]> git.pond.sub.org Git - eow/blobdiff - static/dojo-release-1.1.1/dojo/tests/number.js
add Dojo 1.1.1
[eow] / static / dojo-release-1.1.1 / dojo / tests / number.js
diff --git a/static/dojo-release-1.1.1/dojo/tests/number.js b/static/dojo-release-1.1.1/dojo/tests/number.js
new file mode 100644 (file)
index 0000000..902484a
--- /dev/null
@@ -0,0 +1,966 @@
+if(!dojo._hasResource["tests.number"]){ //_hasResource checks added by build. Do not use _hasResource directly in your code.
+dojo._hasResource["tests.number"] = true;
+dojo.provide("tests.number");
+
+dojo.require("dojo.number");
+
+/**
+ * Refer to ICU4J's NumberFormatTest.expect(...) 
+ */
+tests.number.check=function(t,options,sourceInput,expectResult){
+       tests.number.checkFormatParseCycle(t, t,options,sourceInput,expectResult,false);
+       tests.number.checkParse(t, t,options,expectResult,sourceInput); 
+}
+
+/**
+ * Perform a single formatting check or a backward check
+ * backward check:number1 -(formatted)-> string1 -(parsed)-> number2 -(formated)-> string2
+ * then number should == number2; string1 should == string2
+ */
+tests.number.checkFormatParseCycle=function(t,options,sourceInput,expectResult,
+                backwardCheck/*boolean,indicates whether need a backward chain check,like formate->parse->format*/){   
+       if(null != options){
+               var pattern = options.pattern;
+               var locale = options.locale;
+               //TODO: add more fields
+       }
+       
+       //print("\n");
+       var str = null==pattern?"default":pattern;
+       //print("pattern:" + str + "| locale:" + locale);
+       //print("input:" + sourceInput);
+       var result = dojo.number.format(sourceInput,options);
+       //print("result:" + result);
+       if(null != expectResult){
+           t.is(expectResult,result);
+       }
+       if(backwardCheck){
+               var resultParsed = dojo.number.parse(result,options);
+               //print("resultParsed:" + resultParsed);                
+               if(!tests.number._decimalNumberDiff(sourceInput,resultParsed)){ 
+                   t.is(sourceInput,resultParsed);
+               }               
+               var resultParsedReformatted = dojo.number.format(resultParsed,options);
+               //print("resultParsedReformatted:" + resultParsedReformatted);
+           if(!tests.number._decimalNumberDiff(result,resultParsedReformatted)){
+                       t.is(result,resultParsedReformatted);
+               }
+       }       
+}
+
+/**
+ * Perform a single parsing check
+ */
+tests.number.checkParse=function(t,options,sourceInput,expectResult){
+       var str = "default";
+       if(null != options && null != options.pattern){ 
+               str = options.pattern;
+       }
+       //print("input:" + sourceInput);
+       var result = dojo.number.parse(sourceInput,options);
+       //print("result :" + result);
+       if(null != expectResult){
+           t.is(expectResult,result);
+       }
+}
+
+/**
+ * //TODO:Round a given number
+ */
+tests.number.rounding = function(t,number,maxFractionDigits,expected){
+       var pattern="#0.";
+       for(var i=0; i<maxFractionDigits; i++){pattern += "#";}
+       var result = dojo.number.format(number,{pattern:pattern});
+       t.is(expected,result);  
+}
+
+/**
+ * Run a batch parsing
+ */
+function runBatchParse(options,dataArray/*array*/,pass/*boolean*/){
+       var exception = null;
+       var result;
+       var i=0;
+       var str = (null==options.pattern)?"default":options.pattern;
+       
+       //print("\n");
+       for(; i<dataArray.length; i++){
+               try{
+                       //print("["+i+"]"+"input:"+dataArray[i]);
+                       result = dojo.number.parse(dataArray[i],options);                       
+                       if(isNaN(result)){
+                               throw "\"" + dataArray[i] + "\" is parsed to NaN with pattern " + str;
+                       }
+                       //print("["+i+"]"+"output:"+result);    
+               }catch(e){
