
    {"#x2022","•"},
    {"#x2044","⁄"},
    {"#x2030","‰"},
    {"#x2122","™"},
    {"#x20AC","€"},
    {NULL}
};

char* strip_html_unicode(const char * string)
{
    if(string == NULL) return NULL;

    char *p   = (char*)string;

    // Be safe.
    size_t cpy_len = strlen(string);
    char *cpy = malloc(cpy_len + 1);
    memset(cpy,0,cpy_len+1);

    int x = 0;
    bool is_uc = false;

    while(*p)
    {
        char *sub = " ";
        is_uc = false;

        if(*p == '&')
        {
            int uit = 0;
            char uit_buf[8];
            char *p2 = p;

            // Skip silly "unwarned result" warning
            p2 = p2 + sizeof(char);

            while(*p2 && *p2 != ';')
            {
                uit_buf[uit++] = *p2++;
            }
            uit_buf[uit] = 0;

            if(*p2 == ';' && uit != 0)
            {
                int uma = 0;
                while(umap[uma] && umap[uma][0])
                {
                    if(!strcmp(umap[uma][0],uit_buf))
                    {
                        sub = (char*)umap[uma][1];
                        is_uc = true;
                        p = p2;
                        break;
                    }
                    uma++;
                }
            }
        }

        if(is_uc == false) cpy[x++] = *p;
        else  while(*sub) cpy[x++] = *sub++;

        p = p + sizeof(char);
    }
    cpy[x] = '\0';
    return cpy;
}


int main(int argc, char *argv[])
{
    /* Some human readable formula should be printed. */
    const char *testme  = "f(x) -> &#x2211; &#x222B; &#x221A;x+1  [ &ETH; = { |R0 } ]  (&#x394; &#x2248; 0,11 +/-)";
    char *result = strip_html_unicode(testme);

    puts(result);

    if(result) free(result);
    return EXIT_SUCCESS;
}