Add string comparators that can compare ASCII strings ignoring case.

lib/src/comparators.dart

Index: lib/src/comparators.dart
diff --git a/lib/src/comparators.dart b/lib/src/comparators.dart
new file mode 100644
index 0000000000000000000000000000000000000000..b9f496a8310444860afaaa3cce43a59f8d3cbf47
--- /dev/null
+++ b/lib/src/comparators.dart
@@ -0,0 +1,397 @@
+// Copyright (c) 2015, 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.
+
+library dart.pkg.collection.comparators;
+
+// Character constants.
+const int _zero         = 0x30;
+const int _upperCaseA   = 0x41;
+const int _upperCaseZ   = 0x5d;

[Lasse Reichstein Nielsen, 2015/11/19 07:15:37]

Whoops, that should be 0x5a, and 0x7a below. Will add to test!

+const int _lowerCaseA   = 0x61;
+const int _lowerCaseZ   = 0x7D;
+const int _asciiCaseBit = 0x20;
+
+/// Checks if strings [a] and [b] differ only on the case of ASCII letters.
+///
+/// Strings are equal if they have the same length, and the characters at
+/// each index are the same, or they are ASCII letters where one is upper-case
+/// and the other is the lower-case version of the same letter.
+///
+/// The comparison does not ignore the case of non-ASCII letters, so
+/// a capital ae-ligature (Æ) is different from a lower case ae-ligature (æ).
+///
+/// Ignoring non-ASCII letters is not generally a good idea, but it makes sense
+/// for situations where the strings are known to be ASCII. Examples could
+/// be Dart identifiers, base-64 or hex encoded strings, GUIDs or similar
+/// strings with a known structure.
+bool equalsIgnoreAsciiCase(String a, String b) {
+  if (a.length != b.length) return false;
+  for (int i = 0; i < a.length; i++) {
+    var aChar = a.codeUnitAt(i);
+    var bChar = b.codeUnitAt(i);
+    if (aChar == bChar) continue;
+    // Quick-check for whether this may be different cases of the same letter.
+    if (aChar ^ bChar != _asciiCaseBit) return false;
+    // If it's possible, then check if either character is actually an ASCII
+    // letter.
+    int aCharUpperCase = aChar | _asciiCaseBit;
+    if (_upperCaseA <= aCharUpperCase && aCharUpperCase <= _upperCaseZ) {
+      continue;
+    }
+    return false;
+  }
+  return true;
+}
+
+
+/// Hash code for a string which is compatible with [equalsIgnoreAsciiCase].
+///
+/// The hash code is unaffected by changing the case of ASCII letters, but
+/// the case of non-ASCII letters do affect the result.
+int hashIgnoreAsciiCase(String string) {
+  // Jenkins hash code ( http://en.wikipedia.org/wiki/Jenkins_hash_function).

[sra1, 2015/11/18 23:40:16]

Jenkins hash code modified to operate on Smi values.

[Lasse Reichstein Nielsen, 2015/11/19 07:15:37]

Done.

+  // Same hash used by dart2js for strings, modified to ignore ASCII letter
+  // case.
+  int hash = 0;
+  for (int i = 0; i < string.length; i++) {
+    int char = string.codeUnitAt(i);
+    // Convert lower-case ASCII letters to upper case.upper
+    // This ensures that strings that differ only in case will have the
+    // same hash code.
+    if (_lowerCaseA <= char && char <= _lowerCaseZ) char -= _asciiCaseBit;
+    hash = 0x1fffffff & (hash + char);
+    hash = 0x1fffffff & (hash + ((0x0007ffff & hash) << 10));
+    hash >>= 6;
+  }
+  hash = 0x1fffffff & (hash + ((0x03ffffff & hash) << 3));
+  hash >>= 11;
+  return 0x1fffffff & (hash + ((0x00003fff & hash) << 15));
+}
+
+
+/// Compares [a] and [b] lexically, converting ASCII letters to upper case.
+///
+/// Comparison treats all lower-case ASCII letters as upper-case letters,
+/// but does no case conversion for non-ASCII letters.
+///
+/// If two strings differ only on the case of ASCII letters, the one with the
+/// capital letter at the first difference will compare as less than the other
+/// string. This tie-breaking ensures that the comparison is a total ordering
+/// on strings and is compatible with equality.
+///
+/// Ignoring non-ASCII letters is not generally a good idea, but it makes sense
+/// for situations where the strings are known to be ASCII. Examples could
+/// be Dart identifiers, base-64 or hex encoded strings, GUIDs or similar
+/// strings with a known structure.
+int compareAsciiUpperCase(String a, String b) {
+  int defaultResult = 0; // Returned if no difference found.
+  for (int i = 0; i < a.length; i++) {
+    if (i == b.length) return 1;
+    var aChar = a.codeUnitAt(i);
+    var bChar = b.codeUnitAt(i);
+    if (aChar == bChar) continue;
+    // Upper-case if letters.
+    int aUpperCase = aChar;
+    int bUpperCase = bChar;
+    if (_lowerCaseA <= aChar && aChar <= _lowerCaseZ) {
+      aUpperCase -= _asciiCaseBit;
+    }
+    if (_lowerCaseA <= bChar && bChar <= _lowerCaseZ) {
+      bUpperCase -= _asciiCaseBit;
+    }
+    if (aUpperCase != bUpperCase) return (aUpperCase - bUpperCase).sign;
+    if (defaultResult == 0) defaultResult = (aChar - bChar);
+  }
+  if (b.length > a.length) return -1;
+  return defaultResult.sign;
+}
+
+
+/// Compares [a] and [b] lexically, converting ASCII letters to lower case.
+///
+/// Comparison treats all upper-case ASCII letters as lower-case letters,
+/// but does no case conversion for non-ASCII letters.
+///
+/// If two strings differ only on the case of ASCII letters, the one with the
+/// capital letter at the first difference will compare as less than the other
+/// string. This tie-breaking ensures that the comparison is a total ordering
+/// on strings.
+///
+/// Ignoring non-ASCII letters is not generally a good idea, but it makes sense
+/// for situations where the strings are known to be ASCII. Examples could
+/// be Dart identifiers, base-64 or hex encoded strings, GUIDs or similar
+/// strings with a known structure.
+int compareAsciiLowerCase(String a, String b) {

