Swift String ranges and NSString ranges are not "compatible".
For example, an emoji like 😄 counts as one Swift character, but as two NSString
characters (a so-called UTF-16 surrogate pair).
Therefore your suggested solution will produce unexpected results if the string
contains such characters. Example:
let text = "😄😄😄Long paragraph saying!"
let textRange = text.startIndex..<text.endIndex
let attributedString = NSMutableAttributedString(string: text)
text.enumerateSubstringsInRange(textRange, options: NSStringEnumerationOptions.ByWords, { (substring, substringRange, enclosingRange, stop) -> () in
let start = distance(text.startIndex, substringRange.startIndex)
let length = distance(substringRange.startIndex, substringRange.endIndex)
let range = NSMakeRange(start, length)
if (substring == "saying") {
attributedString.addAttribute(NSForegroundColorAttributeName, value: NSColor.redColor(), range: range)
}
})
println(attributedString)
Output:
😄😄😄Long paragra{
}ph say{
NSColor = "NSCalibratedRGBColorSpace 1 0 0 1";
}ing!{
}
As you see, "ph say" has been marked with the attribute, not "saying".
Since NS(Mutable)AttributedString ultimately requires an NSString and an NSRange, it is actually
better to convert the given string to NSString first. Then the substringRange
is an NSRange and you don't have to convert the ranges anymore:
let text = "😄😄😄Long paragraph saying!"
let nsText = text as NSString
let textRange = NSMakeRange(0, nsText.length)
let attributedString = NSMutableAttributedString(string: nsText)
nsText.enumerateSubstringsInRange(textRange, options: NSStringEnumerationOptions.ByWords, { (substring, substringRange, enclosingRange, stop) -> () in
if (substring == "saying") {
attributedString.addAttribute(NSForegroundColorAttributeName, value: NSColor.redColor(), range: substringRange)
}
})
println(attributedString)
Output:
😄😄😄Long paragraph {
}saying{
NSColor = "NSCalibratedRGBColorSpace 1 0 0 1";
}!{
}
Update for Swift 2:
let text = "😄😄😄Long paragraph saying!"
let nsText = text as NSString
let textRange = NSMakeRange(0, nsText.length)
let attributedString = NSMutableAttributedString(string: text)
nsText.enumerateSubstringsInRange(textRange, options: .ByWords, usingBlock: {
(substring, substringRange, _, _) in
if (substring == "saying") {
attributedString.addAttribute(NSForegroundColorAttributeName, value: NSColor.redColor(), range: substringRange)
}
})
print(attributedString)
Update for Swift 3:
let text = "😄😄😄Long paragraph saying!"
let nsText = text as NSString
let textRange = NSMakeRange(0, nsText.length)
let attributedString = NSMutableAttributedString(string: text)
nsText.enumerateSubstrings(in: textRange, options: .byWords, using: {
(substring, substringRange, _, _) in
if (substring == "saying") {
attributedString.addAttribute(NSForegroundColorAttributeName, value: NSColor.red, range: substringRange)
}
})
print(attributedString)
Update for Swift 4:
As of Swift 4 (Xcode 9), the Swift standard library
provides method to convert between Range<String.Index> and NSRange.
Converting to NSString is no longer necessary:
let text = "😄😄😄Long paragraph saying!"
let attributedString = NSMutableAttributedString(string: text)
text.enumerateSubstrings(in: text.startIndex..<text.endIndex, options: .byWords) {
(substring, substringRange, _, _) in
if substring == "saying" {
attributedString.addAttribute(.foregroundColor, value: NSColor.red,
range: NSRange(substringRange, in: text))
}
}
print(attributedString)
Here substringRange is a Range<String.Index>, and that is converted to the
corresponding NSRange with
NSRange(substringRange, in: text)