+                       exception = e;
+                       break;
+               }
+       }
+               
+       if(!pass && (exception == null)) {
+               throw "runBatchParse() - stric parse failed, no exception when parsing illegal data"; 
+       }else if(exception != null){
+               if(!pass && i == 0){
+                       //strict parsing should fail for all the dataArray elements as expected
+                       //pass condition for strict parsing
+                       return;
+               }
+               throw "runBatchParse() failed: " + exception;           
+       }
+}
+
+/**
+ * Check whether the given two numbers differ under the decimal bound
+ * 
+ */
+tests.number._decimalNumberDiff = function(num1,num2){
+       //TODO: should be more accurate when dojo.number finish rounding in the future
+       var diffBound = 1e-3;
+       var diff = num1 - num2;
+       //print("Math.abs(diff) " + Math.abs(diff));
+       if(Math.abs(diff) < diffBound ){
+               return true;
+       }else if(isNaN(Math.abs(diff))){        
+               var s = num1.toString().split(num2);
+               s[1] = s[1].replace(",","0");
+               s[1] = s[1].replace('\u066b','0');
+               return (new Number(s[1])< diffBound);
+       }
+       return false;   
+}
+
+tests.register("tests.number", 
+       [
+               {
+                       // Test formatting and parsing of currencies in various locales pre-built in dojo.cldr
+                       // NOTE: we can't set djConfig.extraLocale before bootstrapping unit tests, so directly
+                       // load resources here for specific locales:
+
+                       name: "number",
+                       setUp: function(){
+                               var partLocaleList = ["en-us", "fr-fr", "de-de"];
+
+                               for(var i = 0 ; i < partLocaleList.length; i ++){
+                                       dojo.requireLocalization("dojo.cldr","number",partLocaleList[i], "zh-cn,zh,ko-kr,pt,en-us,en-gb,de,ja,ja-jp,en,ROOT,en-au,fr,es,ko,zh-tw,it,es-es,de-de");
+                               }
+                       },
+                       runTest: function(t){
+                       }
+               },
+               {
+                       name: "format", // old tests
+                       runTest: function(t){
+
+       t.is("0123", dojo.number.format(123, {pattern: "0000"}));
+       t.is("-12,34,567.890", dojo.number.format(-1234567.89, {pattern: "#,##,##0.000##"}));
+       t.is("-12,34,567.89012", dojo.number.format(-1234567.890123, {pattern: "#,##,##0.000##"}));
+       t.is("(1,234,567.89012)", dojo.number.format(-1234567.890123, {pattern: "#,##0.000##;(#,##0.000##)"}));
+       t.is("(1,234,567.89012)", dojo.number.format(-1234567.890123, {pattern: "#,##0.000##;(#)"}));
+       t.is("50.1%", dojo.number.format(0.501, {pattern: "#0.#%"}));
+       t.is("98", dojo.number.format(1998, {pattern: "00"}));
+       t.is("01998", dojo.number.format(1998, {pattern: "00000"}));
+       t.is("0.13", dojo.number.format(0.125, {pattern: "0.##"})); //NOTE: expects round_half_up, not round_half_even
+       t.is("0.1250", dojo.number.format(0.125, {pattern: "0.0000"}));
+       t.is("0.1", dojo.number.format(0.100004, {pattern: "0.####"}));
+
+       t.is("-12", dojo.number.format(-12.3, {places:0, locale: "en-us"}));
+       t.is("-1,234,567.89", dojo.number.format(-1234567.89, {locale: "en-us"}));
+//     t.is("-12,34,567.89", dojo.number.format(-1234567.89, {locale: "en-in"}));
+       t.is("-1,234,568", dojo.number.format(-1234567.89, {places:0, locale: "en-us"}));
+//     t.is("-12,34,568", dojo.number.format(-1234567.89, {places:0, locale: "en-in"}));
+       t.is("-1\xa0000,10", dojo.number.format(-1000.1, {places:2, locale: "fr-fr"}));
+       t.is("-1,000.10", dojo.number.format(-1000.1, {places:2, locale: "en-us"}));
+       t.is("-1\xa0000,10", dojo.number.format(-1000.1, {places:2, locale: "fr-fr"}));
+       t.is("-1.234,56", dojo.number.format(-1234.56, {places:2, locale: "de-de"}));
