Handle escapes in string literals.

frog/leg/ssa/nodes.dart

Index: frog/leg/ssa/nodes.dart
diff --git a/frog/leg/ssa/nodes.dart b/frog/leg/ssa/nodes.dart
index 27a8588cf14a1836f168c41508cb89034627a944..5449c5719425d5652fa2714358b8f4afe5ffe22e 100644
--- a/frog/leg/ssa/nodes.dart
+++ b/frog/leg/ssa/nodes.dart
@@ -1585,15 +1585,15 @@ class HLoopBranch extends HConditionalBranch {
     int flags = 0;
     int start = 0;
     int quoteChar = source.next();
-    if (quoteChar == '@'.charCodeAt(0)) {
+    if (quoteChar == $AT) {
       flags |= RAW;
       start = 1;
       quoteChar = source.next();
     }
-    if (quoteChar == '\''.charCodeAt(0)) {
+    if (quoteChar == $SQ) {
       flags |= SINGLE_QUOTED;
     } else {
-      assert(quoteChar == '"'.charCodeAt(0));
+      assert(quoteChar == $DQ);
     }
     // String has one quote. Check it if has three.
     // If it only have two, the string must be an empty string literal,
@@ -1634,6 +1634,7 @@ class HLoopBranch extends HConditionalBranch {
   bool get isMultiLine() => (flags & MULTI_LINE) != 0;
   bool get isRaw() => (flags & RAW) != 0;
   String get quoteChar() => ((flags & SINGLE_QUOTED) != 0) ? "'" : '"';
+  int get quoteCharCode() => ((flags & SINGLE_QUOTED) != 0) ? $SQ : $DQ;
 
   int get leftQuoteLength() =>
      hasLeftQuote ? (isRaw ? 1 : 0) + (isMultiLine ? 3 : 1) : 0;
@@ -1650,6 +1651,122 @@ class HLoopBranch extends HConditionalBranch {
 
   bool isEmpty() => unquotedSource().isEmpty();
 
+  static int hexValue(int hexDigit) {
+    // hexDigit is one of '0'..'9', 'A'..'F' and 'a'..'f'.
+    if (hexDigit <= $9) {
+      return hexDigit - $0;
+    }
+    // Make letters lowercase.
+    hexDigit |= $a - $A;

[floitsch, 2012/01/16 14:29:50]

This "trick" depends on an ascii property where lower/upper case is just
expressed by a bit. The x |= $a - $A would not (necessarily) work in other
encodings. Maybe:

assert((hexDigit | 0x20) == hexDigit < $a ? hexDigit + $a - $A : hexDigit);
hexDigit |= 0x20;

There is a magic constant, but at least it shows what code would have been there
to without the trick.

[Lasse Reichstein Nielsen, 2012/01/17 08:50:51]

I assume Unicode. There are no other encodings. FINI! 

Everything else in this method assume Unicode too (as does the characters.dart
file), so checking just this little bit seems spurious.

I've changed it to 
  |= $a ^ $A;
instead. More readable, arguably.

+    hexDigit -= $a - 10;
+    assert(10 <= hexDigit && hexDigit <= 15);
+    return hexDigit;
+  }
+
+
+  static void writeUnicodeEscape(Iterator<int> iter, StringBuffer buffer) {
+    buffer.add(@'\u');
+    int code = iter.next();
+    if (code == $OPEN_CURLY_BRACKET) {
+      // In Dart, '\u{'x{0..7}'}' is a valid escape, but not in
+      // JS. Convert to a \uxxxx escape.
+      int value = 0;
+      int hexDigit = iter.next();
+      do {
+        value = value * 16 + hexValue(hexDigit);
+        hexDigit = iter.next();
+      } while (hexDigit !== $CLOSE_CURLY_BRACKET);  // until '}'.

[floitsch, 2012/01/16 14:29:50]

remove comment.

