Handle speech-mark at end of raw string

[peterfry]

I noticed that syntax highlighting breaks if you have a speech-mark at the end of your raw string. e.g.

test = """"foo""""

Any subsequent lines will be highlighted as string content.

If there's a space between the speech-mark and the closing triple-speech-marks, it works fine. e.g.

test = """"foo" """

It also works if there are two speech marks. e.g.

test = """"foo"""""

I suspect as soon as it sees the first three speech-marks at the end, it ends the string literal. The fourth speech-mark then incorrectly begins a new string literal.

[htonheim]

Seeing the same issue in a build file from TeamCity. IntelliJ IDEA parses it ok, but in VS Code the lower half of the .kts file is considered a comment. Or rather, the code is considered comments and the comments are considered code.

Happens on this bit of code:

features {
    feature {
        type = "JetBrains.AssemblyInfo"
        param("file-format", "%system.build.number%")
    }
    swabra {
        filesCleanup = Swabra.FilesCleanup.DISABLED
        lockingProcesses = Swabra.LockingProcessPolicy.KILL
    }
    replaceContent {
        fileRules = "**/*.edmx"
        pattern = """ProviderManifestToken=\".{4,6}\""""
        replacement = """ProviderManifestToken="2008""""
    }
}

requirements {
    exists("DotNetFramework4.7_x64")
}

In the line pattern = """ProviderManifestToken=\".{4,6}\"""" the backslashes are treated as escape characters, which they are not since this is a raw string. This however stops the plugin form treating everything after this as a comment. On the next line however, there is no backslashes and the last four quotation marks makes the plugin treat everything afterwards as a comment.

[nishtahir]

Oh that shouldn't be happening 🤔. I'll look into this