nsstring-null-tail-padding.m

//
// Behavior of NSString with NULL character tail-padding.
//
// **I thought this behavior is wrong, but now, I see it's correct as NSString.**
//
// Usage:
//     $ gcc nsstring-null-tail-padding.m -framework Foundation
//     $ ./a.out
//
#import <Foundation/Foundation.h>

@interface Exp : NSObject
@end

@implementation Exp

+ (void)exp1
{
    NSString *str1 = @"abcdefg";
    NSData *data1 = [str1 dataUsingEncoding:NSUTF8StringEncoding];
    NSData *data2 = [NSData dataWithBytes:data1.bytes length:data1.length+4]; // append NULL char
    NSString *str2 = [[NSString alloc] initWithData:data2 encoding:NSUTF8StringEncoding]; // no error
    NSLog(@"str1:%lu: %@", str1.length, str1);    // => str1:7: abcdefg
    NSLog(@"data1:%lu: %@", data1.length, data1); // => data1:7: <61626364 656667>
    NSLog(@"data2:%lu: %@", data2.length, data2); // => data2:11: <61626364 65666700 000000>
    NSLog(@"str2:%lu: %@", str2.length, str2);    // => str2:11: abcdefg
}

+ (NSString *)generatePaddingStringWithString:(NSString *)str paddingSize:(int)paddingSize
{
    NSData *d = [str dataUsingEncoding:NSUTF8StringEncoding];
    d = [NSData dataWithBytes:d.bytes length:d.length+paddingSize];
    return [[NSString alloc] initWithData:d encoding:NSUTF8StringEncoding];
}

+ (void)exp2
{
    NSString *s1 = @"foo";
    NSString *s2 = @"bar";
    NSString *s3 = [s1 stringByReplacingCharactersInRange:NSMakeRange(1, 1) withString:s2];
    NSLog(@"s1:%lu: %@", s1.length, s1);  // => s1:3: foo
    NSLog(@"s2:%lu: %@ ", s2.length, s2); // => s2:3: bar
    NSLog(@"s3:%lu: %@", s3.length, s3);  // => s3:5: fbaro # ok
}

+ (void)exp3
{
    NSString *s1 = [self generatePaddingStringWithString:@"foo" paddingSize:3];
    NSString *s2 = [self generatePaddingStringWithString:@"bar" paddingSize:3];
    NSString *s3 = [s1 stringByReplacingCharactersInRange:NSMakeRange(1, 1) withString:s2];
    NSLog(@"s1:%lu: %@", s1.length, s1);  // => s1:6: foo
    NSLog(@"s2:%lu: %@ ", s2.length, s2); // => s2:6: bar
    NSLog(@"s3:%lu: %@", s3.length, s3);  // => s3:11: fbaro # "fbaro" is expected ---> No, correct behavior as NSString. See `exp4`.
}

+ (void)exp4
{
    NSString *s1 = @"foo\0\0bar";
    NSLog(@"s1:%lu: %@", s1.length, s1);  // => s1:8: foo
    // Oops! This is correct behavior... NSString can handle buffers containing NULL characters.
}

@end

int main (int argc, char** argv) {
    @autoreleasepool {
        [Exp exp1];
        [Exp exp2];
        [Exp exp3];
        [Exp exp4];
    }
    return 0;
}

Environment:

• OS X 10.9.4
• XCode 6.1

$ gcc --version
Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/usr/include/c++/4.2.1
Apple LLVM version 6.0 (clang-600.0.54) (based on LLVM 3.5svn)
Target: x86_64-apple-darwin13.3.0
Thread model: posix