[floitsch, 2015/11/18 22:50:11]

How is this one different from compareAsciiUpperCase?

[Lasse Reichstein Nielsen, 2015/11/19 07:15:37]

Only in how it orders, e.g., "a" vs. "_".  Since "_" is between "A" and "a", it
will come before "a" when you do lower-case comparison and after "a" when you do
upper-case comparison.

So, basically, it differs in the ordering of only six characters when they are
compared to letters (0x5b-0x61).

You won't need both, but existing code is likely to have used either  (a, b) =>
a.toUpperCase().compareTo(b.toUpperCase())  or  (a, b) =>
a.toLowerCase().compareTo(b.toLowerCase()) , and will want to replace it with an
equivalent version from here without having to change tests.

+  int defaultResult = 0;
+  for (int i = 0; i < a.length; i++) {
+    if (i == b.length) return 1;
+    var aChar = a.codeUnitAt(i);
+    var bChar = b.codeUnitAt(i);
+    if (aChar == bChar) continue;
+    int aLowerCase = aChar;
+    int bLowerCase = bChar;
+    // Upper case if ASCII letters.
+    if (_upperCaseA <= bChar && bChar <= _upperCaseZ) {
+      bLowerCase += _asciiCaseBit;
+    }
+    if (_upperCaseA <= aChar && aChar <= _upperCaseZ) {
+      aLowerCase += _asciiCaseBit;
+    }
+    if (aLowerCase != bLowerCase) return (aLowerCase - bLowerCase).sign;
+    if (defaultResult == 0) defaultResult = aChar - bChar;
+  }
+  if (b.length > a.length) return -1;
+  return defaultResult.sign;
+}
+
+/// Compares strings [a] and [b] according to [natural sort ordering].
+///
+/// A natural sort ordering is a lexical ordering where embedded
+/// numerals (digit sequeunces) are treated as a single unit and ordered by
+/// numerical value.
+/// This means that `"a10b"` will be ordered after `"a7b"` in natural
+/// ordering, where leixcal ordering would put the `1` before the `7`, ignoring

[floitsch, 2015/11/18 22:50:11]

lexical

+/// that the `1` is part of a larger number.
+///
+/// Example:
+/// The following strings are in the order they would be sorted by using this
+/// comparison function:
+///
+///     "a", "a0", "a0b", "a1", "a01", "a9", "a10", "a100", "a100b", "aa"
+///
+/// [natural sort ordering]: https://en.wikipedia.org/wiki/Natural_sort_order
+int compareNatural(String a, String b) {
+  for (int i = 0; i < a.length; i++) {
+    if (i == b.length) return 1;

[sra1, 2015/11/18 23:40:16]

Use

  if (i >= b.length)

to improve constraint-based bounds checking.

[Lasse Reichstein Nielsen, 2015/11/19 07:15:37]

Done.

