Question

I need to split a phrase into words, digits, punctuation marks, and spaces/tabs. I also would like to preserve the order of things.

NSString *text = [NSString stringWithFormat:@"The 3 quick:\"brown fox, jump's\" over."];

This is the kind of list I need to yield:

['The', ' ', '3', ' ', 'quick, ':', '"', 'brown', ' ', 'fox', ',', ' ', 'jump's', ' ', '.']

Thank you!!

Answer 1

Try out this category I wrote using a NSScanner & NSCharacterSet:

@interface NSString(Splitting)

-(NSArray *) arrayBySeparatingComponentsInCharacterSet:(NSCharacterSet *) charSet;

@end

@implementation NSString(Splitting)

BOOL scanOneCharacterFromSetIntoString(NSScanner *self, NSCharacterSet * charSet, NSString **outStr);
BOOL scanOneCharacterFromSetIntoString(NSScanner *self, NSCharacterSet * charSet, NSString **outStr)
{
    // check for index out of bounds
    NSString *inStr = self.string;

    if (self.scanLocation >= inStr.length)
    {
        return NO;
    }

    unichar ch = [inStr characterAtIndex:self.scanLocation];

    if (![charSet characterIsMember:ch])
    {
        return NO;
    }

    self.scanLocation++;
    if (outStr)
    {
        *outStr = [NSString stringWithCharacters:&ch length:1];
    }

    return YES;
}

-(NSArray *) arrayBySeparatingComponentsInCharacterSet:(NSCharacterSet *)charSet
{
    NSScanner *scanner = [NSScanner scannerWithString:self];
    NSMutableArray *result = [NSMutableArray array];

    NSString *temp = nil;
    while ([scanner scanUpToCharactersFromSet:charSet intoString:&temp] || scanOneCharacterFromSetIntoString(scanner, charSet, &temp)) {;
        [result addObject:temp];

        if ([scanner scanLocation] >= [self length])
        {
            break;
        }

        unichar temp2 = [self characterAtIndex:[scanner scanLocation]];

        if ([charSet characterIsMember:temp2])
        {
            [result addObject:[NSString stringWithFormat:@"%c", temp2]];
            // only update the scan location if the scan was sucessful
            [scanner setScanLocation:[scanner scanLocation] + 1];
        }
    }

    return result;
}

@end

int main (int argc, const char * argv[])
{
    @autoreleasepool {

        NSString *str = @"The 3 quick:\"brown fox, jump's\" over.";
        NSArray *array = [str arrayBySeparatingComponentsInCharacterSet:[NSCharacterSet characterSetWithCharactersInString:@" :\",'."]];
        NSLog(@"%@", array);
    }
}

Should be what you need, just change the Character Set to what you need. Also note that this was compiled with ARC enabled, so it may or may not work properly with memory management in a reference counted environment.