Check if the string contains more Unicode code points than a certain number. This is more efficient than counting all code points in the entire string and comparing that number with a threshold. This function may not need to scan the string at all if the length is known (not -1 for NUL-termination) and falls within a certain range, and never needs to count more than 'number+1' code points. Logically equivalent to (u_countChar32(s, length)>number). A Unicode code point may occupy either one or two UChar code units.
Definition at line 1045 of file ustring.c. References FALSE, NULL, TRUE, U16_IS_LEAD, and U16_IS_TRAIL. {
if(number<0) {
return TRUE;
}
if(s==NULL || length<-1) {
return FALSE;
}
if(length==-1) {
/* s is NUL-terminated */
UChar c;
/* count code points until they exceed */
for(;;) {
if((c=*s++)==0) {
return FALSE;
}
if(number==0) {
return TRUE;
}
if(U16_IS_LEAD(c) && U16_IS_TRAIL(*s)) {
++s;
}
--number;
}
} else {
/* length>=0 known */
const UChar *limit;
int32_t maxSupplementary;
/* s contains at least (length+1)/2 code points: <=2 UChars per cp */
if(((length+1)/2)>number) {
return TRUE;
}
/* check if s does not even contain enough UChars */
maxSupplementary=length-number;
if(maxSupplementary<=0) {
return FALSE;
}
/* there are maxSupplementary=length-number more UChars than asked-for code points */
/*
* count code points until they exceed and also check that there are
* no more than maxSupplementary supplementary code points (UChar pairs)
*/
limit=s+length;
for(;;) {
if(s==limit) {
return FALSE;
}
if(number==0) {
return TRUE;
}
if(U16_IS_LEAD(*s++) && s!=limit && U16_IS_TRAIL(*s)) {
++s;
if(--maxSupplementary<=0) {
/* too many pairs - too few code points */
return FALSE;
}
}
--number;
}
}
}
|