+    var aChar = a.codeUnitAt(i);
+    var bChar = b.codeUnitAt(i);
+    if (aChar != bChar) {
+      return _compareNaturally(a, b, i, aChar, bChar);
+    }
+  }
+  if (b.length > a.length) return -1;
+  return 0;
+}
+
+/// Compares strings [a] and [b] according to lower-case
+/// [natural sort ordering].
+///
+/// ASCII letters are converted to lower case before being compared, like
+/// for [compareAsciiLowerCase], then the result is compared like for
+/// [compareNatural].
+///
+/// If two strings differ only on the case of ASCII letters, the one with the
+/// capital letter at the first difference will compare as less than the other
+/// string. This tie-breaking ensures that the comparison is a total ordering
+/// on strings.
+///
+/// [natural sort ordering]: https://en.wikipedia.org/wiki/Natural_sort_order
+int compareAsciiLowerCaseNatural(String a, String b) {
+  int defaultResult = 0; // Returned if no difference found.
+  for (int i = 0; i < a.length; i++) {
+    if (i == b.length) return 1;
+    var aChar = a.codeUnitAt(i);
+    var bChar = b.codeUnitAt(i);
+    if (aChar == bChar) continue;
+    int aLowerCase = aChar;
+    int bLowerCase = bChar;
+    if (_upperCaseA <= aChar && aChar <= _upperCaseZ) {
+      aLowerCase += _asciiCaseBit;
+    }
+    if (_upperCaseA <= bChar && bChar <= _upperCaseZ) {
+      bLowerCase += _asciiCaseBit;
+    }
+    if (aLowerCase != bLowerCase) {
+      return _compareNaturally(a, b, i, aLowerCase, bLowerCase);
+    }
+    if (defaultResult == 0) defaultResult = aChar - bChar;
+  }
+  if (b.length > a.length) return -1;
+  return defaultResult.sign;
+}
+
+/// Compares strings [a] and [b] according to upper-case
+/// [natural sort ordering].
+///
+/// ASCII letters are converted to upper case before being compared, like
+/// for [compareAsciiUpperCase], then the result is compared like for
+/// [compareNatural].
+///
+/// If two strings differ only on the case of ASCII letters, the one with the
+/// capital letter at the first difference will compare as less than the other
+/// string. This tie-breaking ensures that the comparison is a total ordering
+/// on strings
+///
+/// [natural sort ordering]: https://en.wikipedia.org/wiki/Natural_sort_order
+int compareAsciiUpperCaseNatural(String a, String b) {

[floitsch, 2015/11/18 22:50:11]

How is this one different from compareAsciiLowerCaseNatural.

[Lasse Reichstein Nielsen, 2015/11/19 07:15:37]

Same as above, differs on six characters that lie between upper case and lower
case letters.

