escape sequence in `Elm.CodeGen.string` sometimes isn't escaped in pretty printed expression

[lue-bird]

While all the usual escape sequences like Elm.CodeGen.string "\n" or with "\\", "\t" or "\"" get printed as "\n", "\\" and "\t", "\"", the same is not true for carriage return and backslash u{...} sequences:

Test.describe "escape sequence in `Elm.CodeGen.string` sometimes isn't escaped in pretty printed expression"
    [ Test.test "string with escaped carriage return is pretty printed correctly"
        (\() ->
            Elm.CodeGen.string "\u{000D}"
                |> Elm.Pretty.prettyExpression
                |> Pretty.pretty 100
                |> Expect.equal "\"\\r\""
        )
    , Test.test "string with escaped combo emoji is pretty printed correctly"
        (\() ->
            Elm.CodeGen.string "👩\u{200D}🚒"
                |> Elm.Pretty.prettyExpression
                |> Pretty.pretty 100
                |> String.toList
                |> Expect.equalLists ("\"👩\\u{200D}🚒\"" |> String.toList)
        )
    ]

the elm-test output is

✗ string with escaped carriage return is pretty printed correctly

    "\"\r\""
    ╷
    │ Expect.equal
    ╵
    "\"\\r\""

✗ string with escaped combo emoji is pretty printed correctly

    '"', '👩', '‍', '🚒', '"'
    ╷
    │ Expect.equalLists
    ╵
    '"', '👩', '\\', 'u', '{', '2', '0', '0', 'D', '}', '🚒', '"'

    The first diff is at index 2: it was `'‍'`, but `'\\'` was expected.

This was noticed by Ed Kelly in https://github.com/lue-bird/elm-review-documentation-code-snippet/issues/5

The fix would be recognizing and handling those in https://github.com/the-sett/elm-syntax-dsl/blob/master/src/Elm/Pretty.elm#L1784 which needs to check both

• is it backslash r
• is it not printable according to the isPrint in https://hackage.haskell.org/package/base-4.20.0.1/docs/Data-Char.html (used by elm-format https://github.com/avh4/elm-format/blob/c99aa91b1b239653185e55a2085bceebcd1b5c4b/elm-format-lib/src/ElmFormat/Render/Box.hs#L1978) which could be checked using https://dark.elm.dmy.fr/packages/miniBill/elm-unicode/latest/Unicode

[rupertlssmith]

Been a while since I looked at this, but I definitely remember problems with escaping unicode that were never fixed.

Will you be submitting a PR with a fix?

[lue-bird]

Sure, let me make a PR for this.

Edit: done in #51

[rupertlssmith]

Published as 6.0.3