Fix scanning of unterminated strings

pkg/front_end/lib/src/fasta/scanner/string_scanner.dart

Index: pkg/front_end/lib/src/fasta/scanner/string_scanner.dart
diff --git a/pkg/front_end/lib/src/fasta/scanner/string_scanner.dart b/pkg/front_end/lib/src/fasta/scanner/string_scanner.dart
index 2c90b52e08dc999aabe3e684b3656cac90a80a4a..8e45e86cb0dbe23481e5aeb06883bdfd6106314e 100644
--- a/pkg/front_end/lib/src/fasta/scanner/string_scanner.dart
+++ b/pkg/front_end/lib/src/fasta/scanner/string_scanner.dart
@@ -60,7 +60,7 @@ class StringScanner extends ArrayBasedScanner {
       TokenType type, int start, bool asciiOnly, String closingQuotes) {
     String source = string.substring(start, scanOffset);
     return new SyntheticStringToken(
-        type, source + closingQuotes, start, source.length);
+        type, source + closingQuotes, tokenStart, source.length);

[ahe, 2017/07/21 14:09:09]

Why this change?

[danrubel, 2017/07/21 17:38:18]

It should have been tokenStart all along similar to the createSubstringToken
method above and to the parallel methods in Utf8BytesScanner

[ahe, 2017/08/01 14:19:57]

I get it now. tokenStart is recording character offsets, and start is byte
offsets. In this case, they should be the same.

We should come up with a naming convention for bytes and character offsets. What
do you think about using, for example, ByteOffset and CharOffset on all the int
variables representing offsets?

[danrubel, 2017/08/01 21:23:19]

That would be an improvement.

byteOffset     charOffset
byteStart      charStart
byteLength     charLength
byteEnd        charEnd

   }
 
   @override