I create the following for truncating a string in java to a new string with a given number of bytes.
String truncatedValue = \"\";
String curren
As noted, Peter Lawrey solution has major performance disadvantage (~3,500msc for 10,000 times), Rex Kerr was much better (~500msc for 10,000 times) but the result not was accurate - it cut much more than it needed (instead of remaining 4000 bytes it remainds 3500 for some example). attached here my solution (~250msc for 10,000 times) assuming that UTF-8 max length char in bytes is 4 (thanks WikiPedia):
public static String cutWord (String word, int dbLimit) throws UnsupportedEncodingException{
double MAX_UTF8_CHAR_LENGTH = 4.0;
if(word.length()>dbLimit){
word = word.substring(0, dbLimit);
}
if(word.length() > dbLimit/MAX_UTF8_CHAR_LENGTH){
int residual=word.getBytes("UTF-8").length-dbLimit;
if(residual>0){
int tempResidual = residual,start, end = word.length();
while(tempResidual > 0){
start = end-((int) Math.ceil((double)tempResidual/MAX_UTF8_CHAR_LENGTH));
tempResidual = tempResidual - word.substring(start,end).getBytes("UTF-8").length;
end=start;
}
word = word.substring(0, end);
}
}
return word;
}