+  int defaultResult = 0;
+  for (int i = 0; i < a.length; i++) {
+    if (i == b.length) return 1;
+    var aChar = a.codeUnitAt(i);
+    var bChar = b.codeUnitAt(i);
+    if (aChar == bChar) continue;
+    int aUpperCase = aChar;
+    int bUpperCase = bChar;
+    if (_lowerCaseA <= aChar && aChar <= _lowerCaseZ) {
+      aUpperCase -= _asciiCaseBit;
+    }
+    if (_lowerCaseA <= bChar && bChar <= _lowerCaseZ) {
+      bUpperCase -= _asciiCaseBit;
+    }
+    if (aUpperCase != bUpperCase) {
+      return _compareNaturally(a, b, i, aUpperCase, bUpperCase);
+    }
+    if (defaultResult == 0) defaultResult = aChar - bChar;
+  }
+  if (b.length > a.length) return -1;
+  return defaultResult.sign;
+}
+
+/// Check for numbers overlapping the current mismatched characters.
+///
+/// If both [aChar] and [bChar] are digits, use numerical comparison.
+/// Check if the previous characters is a non-zero number, and if not,
+/// skip - but count - leading zeros before comparing numbers.
+///
+/// If one is a digit and the other isn't, check if the previous character
+/// is a digit, and if so, the the one with the digit is the greater number.
+///
+/// Otherwise just returns the difference between [aChar] and [bChar].
+int _compareNaturally(
+    String a, String b, int index, int aChar, int bChar) {
+  assert(aChar != bChar);
+  var aIsDigit = _isDigit(aChar);
+  var bIsDigit = _isDigit(bChar);
+  if (aIsDigit) {
+    if (bIsDigit) {
+      return _compareNumerically(a, b, aChar, bChar, index);
+    } else if (index > 0 && _isDigit(a.codeUnitAt(index - 1))) {
+      // aChar is the continuation of a longer number.
+      return 1;
+    }
+  } else if (bIsDigit && index > 0 && _isDigit(b.codeUnitAt(index - 1))) {
+    // bChar is the continuation of a longer number.
+    return -1;
+  }
+  // Characters are both non-digits, or not continuation of earlier number.
+  return (aChar - bChar).sign;
+}
+
+/// Compare numbers overlapping [aChar] and [bChar] numerically.
+///
+/// If the numbers have the same numerical value, but one has more leading
+/// zeros, the longer number is considered greater than the shorter one.
+///
+/// This ensures a total ordering on strings compatible with equality.
+int _compareNumerically(String a, String b, int aChar, int bChar, int index) {
+  // Both are digits. Find the first significant different digit, then find
+  // the length of the numbers.
+  if (_isNumberSuffix(a, index)) {
+    // Part of a longer number, differs at this index, just count the length.
+    int result = _compareDigitCount(a, b, index, index);
+    if (result != 0) return result;
+    // If same length, the current character is the most significant differing
+    // digit.
+    return (aChar - bChar).sign;
+  }
+  // Not part of larger (non-zero) number, so skip leading zeros before
+  // comparing numbers.
+  int aIndex = index;
+  int bIndex = index;
+  if (aChar == _zero) {
+    do {
+      aIndex++;
+      if (aIndex == a.length) return -1;  // number in a is zero, b is not.
+      aChar = a.codeUnitAt(aIndex);
+    } while (aChar == _zero);
+    if (!_isDigit(aChar)) return -1;
+  } else if (bChar == _zero) {
+    do {
+      bIndex++;
+      if (bIndex == b.length) return 1;  // number in b is zero, a is not.
+      bChar = b.codeUnitAt(bIndex);
+    } while (bChar == _zero);
+    if (!_isDigit(bChar)) return 1;
+  }
+  if (aChar != bChar) {
+    int result = _compareDigitCount(a, b, aIndex, bIndex);
+    if (result != 0) return result;
+    return (aChar - bChar).sign;
+  }
+  // Same leading digit, one had more leading zeros.
+  // Compare digits until reaching a difference.
+  while (true) {
+    var aIsDigit = false;
+    var bIsDigit = false;
+    aChar = 0;
+    bChar = 0;
+    if (++aIndex < a.length) {
+      aChar = a.codeUnitAt(aIndex);
+      aIsDigit = _isDigit(aChar);
+    }
+    if (++bIndex < b.length) {
+      bChar = b.codeUnitAt(bIndex);
+      bIsDigit = _isDigit(bChar);
+    }
+    if (aIsDigit) {
+      if (bIsDigit) {
+        if (aChar == bChar) continue;
+        // First different digit found.
+        break;
+      }
+      // bChar is non-digit, so a has longer number.
+      return 1;
+    } else if (bIsDigit) {
+      return -1;  // b has longer number.
+    } else {
+      // Neither is digit, so numbers had same numerical value.
+      // Fall back on number of leading zeros
+      // (reflected by difference in indices).
+      return (aIndex - bIndex).sign;
+    }
+  }
+  // At first differing digits.
+  int result = _compareDigitCount(a, b, aIndex, bIndex);
+  if (result != 0) return result;
+  return (aChar - bChar).sign;
+}
+
+/// Checks which of [a] and [b] has the longest sequence of digits.
+///
+/// Starts counting from `i + 1` and `j + 1` (assumes that `a[i]` and `b[j]` are
+/// both already known to be digits).
+int _compareDigitCount(String a, String b, int i, int j) {
+  while (++i < a.length) {
+    bool aIsDigit = _isDigit(a.codeUnitAt(i));
+    if (++j == b.length) return aIsDigit ? 1 : 0;
+    bool bIsDigit = _isDigit(b.codeUnitAt(j));
+    if (aIsDigit) {
+      if (bIsDigit) continue;
+      return 1;
+    } else if (bIsDigit) {
+      return -1;
+    } else {
+      return 0;
+    }
+  }
+  if (++j < b.length && _isDigit(b.codeUnitAt(j))) {
+    return -1;
+  }
+  return 0;
+}
+
+bool _isDigit(int charCode) => (charCode ^ _zero) <= 9;
+
+/// Check if the digit at [index] is continuing a non-zero number.
+///
+/// If there is no non-zero digits before, then leading zeros at [index]
+/// are also ignored when comparing numerically. If there is a non-zero digit
+/// before, then zeros at [index] are significant.
+bool _isNumberSuffix(String string, int index) {

[sra1, 2015/11/18 23:40:16]

maybe call it _isNonZeroNumberSuffix

[Lasse Reichstein Nielsen, 2015/11/19 07:15:37]

Done.

+  while (--index >= 0) {
+    int char = string.codeUnitAt(index);
+    if (char != _zero) return _isDigit(char);
+  }
+  return false;
+}