unicode string representation

[thautwarm]

concerned issues

purescript/purescript#3662

#21 (comment)

@Hong-Xiang when you have time, please give some comments about codeUnits issues which occurred while you was writing tests for purescript-quickcheck.

[Hong-Xiang]

I suggest checking unit tests for purescript-strings may be more clear for codePoints issues, it seems codeUnits passed tests.

Related code

The error happens in runnning import Test.Data.String.CodePoints (testStringCodePoints), which locates here in original purescript-strings repo.

Details about test case

in test case, the test string is defined as

str :: String
str = "a\xDC00\xD800\xD800\x16805\x16A06z"

and tests is accessing it by index, with native python's str implemetation, when accessing it with index 1 as the following test case

import Data.String.CodePoints as SCP

assertEqual
    { actual: map show (SCP.codePointAt 1 str)
    , expected: Just "(CodePoint 0xDC00)"
    }

The results is

Exception: Expected: (Just "(CodePoint 0xDC00)")
Actual:   (Just "(CodePoint 0xFFFD)")