dom/encoding/test/unit/test_utf.js

Wed, 31 Dec 2014 06:09:35 +0100

author
Michael Schloh von Bennewitz <michael@schloh.com>
date
Wed, 31 Dec 2014 06:09:35 +0100
changeset 0
6474c204b198
permissions
-rw-r--r--

Cloned upstream origin tor-browser at tor-browser-31.3.0esr-4.5-1-build1
revision ID fc1c9ff7c1b2defdbc039f12214767608f46423f for hacking purpose.

     1 // NOTE: Requires testharness.js
     2 // http://www.w3.org/2008/webapps/wiki/Harness
     4 // Extension to testharness.js API which avoids logging enormous strings
     5 // on a coding failure.
     6 function assert_string_equals(actual, expected, description) {
     7   // short circuit success case
     8   if (actual === expected) {
     9     assert_true(true, description + ": <actual> === <expected>");
    10     return;
    11   }
    13   // length check
    14   assert_equals(actual.length, expected.length,
    15                 description + ": string lengths")
    17   var i, a, b;
    18   for (i = 0; i < actual.length; i++) {
    19     a = actual.charCodeAt(i);
    20     b = expected.charCodeAt(i);
    21     if (a !== b)
    22       assert_true(false,
    23                   description +
    24                   ": code unit " + i.toString() + " unequal: " +
    25                   cpname(a) + " != " + cpname(b)); // doesn't return
    26   }
    28   // It should be impossible to get here, because the initial
    29   // comparison failed, so either the length comparison or the
    30   // codeunit-by-codeunit comparison should also fail.
    31   assert_true(false, description + ": failed to detect string difference");
    32 }
    34 // Inspired by:
    35 // http://ecmanaut.blogspot.com/2006/07/encoding-decoding-utf8-in-javascript.html
    36 function encode_utf8(string) {
    37   var utf8 = unescape(encodeURIComponent(string));
    38   var octets = new Uint8Array(utf8.length), i;
    39   for (i = 0; i < utf8.length; i += 1) {
    40     octets[i] = utf8.charCodeAt(i);
    41   }
    42   return octets;
    43 }
    45 function decode_utf8(octets) {
    46   var utf8 = String.fromCharCode.apply(null, octets);
    47   return decodeURIComponent(escape(utf8));
    48 }
    50 // Helpers for test_utf_roundtrip.
    51 function cpname(n) {
    52   if (n+0 !== n)
    53     return n.toString();
    54   var w = (n <= 0xFFFF) ? 4 : 6;
    55   return 'U+' + ('000000' + n.toString(16).toUpperCase()).slice(-w);
    56 }
    58 function genblock(from, len) {
    59   var i, j, point, offset;
    60   var size, block;
    62   // determine size required:
    63   //    1 unit   for each point from U+000000 through U+00D7FF
    64   //    0 units                      U+00D800 through U+00DFFF
    65   //    1 unit                       U+00E000 through U+00FFFF
    66   //    2 units                      U+010000 through U+10FFFF
    67   function overlap(min1, max1, min2, max2) {
    68     return Math.max(0, Math.min(max1, max2) - Math.max(min1, min2));
    69   }
    70   size = (overlap(from, from+len, 0x000000, 0x00D800) +
    71           overlap(from, from+len, 0x00E000, 0x010000) +
    72           overlap(from, from+len, 0x010000, 0x110000)*2);
    74   block = new Uint16Array(size);
    75   for (i = 0, j = 0; i < len; i++) {
    76     point = from + i;
    77     if (0xD800 <= point && point <= 0xDFFF)
    78       continue;
    79     else if (point <= 0xFFFF)
    80       block[j++] = point;
    81     else {
    82       offset = point - 0x10000;
