fixes elm#970

This does not fully solve the problem of detecting case in Unicode, as it can also vary by locale.
This does make the isUpper/Lower and toUpper/Lower functions consistent.

Make Char.isLower and Char.isUpper Unicode-aware
This allows people with non-ASCII alphabets work with `Char.isLower` and `Char.isUpper`. Uses `toUpper` and `toLower`
underneath, which use Javascript's `String.prototype.toLower/UpperCase()`.
The second condition in the functions is there to distinguish between characters that have an upper/lower-case pairing,
and those that don't (`'0' == Char.toLower '0'` but we don't want `isLower '0'` to be true).

Author: Janiczek

src/Char.elm

@@ -66,46 +66,42 @@ type Char = Char -- NOTE: The compiler provides the real implementation.
 -- CLASSIFICATION
 
 
-{-| Detect upper case ASCII characters.
+{-| Detect upper case Unicode characters.
 
     isUpper 'A' == True
     isUpper 'B' == True
     ...
     isUpper 'Z' == True
+    
+    isUpper 'Σ' == True
 
     isUpper '0' == False
     isUpper 'a' == False
     isUpper '-' == False
-    isUpper 'Σ' == False
 -}
 isUpper : Char -> Bool
 isUpper char =
-  let
-    code =
-      toCode char
-  in
-    code <= 0x5A && 0x41 <= code
+  (char == Char.toUpper char)
+    && (char /= Char.toLower char)
 
 
-{-| Detect lower case ASCII characters.
+{-| Detect lower case Unicode characters.
 
     isLower 'a' == True
     isLower 'b' == True
     ...
     isLower 'z' == True
+    
+    isLower 'π' == True
 
     isLower '0' == False
     isLower 'A' == False
     isLower '-' == False
-    isLower 'π' == False
 -}
 isLower : Char -> Bool
 isLower char =
-  let
-    code =
-      toCode char
-  in
-    0x61 <= code && code <= 0x7A
+  (char == Char.toLower char)
+    && (char /= Char.toUpper char)
 
 
 {-| Detect upper case and lower case ASCII characters.