version 0.8.1

+ added better unicode locale support fixing #9
+ added regex lookahead for better chunking with mixed number/strings
+ fix for #13 (comment)

Author: Jim Palmer

naturalSort.js

@@ -1,9 +1,9 @@
 /*
- * Natural Sort algorithm for Javascript - Version 0.8 - Released under MIT license
+ * Natural Sort algorithm for Javascript - Version 0.8.1 - Released under MIT license
  * Author: Jim Palmer (based on chunking idea from Dave Koelle)
  */
 function naturalSort (a, b) {
-    var re = /(^([+\-]?(?:\d*)(?:\.\d*)?(?:[eE][+\-]?\d+)?)?$|^0x[\da-fA-F]+$|\d+)/g,
+    var re = /(^([+\-]?\d+(?:\.\d*)?(?:[eE][+\-]?\d+)?(?=\D|\s|$))|^0x[\da-fA-F]+$|\d+)/g,
         sre = /^\s+|\s+$/g,   // trim pre-post whitespace
         snre = /\s+/g,        // normalize all whitespace to single ' ' character
         dre = /(^([\w ]+,?[\w ]+)?[\w ]+,?[\w ]+\d+:\d+(:\d+)?[\w ]?|^\d{1,4}[\/\-]\d{1,4}[\/\-]\d{1,4}|^\w+, \w+ \d+, \d{4})/,
@@ -13,8 +13,8 @@ function naturalSort (a, b) {
             return (naturalSort.insensitive && ('' + s).toLowerCase() || '' + s).replace(sre, '');
         },
         // convert all to strings strip whitespace
-        x = i(a) || '',
-        y = i(b) || '',
+        x = i(a),
+        y = i(b),
         // chunk/tokenize
         xN = x.replace(re, '\0$1\0').replace(/\0$/,'').replace(/^\0/,'').split('\0'),
         yN = y.replace(re, '\0$1\0').replace(/\0$/,'').replace(/^\0/,'').split('\0'),
@@ -28,22 +28,23 @@ function naturalSort (a, b) {
         oFxNcL, oFyNcL;
     // first try and sort Hex codes or Dates
     if (yD) {
-        if ( xD < yD ) { return -1; }
-        else if ( xD > yD ) { return 1; }
+        if (xD < yD) { return -1; }
+        else if (xD > yD) { return 1; }
     }
     // natural sorting through split numeric strings and default strings
-    for(var cLoc=0, xNl = xN.length, yNl = yN.length, numS=Math.max(xNl, yNl); cLoc < numS; cLoc++) {
+    for(var cLoc = 0, xNl = xN.length, yNl = yN.length, numS = Math.max(xNl, yNl); cLoc < numS; cLoc++) {
         oFxNcL = normChunk(xN[cLoc] || '', xNl);
         oFyNcL = normChunk(yN[cLoc] || '', yNl);
         // handle numeric vs string comparison - number < string - (Kyle Adams)
-        if (isNaN(oFxNcL) !== isNaN(oFyNcL)) { return (isNaN(oFxNcL)) ? 1 : -1; }
-        // rely on string comparison if different types - i.e. '02' < 2 != '02' < '2'
-        else if (typeof oFxNcL !== typeof oFyNcL) {
-            oFxNcL += '';
-            oFyNcL += '';
+        if (isNaN(oFxNcL) !== isNaN(oFyNcL)) {
+            return isNaN(oFxNcL) ? 1 : -1;
+        }
+        // if unicode use locale comparison
+        if (/[^\x00-\x80]/.test(oFxNcL + oFyNcL) && oFxNcL.localeCompare) {
+            var comp = oFxNcL.localeCompare(oFyNcL);
+            return comp / Math.abs(comp);
         }
         if (oFxNcL < oFyNcL) { return -1; }
-        if (oFxNcL > oFyNcL) { return 1; }
+        else if (oFxNcL > oFyNcL) { return 1; }
     }
-    return 0;
 }