+       t.is("-1,000.10", dojo.number.format(-1000.1, {places:2, locale: "en-us"}));
+       t.is("123.46%", dojo.number.format(1.23456, {places:2, locale: "en-us", type: "percent"}));
+
+       //rounding
+       t.is("-1,234,568", dojo.number.format(-1234567.89, {places:0, locale: "en-us"}));
+//     t.is("-12,34,568", dojo.number.format(-1234567.89, {places:0, locale: "en-in"}));
+       t.is("-1,000.11", dojo.number.format(-1000.114, {places:2, locale: "en-us"}));
+       t.is("-1,000.11", dojo.number.format(-1000.115, {places:2, locale: "en-us"}));
+       t.is("-1,000.12", dojo.number.format(-1000.116, {places:2, locale: "en-us"}));
+       t.is("-0.00", dojo.number.format(-0.0001, {places:2, locale: "en-us"}));
+       t.is("0.00", dojo.number.format(0, {places:2, locale: "en-us"}));
+
+       //change decimal places
+       t.is("-1\xa0000,100", dojo.number.format(-1000.1, {places:3, locale: "fr-fr"}));
+       t.is("-1,000.100", dojo.number.format(-1000.1, {places:3, locale: "en-us"}));
+                       }
+               },
+               {
+                       name: "parse", // old tests
+                       runTest: function(t){
+       t.is(1000, dojo.number.parse("1000", {locale: "en-us"}));
+       t.is(1000.123, dojo.number.parse("1000.123", {locale: "en-us"}));
+       t.is(1000, dojo.number.parse("1,000", {locale: "en-us"}));
+       t.is(-1000, dojo.number.parse("-1000", {locale: "en-us"}));
+       t.is(-1000.123, dojo.number.parse("-1000.123", {locale: "en-us"}));
+       t.is(-1234567.89, dojo.number.parse("-1,234,567.89", {locale: "en-us"}));
+       t.is(-1234567.89, dojo.number.parse("-1 234 567,89", {locale: "fr-fr"}));
+       t.t(isNaN(dojo.number.parse("-1 234 567,89", {locale: "en-us"})));
+
+       t.t(isNaN(dojo.number.parse("10,00", {locale: "en-us"})));
+       t.t(isNaN(dojo.number.parse("1000.1", {locale: "fr-fr"})));
+
+       t.t(isNaN(dojo.number.parse("")));
+       t.t(isNaN(dojo.number.parse("abcd")));
+
+       //test whitespace
+//     t.is(-1234567, dojo.number.parse("  -1,234,567  ", {locale: "en-us"}));
+
+//     t.t(dojo.number.parse("9.1093826E-31"));
+       t.is(0.501, dojo.number.parse("50.1%", {pattern: "#0.#%"}));
+
+       t.is(123.4, dojo.number.parse("123.4", {pattern: "#0.#"}));
+       t.is(-123.4, dojo.number.parse("-123.4", {pattern: "#0.#"}));
+       t.is(123.4, dojo.number.parse("123.4", {pattern: "#0.#;(#0.#)"}));
+       t.is(-123.4, dojo.number.parse("(123.4)", {pattern: "#0.#;(#0.#)"}));
+
+       t.is(null, dojo.number.format("abcd", {pattern: "0000"}));
+       
+       t.is(123, dojo.number.parse("123", {places:0}));
+       t.is(123, dojo.number.parse("123", {places:'0'}));
+       t.is(123.4, dojo.number.parse("123.4", {places:1}));
+       t.is(123.45, dojo.number.parse("123.45", {places:'1,3'}));
+       t.is(123.45, dojo.number.parse("123.45", {places:'0,2'}));
+                       }
+               },
+               {
+                       name: "format_icu4j3_6",
+                       runTest: function(t){
+
+/*************************************************************************************************
+ * Evan:The following test cases are referred from ICU4J 3.6 (NumberFormatTest etc.) 
+ * see http://icu.sourceforge.net/download/3.6.html#ICU4J
+ *************************************************************************************************/
+
+
+/**
+ * In ICU4J, testing logic for NumberFormat.format() is seperated into 
+ * differernt single tese cases. So part of these logic are 
+ * collected together in this single method.
+ * 
+ * !!Failed cases are as follows:
+ * 1.1234567890987654321234567890987654321 should be formatted as 
+ *   1,234,567,890,987,654,321,234,567,890,987,654,321 with all the default parameters,
+ *   but got 1.234 instead, may due to the unimplemeted exponent.
+ * 2.\u00a4 and ' are not replaced
+ *      with pattern "'*&'' '\u00a4' ''&*' #,##0.00"
+ *   1.0 should be formatted to "*&' Re. '&* 1.00",but got "'*&'' '\u00a4' ''&*' 1.00" instead
+ *   etc.   