    83       block[j++] = 0xD800 + (offset >> 10);
    84       block[j++] = 0xDC00 + (offset & 0x3FF);
    85     }
    86   }
    87   return String.fromCharCode.apply(null, block);
    88 }
    90 function test_utf_roundtrip () {
    91   var MIN_CODEPOINT = 0;
    92   var MAX_CODEPOINT = 0x10FFFF;
    93   var BLOCK_SIZE = 0x1000;
    95   var block, block_tag, i, j, encoded, decoded, exp_encoded, exp_decoded;
    97   var TE_U16LE = new TextEncoder("UTF-16LE");
    98   var TD_U16LE = new TextDecoder("UTF-16LE");
   100   var TE_U16BE = new TextEncoder("UTF-16BE");
   101   var TD_U16BE = new TextDecoder("UTF-16BE");
   103   var TE_U8    = new TextEncoder("UTF-8");
   104   var TD_U8    = new TextDecoder("UTF-8");
   106   for (i = MIN_CODEPOINT; i < MAX_CODEPOINT; i += BLOCK_SIZE) {
   107     block_tag = cpname(i) + " - " + cpname(i + BLOCK_SIZE - 1);
   108     block = genblock(i, BLOCK_SIZE);
   110     // test UTF-16LE, UTF-16BE, and UTF-8 encodings against themselves
   111     encoded = TE_U16LE.encode(block);
   112     decoded = TD_U16LE.decode(encoded);
   113     assert_string_equals(block, decoded, "UTF-16LE round trip " + block_tag);
   115     encoded = TE_U16BE.encode(block);
   116     decoded = TD_U16BE.decode(encoded);
   117     assert_string_equals(block, decoded, "UTF-16BE round trip " + block_tag);
   119     encoded = TE_U8.encode(block);
   120     decoded = TD_U8.decode(encoded);
   121     assert_string_equals(block, decoded, "UTF-8 round trip " + block_tag);
   123     // test TextEncoder(UTF-8) against the older idiom
   124     exp_encoded = encode_utf8(block);
   125     assert_array_equals(encoded, exp_encoded,
   126                         "UTF-8 reference encoding " + block_tag);
   128     exp_decoded = decode_utf8(exp_encoded);
   129     assert_string_equals(decoded, exp_decoded,
   130                          "UTF-8 reference decoding " + block_tag);
   131   }
   132 }
   134 function test_utf_samples () {
   135   // z, cent, CJK water, G-Clef, Private-use character
   136   var sample = "z\xA2\u6C34\uD834\uDD1E\uDBFF\uDFFD";
   137   var cases = [
   138     { encoding: "utf-8",
   139       expected: [0x7A, 0xC2, 0xA2, 0xE6, 0xB0, 0xB4, 0xF0, 0x9D, 0x84, 0x9E, 0xF4, 0x8F, 0xBF, 0xBD] },
   140     { encoding: "utf-16le",
   141       expected: [0x7A, 0x00, 0xA2, 0x00, 0x34, 0x6C, 0x34, 0xD8, 0x1E, 0xDD, 0xFF, 0xDB, 0xFD, 0xDF] },
   142     { encoding: "utf-16",
   143       expected: [0x7A, 0x00, 0xA2, 0x00, 0x34, 0x6C, 0x34, 0xD8, 0x1E, 0xDD, 0xFF, 0xDB, 0xFD, 0xDF] },
   144     { encoding: "utf-16be",
   145       expected: [0x00, 0x7A, 0x00, 0xA2, 0x6C, 0x34, 0xD8, 0x34, 0xDD, 0x1E, 0xDB, 0xFF, 0xDF, 0xFD] }
   146   ];
   148   cases.forEach(
   149     function(t) {
   150       var encoded = new TextEncoder(t.encoding).encode(sample);
   151       assert_array_equals(encoded, t.expected,
   152                           "expected equal encodings - " + t.encoding);
   154       var decoded = new TextDecoder(t.encoding)
   155                         .decode(new Uint8Array(t.expected));
   156       assert_equals(decoded, sample,
   157                     "expected equal decodings - " + t.encoding);
   158     });
   159 }
   161 test(test_utf_samples,
   162      "UTF-8, UTF-16LE, UTF-16BE - Encode/Decode - reference sample");
   164 test(test_utf_roundtrip,
   165      "UTF-8, UTF-16LE, UTF-16BE - Encode/Decode - full roundtrip and "+
   166      "agreement with encode/decodeURIComponent");

mercurial