Handle escapes in string literals.

frog/leg/ssa/nodes.dart

 // Copyright (c) 2011, the Dart project authors.  Please see the AUTHORS file
 // for details. All rights reserved. Use of this source code is governed by a
 // BSD-style license that can be found in the LICENSE file.
 
 interface HVisitor<R> {
   R visitAdd(HAdd node);
   R visitBailoutTarget(HBailoutTarget node);
   R visitBitAnd(HBitAnd node);
   R visitBitNot(HBitNot node);
   R visitBitOr(HBitOr node);
@@ skipping 1567 matching lines @@
   /**
    * Finds the quote type for a string given a part of it containing the
    * starting quote. Returns flags, but doesn't include HAS_LEFT_QUOTE
    * or HAS_RIGHT_QUOTE.
    */
   static int flagsFromLeftQuote(SourceString sourceString) {
     Iterator<int> source = sourceString.iterator();
     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,
     // and end after the second quote.
     if (source.hasNext() && source.next() == quoteChar && source.hasNext()) {
       assert(source.next() == quoteChar);
       flags |= MULTI_LINE;
     }
     return flags;
   }
@@ skipping 20 matching lines @@
     if ((flags & HAS_BOTH_QUOTES) == 0) return wrappedString;
     return wrappedString.copyWithoutQuotes(leftQuoteLength,
                                            rightQuoteLength);
   }
 
   bool get hasLeftQuote() => (flags & HAS_LEFT_QUOTE) != 0;
   bool get hasRightQuote() => (flags & HAS_RIGHT_QUOTE) != 0;
   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;
   int get rightQuoteLength() =>
      hasRightQuote ? (isMultiLine ? 3 : 1) : 0;
   static int leftQuoteLengthFromFlags(int flags) {
     if ((flags & HAS_LEFT_QUOTE) == 0) return 0;
     return (flags & (RAW | MULTI_LINE)) + 1;
   }
   static int rightQuoteLengthFromFlags(int flags) {
     if ((flags & HAS_RIGHT_QUOTE) == 0) return 0;
     return (flags & MULTI_LINE) + 1;
   }
 
   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.
     * Returns false if the strings are different, or if it's not possible
     * to (quickly) determine whether they are equal.
     */
   bool definitlyEquals(QuotedString other) {
     return flags == other.flags && wrappedString == other.wrappedString;
   }
 
@@ skipping 444 matching lines @@
 
   HInstruction get expression() => inputs[0];
 
   HType computeType() => HType.BOOLEAN;
   bool hasExpectedType() => true;
 
   accept(HVisitor visitor) => visitor.visitIs(this);
 
   toString() => "$expression is $typeExpression";
 }