michael@0: /* Test case for bug 340714
michael@0:  *
michael@0:  * Uses nsIConverterInputStream to decode UTF-16 text with all combinations
michael@0:  * of UTF-16BE and UTF-16LE with and without BOM.
michael@0:  *
michael@0:  * Sample text is: "Все счастливые семьи похожи друг на друга, каждая несчастливая семья несчастлива по-своему."
michael@0:  *
michael@0:  * The enclosing quotation marks are included in the sample text to test that
michael@0:  * UTF-16LE is recognized even when there is no BOM and the UTF-16LE decoder is
michael@0:  * not explicitly called. This only works when the first character of the text
michael@0:  * is an eight-bit character.
michael@0:  */
michael@0: 
michael@0: const beBOM="%FE%FF";
michael@0: const leBOM="%FF%FE";
michael@0: const sampleUTF16BE="%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";
michael@0: const sampleUTF16LE="%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";
michael@0: const expected = "\"\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.\""; 
michael@0: 
michael@0: function makeText(withBOM, charset)
michael@0: {
michael@0:   var theText = eval("sample" + charset);
michael@0:   if (withBOM) {
michael@0:     if (charset == "UTF16BE") {
michael@0:       theText = beBOM + theText;
michael@0:     } else {
michael@0:       theText = leBOM + theText;
michael@0:     }
michael@0:   }
michael@0:   return theText;
michael@0: }
michael@0: 
michael@0: function testCase(withBOM, charset, charsetDec, decoder, bufferLength)
michael@0: {
michael@0:   var dataURI = "data:text/plain;charset=" + charsetDec + "," +
michael@0:                  makeText(withBOM, charset);
michael@0: 
michael@0:   var IOService = Components.Constructor("@mozilla.org/network/io-service;1",
michael@0: 					 "nsIIOService");
michael@0:   var ConverterInputStream =
michael@0:       Components.Constructor("@mozilla.org/intl/converter-input-stream;1",
michael@0: 			     "nsIConverterInputStream",
michael@0: 			     "init");
michael@0: 
michael@0:   var ios = new IOService();
michael@0:   var channel = ios.newChannel(dataURI, "", null);
michael@0:   var testInputStream = channel.open();
michael@0:   var testConverter = new ConverterInputStream(testInputStream,
michael@0: 					       decoder,
michael@0: 					       bufferLength,
michael@0: 					       0xFFFD);
michael@0: 
michael@0:   if (!(testConverter instanceof
michael@0: 	Components.interfaces.nsIUnicharLineInputStream))
michael@0:       throw "not line input stream";
michael@0: 
michael@0:   var outStr = "";
michael@0:   var more;
michael@0:   do {
michael@0:       // read the line and check for eof
michael@0:       var line = {};
michael@0:       more = testConverter.readLine(line);
michael@0:       outStr += line.value;
michael@0:   } while (more);
michael@0: 
michael@0:   if (outStr != expected) {
michael@0:     dump("Failed with BOM = " + withBOM + "; charset = " + charset +
michael@0: 	 "; charset declaration = " + charsetDec + "; decoder = " + decoder +
michael@0: 	 "; bufferLength = " + bufferLength + "\n");
michael@0:     if (outStr.length == expected.length) {
michael@0:       for (i = 0; i < outStr.length; ++i) {
michael@0: 	if (outStr.charCodeAt(i) != expected.charCodeAt(i)) {
michael@0: 	  dump(i + ": " + outStr.charCodeAt(i).toString(16) + " != " + expected.charCodeAt(i).toString(16) + "\n");
michael@0: 	}
michael@0:       }
michael@0:     }
michael@0:   }
michael@0: 
michael@0:   // escape the strings before comparing for better readability
michael@0:   do_check_eq(escape(outStr), escape(expected));
michael@0: }
michael@0: 
michael@0: function run_test()
michael@0: {
michael@0:     /*       BOM    charset    charset   decoder     buffer
michael@0:                                declaration           length */
michael@0:     testCase(true,  "UTF16LE", "UTF-16", "UTF-16",   64);
michael@0:     testCase(true,  "UTF16BE", "UTF-16", "UTF-16",   64);
michael@0:     testCase(true,  "UTF16LE", "UTF-16", "UTF-16LE", 64);
michael@0:     testCase(true,  "UTF16BE", "UTF-16", "UTF-16BE", 64);
michael@0:     testCase(false, "UTF16LE", "UTF-16", "UTF-16",   64);
michael@0:     testCase(false, "UTF16BE", "UTF-16", "UTF-16",   64);
michael@0:     testCase(false, "UTF16LE", "UTF-16", "UTF-16LE", 64);
michael@0:     testCase(false, "UTF16BE", "UTF-16", "UTF-16BE", 64);
michael@0:     testCase(true,  "UTF16LE", "UTF-16", "UTF-16",   65);
michael@0:     testCase(true,  "UTF16BE", "UTF-16", "UTF-16",   65);
michael@0:     testCase(true,  "UTF16LE", "UTF-16", "UTF-16LE", 65);
michael@0:     testCase(true,  "UTF16BE", "UTF-16", "UTF-16BE", 65);
michael@0:     testCase(false, "UTF16LE", "UTF-16", "UTF-16",   65);
michael@0:     testCase(false, "UTF16BE", "UTF-16", "UTF-16",   65);
michael@0:     testCase(false, "UTF16LE", "UTF-16", "UTF-16LE", 65);
michael@0:     testCase(false, "UTF16BE", "UTF-16", "UTF-16BE", 65);
michael@0: }