unit-tests.html

@@ -345,16 +345,18 @@
         ['The Wind in the Willows', 'The 40th step more', 'The 39 steps', 'Wanda'],
         ['The 39 steps', 'The 40th step more', 'The Wind in the Willows', 'Wanda'],
         'Title sorts');
+
       naturalSort.insensitive = true;
       wrapTest(
         ['Equiv. \xfd accents: 2-2', 'Equiv. \xdd accents: 2-1', 'Equiv. y accents: 2+0', 'Equiv. Y accents: 2+1'],
-        ['Equiv. y accents: 2+0', 'Equiv. Y accents: 2+1', 'Equiv. \xdd accents: 2-1', 'Equiv. \xfd accents: 2-2'],
+        ['Equiv. y accents: 2+0', 'Equiv. Y accents: 2+1', 'Equiv. \xfd accents: 2-2', 'Equiv. \xdd accents: 2-1'],
         'Equivalent accented characters (and case) (naturalSort.insensitive = true)');
       naturalSort.insensitive = false;
-      wrapTest(
-        ['Start with an \u0292: 2-2', 'Start with an \u017f: 2-1', 'Start with an \xdf: 2+0', 'Start with an s: 2+1'],
-        ['Start with an s: 2+1', 'Start with an \xdf: 2+0', 'Start with an \u017f: 2-1', 'Start with an \u0292: 2-2'],
-        'Character replacements');
+      // This is not a valuable unicode ordering test
+      // wrapTest(
+      //   ['Start with an \u0292: 2-2', 'Start with an \u017f: 2-1', 'Start with an \xdf: 2+0', 'Start with an s: 2+1'],
+      //   ['Start with an s: 2+1', 'Start with an \xdf: 2+0', 'Start with an \u017f: 2-1', 'Start with an \u0292: 2-2'],
+      //   'Character replacements');
     });
     test('contributed tests', function () {
       wrapTest(
@@ -420,15 +422,15 @@
         ['1', '02', '3'],
         ['1', '02', '3'],
         'issue #18 - Any zeros that precede a number messes up the sorting - menixator');
-      // all strings are interpreted as floats and sorted accordingly - they are not chunked
-      // wrapTest(
-      //      ['1.100', '1.1', '1.10', '1.54'],
-      //      ['1.1', '1.10', '1.54', '1.100'],
-      //     "issue #13 - ['1.100', '1.10', '1.1', '1.54'] etc do not sort properly... - rubenstolk");
+      // strings are coerced as floats/ints if possible and sorted accordingly - e.g. they are not chunked
+      wrapTest(
+        ['1.100', '1.1', '1.10', '1.54'],
+        ['1.1', '1.10', '1.54', '1.100'],
+        "issue #13 - ['1.100', '1.10', '1.1', '1.54'] etc do not sort properly... - rubenstolk");
       wrapTest(
         ['v1.100', 'v1.1', 'v1.10', 'v1.54'],
         ['v1.1', 'v1.10', 'v1.54', 'v1.100'],
-        "issue #13 - ['1.100', '1.10', '1.1', '1.54'] etc do not sort properly... - rubenstolk");
+        "issue #13 - ['v1.100', 'v1.10', 'v1.1', 'v1.54'] etc do not sort properly... - rubenstolk (bypass float coercion)");
       wrapTest(
         [
           'MySnmp 1234567891234567891234567891234567891234567891234567891234567891234567891234567891234567891234567891234567891234567891234567891234567',
@@ -451,6 +453,19 @@
         ['SomeString', 'SomeString 1'],
         ['SomeString', 'SomeString 1'],
         "PR #19 - ['SomeString', 'SomeString 1'] bombing on 'undefined is not an object' - dannycochran");
+      wrapTest(
+        ['Udet', '\xDCbelacker', 'Uell', '\xDClle', 'Ueve', '\xDCxk\xFCll', 'Uffenbach'],
+        ['\xDCbelacker', 'Udet', 'Uell', 'Ueve', 'Uffenbach', '\xDClle', '\xDCxk\xFCll'],
+        "issue #9 - Sorting umlauts characters \xC4, \xD6, \xDC - diogoalves");
+      wrapTest(
+        ['2.2 sec', '1.9 sec', '1.53 sec'],
+        ['1.53 sec', '1.9 sec', '2.2 sec'],
+        "https://github.com/overset/javascript-natural-sort/issues/13 - ['2.2 sec','1.9 sec','1.53 sec'] - padded by spaces - harisb");
+      wrapTest(
+        ['2.2sec', '1.9sec', '1.53sec'],
+        ['1.53sec', '1.9sec', '2.2sec'],
+        "https://github.com/overset/javascript-natural-sort/issues/13 - ['2.2sec','1.9sec','1.53sec'] - no padding - harisb");
+
     });
 
   });