[parser] implement error reporting for Scanner errors

src/parsing/scanner.cc

Index: src/parsing/scanner.cc
diff --git a/src/parsing/scanner.cc b/src/parsing/scanner.cc
index a197829b35b69a6ecbe7d9b7521554eeb8c8c8c1..507e1ad1d8500ed5515bc0e16c84248a71d3ea27 100644
--- a/src/parsing/scanner.cc
+++ b/src/parsing/scanner.cc
@@ -62,15 +62,19 @@ void Scanner::Initialize(Utf16CharacterStream* source) {
   Scan();
 }
 
-
-template <bool capture_raw>
+template <bool capture_raw, bool unicode>
 uc32 Scanner::ScanHexNumber(int expected_length) {
   DCHECK(expected_length <= 4);  // prevent overflow
 
+  int begin = source_pos() - 2;
   uc32 x = 0;
   for (int i = 0; i < expected_length; i++) {
     int d = HexValue(c0_);
     if (d < 0) {
+      ReportScannerError(Location(begin, begin + expected_length + 2),
+                         unicode
+                             ? MessageTemplate::kInvalidUnicodeEscapeSequence
+                             : MessageTemplate::kInvalidHexEscapeSequence);
       return -1;
     }
     x = x * 16 + d;
@@ -80,20 +84,27 @@ uc32 Scanner::ScanHexNumber(int expected_length) {
   return x;
 }
 
-
 template <bool capture_raw>
 uc32 Scanner::ScanUnlimitedLengthHexNumber(int max_value) {
   uc32 x = 0;
   int d = HexValue(c0_);
-  if (d < 0) {
-    return -1;
-  }
+  int nof_digits = 0;
   while (d >= 0) {
     x = x * 16 + d;
-    if (x > max_value) return -1;
+    if (x > max_value) {
+      return -1;

[adamk, 2016/03/21 19:13:56]

Seems like this is the right place to report the going-over-the-max error, since
that's the constraint this method implements.

[caitp (gmail), 2016/03/21 19:40:05]

Done.

+    }
     Advance<capture_raw>();
     d = HexValue(c0_);
+    nof_digits++;
+  }
+
+  if (!nof_digits || (c0_ != '}')) {

[adamk, 2016/03/21 19:13:56]

I'd leave the closing brace check in the caller, since it has the opening brace
check.

And the InvalidUnicodeEscapeSequence can also live in the caller, once the "max
value" check is reported here. That'll make the code shape exactly the same as
it was before the addition of error messages.

[caitp (gmail), 2016/03/21 19:40:05]

Done.

+    ReportScannerError(source_pos(),
+                       MessageTemplate::kInvalidUnicodeEscapeSequence);
+    return -1;
   }
+
   return x;
 }
 
@@ -856,7 +867,9 @@ Token::Value Scanner::ScanString() {
     uc32 c = c0_;
     Advance();
     if (c == '\\') {
-      if (c0_ < 0 || !ScanEscape<false, false>()) return Token::ILLEGAL;
+      if (c0_ < 0 || !ScanEscape<false, false>()) {
+        return Token::ILLEGAL;
+      }
     } else {
       AddLiteralChar(c);
     }
@@ -888,7 +901,6 @@ Token::Value Scanner::ScanTemplateSpan() {
   StartRawLiteral();
   const bool capture_raw = true;
   const bool in_template_literal = true;
-
   while (true) {
     uc32 c = c0_;
     Advance<capture_raw>();
@@ -1108,18 +1120,19 @@ uc32 Scanner::ScanUnicodeEscape() {
   // Accept both \uxxxx and \u{xxxxxx}. In the latter case, the number of
   // hex digits between { } is arbitrary. \ and u have already been read.
   if (c0_ == '{') {
+    int begin = source_pos() - 2;
     Advance<capture_raw>();
     uc32 cp = ScanUnlimitedLengthHexNumber<capture_raw>(0x10ffff);
     if (cp < 0) {
-      return -1;
-    }
-    if (c0_ != '}') {
+      ReportScannerError(Location(begin, source_pos() + 1),
+                         MessageTemplate::kUndefinedUnicodeCodePoint);
       return -1;
     }
     Advance<capture_raw>();
     return cp;
   }
-  return ScanHexNumber<capture_raw>(4);
+  const bool unicode = true;
+  return ScanHexNumber<capture_raw, unicode>(4);
 }