How to determine length of a string?

[SudoBen]

I was just confronted with the following question:
How do I determine the count of characters contained in a String?

strlen() does not work, as it returns the len in byte (e.g. strlen("Ä") is 2) .


Any Idea?


best Regards

[Gertjan van Schijndel]

To get a real solution to this problem, you will need to report this to your WinCC OA support contact(s).

As workaround you could iterate over the string with the 'string strwalk(string text, int &pos)' function. The pos argument is updated with the position of the next character.

[fmulder]

strlen() works great and does exactly what the 3.12(!) helpfile says. The helpfile says 'length in bytes'.
What you really want is to know the length in utf characters. Which would have been 1

Unfortunately, there is no function for that (yet?)

When you create a new project then you get the option for 'other language' which includes the 'old scool' Ascii. We ran into the same thing and then decide to stick with Ascii for now.

Although we see the power of the utf characters, we were afraid that it would take too much time to go through our existing libraries.

Good luck !

[SudoBen]

@Frenk: Thats exactly what my Question stated before. But I appreciate that you point out that this function actually IS covered by the help :laugh: :laugh:
Seriously: I can't see why it would be so incredibly hard to add such simplest functionality to the code base. It's like missing functions to check if a String coud be cast to an Integer.... oh... :ohmy:

@Gertjan van Schijndel: Thanks for the help. Didn't know about that function yet. I will have a closer look at it (as i have found no documentation about it).

In the meanwhile I've come up with the following solution :

Code: Select all

//Returns the actual length (amount of characters)of a string
int StrLen(string s)
{
  int clen, len,pos;
  clen =1;// first step
  char c;
  while (clen>0)
  {
    c=s[pos];

    clen=getUTFLength(c);

    if (clen>0)
    {// Character recognized--> count
      len++;
      pos+=clen; //step to next character.
    }    
  }
  return len;
      
}

//Determines the length of an utf 32 character. 
int getUTFLength(char c)
{
  char one = 128;  //10000000
  char two = 192;  //11000000
  char three=224;   //11100000
  char four =240;   //11110000 
  char eos =0;      //00000000

  if ((c & four) ==four)
  {
    return 4;
  } 
  if ((c & three) ==three)
  {
    return 3;
  } 
  if ((c & two) ==two)
  {
    return 2;
  } 
  if ((c & one) ==one) //this is just part of a char
  {
    return -1;
  } 
  if ((c|eos)==eos)
  {
    return 0;
  }
  if (c

[jfily]

Hi guys.

I made it in few seconds, hope it will help.
(not sure it works in all case, but i tried with some non-ascii char)

Code: Select all

int _strlen(string str) {
  int length = 0;  
  for(int i = 0; i < strlen(str); i++) {
    if(str[i] > 0x7F) {
      i++;
    }    
    length++;
  }
  return length;
}

[SudoBen]

Its a really cool approach but, unfortunately your solution does not work. :woohoo:
You get false results with characters that use more than 2 Bytes, as you miss to check if the following bytes belong to the same character (e.g. "€" will return 2 as it is encoded as 3 Byte Char).

I' think its a great approach (as it is shorter and more generic) but you need to fix it to something like :

Code: Select all

int _strlen(string str) {
  int length = 0;  
  for(int i = 0; i < strlen(str); i++) {
   
    while((str[i+1] & 0xC0)==0x80&&i