Fix apple docs links

Author: MartinP7r

Documentation/FullTextSearch.md

@@ -737,11 +737,11 @@ The SQLite built-in tokenizers for [FTS3, FTS4](#fts3-and-fts4-tokenizers) and [
 
 Generally speaking, matches may fail when content and query don't use the same [unicode normalization](http://unicode.org/reports/tr15/). SQLite actually exhibits inconsistent behavior in this regard.
 
-For example, for "aimé" to match "aimé", they better have the same normalization: the NFC "aim\u{00E9}" form may not match its NFD "aime\u{0301}" equivalent. Most strings that you get from Swift, UIKit and Cocoa use NFC, so be careful with NFD inputs (such as strings from the HFS+ file system, or strings that you can't trust like network inputs). Use [String.precomposedStringWithCanonicalMapping](https://developer.apple.com/reference/swift/string/1407210-precomposedstringwithcanonicalma) to turn a string into NFC.
+For example, for "aimé" to match "aimé", they better have the same normalization: the NFC "aim\u{00E9}" form may not match its NFD "aime\u{0301}" equivalent. Most strings that you get from Swift, UIKit and Cocoa use NFC, so be careful with NFD inputs (such as strings from the HFS+ file system, or strings that you can't trust like network inputs). Use [String.precomposedStringWithCanonicalMapping](https://developer.apple.com/documentation/foundation/nsstring/1412645-precomposedstringwithcanonicalma) to turn a string into NFC.
 
-Besides, if you want "fi" to match the ligature "fi" (U+FB01), then you need to normalize your indexed contents and inputs to NFKC or NFKD. Use [String.precomposedStringWithCompatibilityMapping](https://developer.apple.com/reference/swift/string/1407834-precomposedstringwithcompatibili) to turn a string into NFKC.
+Besides, if you want "fi" to match the ligature "fi" (U+FB01), then you need to normalize your indexed contents and inputs to NFKC or NFKD. Use [String.precomposedStringWithCompatibilityMapping](https://developer.apple.com/documentation/foundation/nsstring/1412625-precomposedstringwithcompatibili) to turn a string into NFKC.
 
-Unicode normalization is not the end of the story, because it won't help "Encyclopaedia" match "Encyclopædia", "Mueller", "Müller", "Grossmann", "Großmann", or "Diyarbakır", "DIYARBAKIR". The [String.applyingTransform](https://developer.apple.com/reference/swift/string/1643133-applyingtransform) method can help.
+Unicode normalization is not the end of the story, because it won't help "Encyclopaedia" match "Encyclopædia", "Mueller", "Müller", "Grossmann", "Großmann", or "Diyarbakır", "DIYARBAKIR". The [String.applyingTransform](https://developer.apple.com/documentation/foundation/nsstring/1407787-applyingtransform) method can help.
 
 GRDB lets you write [custom FTS5 tokenizers](FTS5Tokenizers.md) that can transparently deal with all these issues. For FTS3 and FTS4, you'll need to pre-process your strings before injecting them in the full-text engine.