test_bug340714.js (5354B)
1 /* Test case for bug 340714 2 * 3 * Uses nsIConverterInputStream to decode UTF-16 text with all combinations 4 * of UTF-16BE and UTF-16LE with and without BOM. 5 * 6 * Sample text is: "Все счастливые семьи похожи друг на друга, каждая несчастливая семья несчастлива по-своему." 7 * 8 * The enclosing quotation marks are included in the sample text to test that 9 * UTF-16LE is recognized even when there is no BOM and the UTF-16LE decoder is 10 * not explicitly called. This only works when the first character of the text 11 * is an eight-bit character. 12 */ 13 14 const { NetUtil } = ChromeUtils.importESModule( 15 "resource://gre/modules/NetUtil.sys.mjs" 16 ); 17 18 const beBOM = "%FE%FF"; 19 const leBOM = "%FF%FE"; 20 const sampleUTF16BE = 21 "%00%22%04%12%04%41%04%35%00%20%04%41%04%47%04%30%04%41%04%42%04%3B%04%38%04%32%04%4B%04%35%00%20%04%41%04%35%04%3C%04%4C%04%38%00%20%04%3F%04%3E%04%45%04%3E%04%36%04%38%00%20%04%34%04%40%04%43%04%33%00%20%04%3D%04%30%00%20%04%34%04%40%04%43%04%33%04%30%00%2C%00%20%04%3A%04%30%04%36%04%34%04%30%04%4F%00%20%04%3D%04%35%04%41%04%47%04%30%04%41%04%42%04%3B%04%38%04%32%04%30%04%4F%00%20%04%41%04%35%04%3C%04%4C%04%4F%00%20%04%3D%04%35%04%41%04%47%04%30%04%41%04%42%04%3B%04%38%04%32%04%30%00%20%04%3F%04%3E%00%2D%04%41%04%32%04%3E%04%35%04%3C%04%43%00%2E%00%22"; 22 const sampleUTF16LE = 23 "%22%00%12%04%41%04%35%04%20%00%41%04%47%04%30%04%41%04%42%04%3B%04%38%04%32%04%4B%04%35%04%20%00%41%04%35%04%3C%04%4C%04%38%04%20%00%3F%04%3E%04%45%04%3E%04%36%04%38%04%20%00%34%04%40%04%43%04%33%04%20%00%3D%04%30%04%20%00%34%04%40%04%43%04%33%04%30%04%2C%00%20%00%3A%04%30%04%36%04%34%04%30%04%4F%04%20%00%3D%04%35%04%41%04%47%04%30%04%41%04%42%04%3B%04%38%04%32%04%30%04%4F%04%20%00%41%04%35%04%3C%04%4C%04%4F%04%20%00%3D%04%35%04%41%04%47%04%30%04%41%04%42%04%3B%04%38%04%32%04%30%04%20%00%3F%04%3E%04%2D%00%41%04%32%04%3E%04%35%04%3C%04%43%04%2E%00%22%00"; 24 const expected = 25 '"\u0412\u0441\u0435 \u0441\u0447\u0430\u0441\u0442\u043B\u0438\u0432\u044B\u0435 \u0441\u0435\u043C\u044C\u0438 \u043F\u043E\u0445\u043E\u0436\u0438 \u0434\u0440\u0443\u0433 \u043D\u0430 \u0434\u0440\u0443\u0433\u0430, \u043A\u0430\u0436\u0434\u0430\u044F \u043D\u0435\u0441\u0447\u0430\u0441\u0442\u043B\u0438\u0432\u0430\u044F \u0441\u0435\u043C\u044C\u044F \u043D\u0435\u0441\u0447\u0430\u0441\u0442\u043B\u0438\u0432\u0430 \u043F\u043E-\u0441\u0432\u043E\u0435\u043C\u0443."'; 26 27 Services.prefs.setBoolPref("security.allow_eval_with_system_principal", true); 28 registerCleanupFunction(() => { 29 Services.prefs.clearUserPref("security.allow_eval_with_system_principal"); 30 }); 31 32 function makeText(withBOM, charset) { 33 const isBE = charset === "UTF16BE"; 34 const sampleText = isBE ? sampleUTF16BE : sampleUTF16LE; 35 const bom = isBE ? beBOM : leBOM; 36 return withBOM ? bom + sampleText : sampleText; 37 } 38 39 function testCase(withBOM, charset, charsetDec, decoder, bufferLength) { 40 var dataURI = 41 "data:text/plain;charset=" + charsetDec + "," + makeText(withBOM, charset); 42 43 var ConverterInputStream = Components.Constructor( 44 "@mozilla.org/intl/converter-input-stream;1", 45 "nsIConverterInputStream", 46 "init" 47 ); 48 49 var channel = NetUtil.newChannel({ 50 uri: dataURI, 51 loadUsingSystemPrincipal: true, 52 }); 53 var testInputStream = channel.open(); 54 var testConverter = new ConverterInputStream( 55 testInputStream, 56 decoder, 57 bufferLength, 58 0xfffd 59 ); 60 61 if (!(testConverter instanceof Ci.nsIUnicharLineInputStream)) { 62 throw new Error("not line input stream"); 63 } 64 65 var outStr = ""; 66 var more; 67 do { 68 // read the line and check for eof 69 var line = {}; 70 more = testConverter.readLine(line); 71 outStr += line.value; 72 } while (more); 73 74 if (outStr != expected) { 75 dump( 76 "Failed with BOM = " + 77 withBOM + 78 "; charset = " + 79 charset + 80 "; charset declaration = " + 81 charsetDec + 82 "; decoder = " + 83 decoder + 84 "; bufferLength = " + 85 bufferLength + 86 "\n" 87 ); 88 if (outStr.length == expected.length) { 89 for (let i = 0; i < outStr.length; ++i) { 90 if (outStr.charCodeAt(i) != expected.charCodeAt(i)) { 91 dump( 92 i + 93 ": " + 94 outStr.charCodeAt(i).toString(16) + 95 " != " + 96 expected.charCodeAt(i).toString(16) + 97 "\n" 98 ); 99 } 100 } 101 } 102 } 103 104 // escape the strings before comparing for better readability 105 Assert.equal(escape(outStr), escape(expected)); 106 } 107 108 function run_test() { 109 /* BOM charset charset decoder buffer 110 declaration length */ 111 testCase(true, "UTF16LE", "UTF-16", "UTF-16BE", 64); 112 testCase(true, "UTF16BE", "UTF-16", "UTF-16LE", 64); 113 testCase(true, "UTF16LE", "UTF-16", "UTF-16LE", 64); 114 testCase(true, "UTF16BE", "UTF-16", "UTF-16BE", 64); 115 testCase(false, "UTF16LE", "UTF-16", "UTF-16LE", 64); 116 testCase(false, "UTF16BE", "UTF-16", "UTF-16BE", 64); 117 testCase(true, "UTF16LE", "UTF-16", "UTF-16BE", 65); 118 testCase(true, "UTF16BE", "UTF-16", "UTF-16LE", 65); 119 testCase(true, "UTF16LE", "UTF-16", "UTF-16LE", 65); 120 testCase(true, "UTF16BE", "UTF-16", "UTF-16BE", 65); 121 testCase(false, "UTF16LE", "UTF-16", "UTF-16LE", 65); 122 testCase(false, "UTF16BE", "UTF-16", "UTF-16BE", 65); 123 }