SQL Database is dropping non-ascii characters in my C program

cov · 01-17-2024, 02:59 AM

Here's my code:

Code:

MYSQL *sqlconnect=NULL;

void initialize_database(char *dBase, char *user, char *pw)
{
  sqlconnect=mysql_init(sqlconnect);
  if (!mysql_real_connect(sqlconnect, "localhost", "dwgman", "x47Y)CmK^o,tr|EE/:oU", dBase, 0, NULL, 0))
  {
    fprintf(stderr, "Unable to connect to MariaDB server\n");
    exit(1);
  }
  mysql_set_character_set(sqlconnect, "utf8mb4");
}
wchar_t decodeTextCodes(char *ch)
{
  wchar_t uni;
  if(ch[0]){
    uni=atoi(ch);
  }else
    uni=0xA3;
  return uni;
}
void to_usChar(wchar_t c, char *uni)
{
  int i=0;
  if(c<128)
    uni[i++]=c&0xFF;
  else if(c<2047){
    uni[i++]=192+((c>>6)&15);//1100+00+7-8
    uni[i++]=128+(c&63);//10+1-6 LSB
  }else if(c<65536){
    uni[i++]=224+((c>>12)&15);//1110+13-16
    uni[i++]=128+((c>>6)&63);//10+7-12
    uni[i++]=128+(c&63);//10+1-6 LSB
  }else if(c<0x110000){
    uni[i++]=240+((c>>18)&7);//11110+19-21
    uni[i++]=128+((c>>12)&63);//10+13-18
    uni[i++]=128+((c>>6)&63);//10+7-12
    uni[i++]=128+(c&63);//10+1-6 LSB
  }
  uni[i]=0;
}
int main(int argc, char **argv)
{
  int i=0;
  char ch[255];
  char query[255];
  wchar_t c;
  if(argc>3)
    initialize_database(argv[1],argv[2],argv[3]);
  if(argc>4)
    strcpy(ch,argv[4]);
  else
    ch[0]=0;
  sprintf(query,"INSERT INTO text(drawing, eID, txt) VALUES(9,14063,'");
  c=decodeTextCodes(ch);
  wchar2char(c,ch);
  strcat(query,ch);
  i=strlen(query);
  query[i++]='5';
  query[i++]='1';
  query[i++]='7';
  query[i++]='9';
  query[i++]=39;
  query[i++]=41;
  query[i++]=59;
  query[i]=0;
  printf("\n====\n%s\n====\n",query);
  if(argc>3){
    mysql_query(sqlconnect,query);
    mysql_close(sqlconnect);
  }
}

This outputs

Quote:

====
INSERT INTO text(drawing, eID, txt) VALUES(9,14063,'£5179');
====

However, the mariadb drops the '£' character.

Code:

+---------+-------+--------------------+
| drawing | eID   | txt                |
+---------+-------+--------------------+
|       9 | 14063 | 5179               |
+---------+-------+--------------------+

Can anyone see what I'm doing wrong?

NevemTeve · 01-17-2024, 03:57 AM

`to_usChar` is defined, but not used; 'wchar2char` is used, but not defined.

Try this in mysql.exe:

Code:

SELECT drawing,eID,hex(txt) FROM text;

cov · 01-17-2024, 02:30 PM

Cheers!

I've been using a number of different routines to try different things to get it to work. I thought that I'd removed them, but obviously not.

Your suggestion of

Code:

SELECT drawing,eID,hex(txt) FROM text;

yields interesting results:

Code:

+---------+-------+--------------+
| drawing | eID   | hex(txt)     |
+---------+-------+--------------+
|       9 | 14063 | C2A335313739 |
+---------+-------+--------------+

Clearly the unicode character is there.

Thanks, I'll need to look at whether it will disply in my PHP front end.