"The options are written in an incorrect format." when saving Trumbowyg settings

[Piedone]

Describe the bug

When saving the Trumbowyg settings, at least for Html Field, you get the "The options are written in an incorrect format." error with the default options.

To Reproduce

1. Set up a fresh site from main with the Blog recipe.
2. Go to /Admin/ContentTypes/Edit/RawHtml.
3. Edit the Content Field.
4. What type of editor should be used? -> Trumbowyg editor.
5. Change nothing, Save.
6. See the "The options are written in an incorrect format." validation error.

Expected behavior

The default options just work.

Screenshots

2024-04-09_17h59_03.mp4

[Piedone]

This is a regression since 1.8.2.

[Piedone]

@MikeAlhayek this looks like a JSON issue too.

[Skrypt]

That's a good example of update regression. 😄

[MikeAlhayek]

I am guessing that Json contains a trailing comma or contains comments.

[Piedone]

The problem is that since 1ee808f the Trumbowyg editor's settings (and since then, a lot of other code using IsJson()) is validated as JSON. However, those settings aren't, and never were, actually JSON. Since this is not JSON, but otherwise a valid JS object:

{
    autogrow: true,
    btns: [
        ["viewHTML"],
        ["undo", "redo"],
        ["formatting"],
        ["strong", "em", "del"],
        ["foreColor", "backColor"],
        ["superscript", "subscript"],
        ["link"],
        ["insertShortcode"],
        ["image"],
        ["align"],
        ["unorderedList", "orderedList"],
        ["horizontalRule"],
        ["removeformat"],
        ["fullscreen"]
    ],
    btnsDef: {
        align: {
            dropdown: ["justifyLeft", "justifyCenter", "justifyRight", "justifyFull"]
            ico: "justifyLeft"
        },
        image: {
            dropdown: ["insertImage", "base64", "noembed"],
            ico: "insertImage"
        }
    }
}

We need to keep this for Trumbowyg.

So, for these, maybe we need JS, not JSON validation instead? Either Jint Engine.PrepareScript(), or going one lower, what it does with new JavaScriptParser().ParseScript() (we mustn't try to execute the script, of course), and similarly catching the exception on error.