JavaTokenParsers stringLiteral Unicode literals not fully correct

[scabug]

The stringLiteral function in scala.util.parsing.combinator.!JavaTokenParsers
tries to parse String literals with Unicode escape
sequence in them

  def stringLiteral: Parser[String] = 
    ("\""+"""([^"\p{Cntrl}\\]|\\[\\/bfnrt]|\\u[a-fA-F0-9]{4})*"""+"\"").r

However, such Unicode escapes can occur elsewhere,
including the first \ of an escape sequence,
or even instead of the " characters themselves (\u0022).

If you wish to support Unicode escapes, they should be
handled in a lower level stream Char => Char parser only,
where they would apply to string literals, numbers,
identifiers, etc.

Compile and run the following program

import scala.util.parsing.combinator.JavaTokenParsers
object StringLiteral extends JavaTokenParsers with Application {
  override def main(args: Array[String]) {
    args.foreach { a => println(a)
                        println(parseAll(stringLiteral, a)) }
  }
}

and pass it the strings:

  scala StringLiteral "\"trivial\"" "\"A backslash \u005c\ character\u0022"

and it parses the first correctly but fails on the second argument

"trivial"
[1.10] parsed: "trivial"
"A backslash \u005c\ character\u0022
[1.1] failure: string matching regex `"([^"\p{Cntrl}\\]|\\[\\/bfnrt]|\\u[a-fA-F0-9]{4})*"' expected but `"' found

"A backslash \u005c\ character\u0022
^

(Be careful about quoting arguments to the shell.)

[scabug]

Imported From: https://issues.scala-lang.org/browse/SI-1566?orig=1
Reporter: @DavidBiesack

[scabug]

@dcsobral said:
I'm siding on removing the unicode treatment with a migration warning.