[Lasse Reichstein Nielsen, 2012/01/17 08:50:51]

Done.

+      // JavaScript only supports characters in the BMP.
+      // TODO(lrn): Handle characters outside the BMP, e.g.,
+      // by having a slow-case string class.
+      if (value > 0xffff) value = 0xfffd;

[floitsch, 2012/01/16 14:29:50]

I would prefer bailing out (compiler.cancel).

[Lasse Reichstein Nielsen, 2012/01/17 08:50:51]

Done.

+
+      for (int j = 12; j >= 0; j -= 4) {
+        int digit = (value >> j) & 0xf;
+        buffer.add("0123456789abcdef"[digit]);
+      }
+      return;
+    }
+    for (int i = 0; i < 4; i++) {
+      if (i > 0) code = iter.next();
+      buffer.add(new String.fromCharCodes([code]));
+    }
+  }
+
+  /**
+   * Write the contents of the quoted string to a [StringBuffer] in
+   * a form that is valid as JavaScript string literal content.
+   * The string is assumed quoted by [quote] characters.
+   * This method doesn't try to make the shortest string, but rather
+   * to be as close to the original string as possible.
+   */
+  void writeEscaped(StringBuffer buffer, int quote) {
+    bool raw = this.isRaw;
+    Iterator<int> iter =
+        wrappedString.copyWithoutQuotes(leftQuoteLength,
+                                        rightQuoteLength).iterator();
+    while (iter.hasNext()) {
+      int code = iter.next();
+      if (code == quote) {
+        // We need to add a backslash before quotes, both in normal
+        // and in raw strings.
+        buffer.add(@'\');
+        buffer.add(code == $SQ ? "'" : '"');
+      } else if (code == $LF) {
+        // Newlines in strings only occour in multiline strings.
+        // They need to be written using escapes in JS..
+        assert(isMultiLine);
+        buffer.add(@'\n');
+      } else if (code == $CR) {
+        assert(isMultiLine);
+        buffer.add(@'\r');
+      } else if (code == $LS || code == $PS) {
+        // These Unicode line terminators are invalid in JS strings.
+        buffer.add(code == $LS ? @'\u2028' : @'\u2029');
+      } else if (code != $BACKSLASH) {
+        buffer.add(new String.fromCharCodes([code]));
+      } else if (raw) {
+        buffer.add(@'\\');
+      } else {
+        code = iter.next();
+        switch (code) {
+          case $u:
+            writeUnicodeEscape(iter, buffer);
+            break;
+          case $x:
+            buffer.add(@'\x');
+            // Remaining character will be copied literally.
+            break;
+          // Character escapes that identical in meaning in JS.
+          case $b: buffer.add(@'\b'); break;
+          case $f: buffer.add(@'\f'); break;
+          case $n: buffer.add(@'\n'); break;
+          case $r: buffer.add(@'\r'); break;
+          case $t: buffer.add(@'\t'); break;
+          case $v: buffer.add(@'\v'); break;
+          // Identity escapes that must be escaped in JS strings.
+          case $BACKSLASH: buffer.add(@'\\'); break;
+          case $LF: buffer.add(@'\n'); break;
+          case $CR: buffer.add(@'\r'); break;
+          case $LS: buffer.add(@'\u2028'); break;
+          case $PS: buffer.add(@'\u2029'); break;
+          // Quotes may or may not need the escape.
+          case $SQ:
+          case $DQ:
+            // Only escape quotes if they match the generated string quotes.
+            if (code == quote) buffer.add(@'\');
+            buffer.add(code == $SQ ? "'" : '"');
+            break;
+          default:
+            // All other escaped characters are identity escapes,
+            // and don't need a backslash in JS.
+            buffer.add(new String.fromCharCodes([code]));
+            break;
+        }
+      }
+    }
+  }
+
   /**
     * Does a conservative test for equality between two quoted strings.
     * Returns true if the two definitly have the same string.