Reject unsupported newlines in tokenizer ASCII fast path (#15262)

Author: lukaszsamson

lib/elixir/src/elixir_interpolation.erl

@@ -80,6 +80,15 @@ extract([$\\ | Rest], Buffer, Output, Line, Column, Scope, Interpol, Last) ->
 
 %% Catch all clause
 
+extract([Char | _Rest], _Buffer, _Output, Line, Column, _Scope, _Interpol, _Last)
+    when ?break(Char) ->
+  Token = io_lib:format("\\u~4.16.0B", [Char]),
+  Pos = io_lib:format(". If you want to use such character, use it in its escaped ~ts form instead", [Token]),
+  {error, {?LOC(Line, Column),
+    {"invalid line break character in string: ",
+     Pos},
+    Token}};
+
 extract([Char1, Char2 | Rest], Buffer, Output, Line, Column, Scope, Interpol, Last)
     when Char1 =< 255, Char2 =< 255 ->
   extract([Char2 | Rest], [Char1 | Buffer], Output, Line, Column + 1, Scope, Interpol, Last);

lib/elixir/test/elixir/kernel/parser_test.exs

@@ -1131,6 +1131,30 @@ defmodule Kernel.ParserTest do
         ],
         ~c"\"this is a \\\u2028\""
       )
+
+      assert_syntax_error(
+        [
+          "nofile:1:12:",
+          "invalid line break character in string: \\u000B. If you want to use such character, use it in its escaped \\u000B form instead"
+        ],
+        :erlang.list_to_binary([34] ++ ~c"this is a " ++ [11, 34])
+      )
+
+      assert_syntax_error(
+        [
+          "nofile:1:12:",
+          "invalid line break character in string: \\u000C. If you want to use such character, use it in its escaped \\u000C form instead"
+        ],
+        :erlang.list_to_binary([34] ++ ~c"this is a " ++ [12, 34])
+      )
+
+      assert_syntax_error(
+        [
+          "nofile:1:12:",
+          "invalid line break character in string: \\u0085. If you want to use such character, use it in its escaped \\u0085 form instead"
+        ],
+        <<34, "this is a ", 194, 133, 34>>
+      )
     end
 
     test "reserved tokens" do