+ *             
+ */
+       //print("test_number_format_icu4j3_6() start..............");
+       /* !!Failed case, 1.234 returned instead
+       //refer to ICU4J's NumberFormatTest.TestCoverage() 
+       var bigNum = 1234567890987654321234567890987654321;
+       var expectResult = "1,234,567,890,987,654,321,234,567,890,987,654,321";
+       tests.number.checkFormatParseCycle(t, null,bigNum,expectResult,false);
+       */      
+       
+       //in icu4j should throw out an exception when formatting a string,
+       //but it seems dojo.number.format can deal with strings
+       //return 123,456,789
+       dojo.number.format("123456789");
+       
+       //!!Failed case, \u00a4 and ' are not replaced
+       /*
+       var options = {pattern:"'*&'' '\u00a4' ''&*' #,##0.00",locale:"en-us"};
+       tests.number.check(t, options,1.0, "*&' Re. '&* 1.00");
+       tests.number.check(t, options,-2.0, "-*&' Rs. '&* 2.00");       
+
+       options = {pattern:"#,##0.00 '*&'' '\u00a4' ''&*'",locale:"en-us"};
+       tests.number.check(t, options,1.0,"1.00 *&' Re. '&*");
+       tests.number.check(t, options,-2.0,"-2.00 *&' Rs. '&*");
+       */
+       //print("test_number_format_icu4j3_6() end..............\n");
+                       }
+               },
+               {
+                       name: "format_patterns",
+                       runTest: function(t){
+
+/**
+ * Refer to ICU4J's NumberFormatTest.TestPatterns() which now only coveres us locale   
+ */
+       //print("test_number_format_Patterns() start..............");
+       var patterns = (["#0.#", "#0.", "#.0", "#"]);
+       var patternsLength = patterns.length;    
+       var num = (["0","0", "0.0", "0"]);
+       var options;
+       //icu4j result seems doesn't work as:
+       //var num = (["0","0.", ".0", "0"]);      
+       for (var i=0; i<patternsLength; ++i)
+       {       
+               options = {pattern:patterns[i]};
+               tests.number.checkFormatParseCycle(t, options,0,num[i],false);       
+       }
+       
+       //!!Failed case
+       //In ICU4J:
+       //        unquoted special characters in the suffix are illegal
+       //        so "000.000|###" is illegal; "000.000'|###'" is legal
+       //dojo.number.format:
+       //        when formatting 1.2 with illegal pattern "000.000|###"
+       //                no exception was thrown but got "001.200|###" instead.
+       
+       /*
+       patterns = (["000.000|###","000.000'|###'"]);
+       var exception = false;
+       var result;
+       for(var i = 0; i < patterns.length; i ++){
+               try{
+                       //"001.200'|###'" is return for "000.000'|###'"
+                       //"001.200|###" is return for "000.000|###"
+                       result = dojo.number.format(1.2,{pattern:patterns[i]});
+                       print("["+i+"] 1.2 is formatted to " + result + " with pattern " + patterns[i]);
+               }catch(e){
+                       exception = true;
+               }
+               if(exception && i==1){
+                       throw "["+i+"]Failed when formatting 1.2 using legal pattern " + patterns[i];
+               }else if(!exception && i==0){
+                       throw "["+i+"]Failed when formatting 1.2 using illegal pattern  " + patterns[i];
+               }
+       }*/     
+       //print("test_number_format_Patterns() end..............\n");
+                       }
+               },
+               {
+                       name: "exponential",
+                       runTest: function(t){
+/**
+ * TODO: For dojo.number future version 
+ * Refer to ICU4J's NumberFormatTest.TestExponential()
+ */
+                       }
+               },
+               {
+                       name: "format_quotes",
+                       runTest: function(t){
+/**
+ * TODO: Failed case
+ * Refer to ICU4J's NumberFormatTest.TestQuotes()
+ */
+       //print("test_number_format_Quotes() start..............");
+       //TODO: add more locales 
+       
+       //TODO:!!Failed case    
+       //Pattern "s'aa''s'c#" should format 6666 to "saa'sc6666", but got s'aa''s'c6666 instead
+       // is this case necessary?
+       /*
+       var pattern = "s'aa''s'c#";   
+       var result = dojo.number.format(6666,{pattern:pattern,locale:"en-us"});
+       var expectResult = "saa'sc6666";
+       t.is(expectResult,result);
+       */
+       //print("test_number_format_Quotes() end..............");     
+                       }
+               },
+               {
+                       name: "format_rounding",
+                       runTest: function(t){
+/**
+ * Refer to ICU4J's NumberFormatTest.TestRounding487() and NumberFormatTest.TestRounding()
+ */
+       //print("test_number_format_rounding() start..............");
+       tests.number.rounding(t,0.000179999, 5, "0.00018");
+       tests.number.rounding(t,0.00099, 4, "0.001");
+       tests.number.rounding(t,17.6995, 3, "17.7");
+       tests.number.rounding(t,15.3999, 0, "15");
+       tests.number.rounding(t,-29.6, 0, "-30");
+       
+       //TODO refer to NumberFormatTest.TestRounding()
+       
+       //print("test_number_format_rounding() end..............");
+                       }
+               },
+               {
+                       name: "format_scientific",
+                       runTest: function(t){
+/**
+ * TODO: For dojo.number future version
+ * Refer to ICU4J's NumberFormatTest.TestScientific()- Exponential testing 
+ * Refer to ICU4J's NumberFormatTest.TestScientific2() 
+ * Refer to ICU4J's NumberFormatTest.TestScientificGrouping()
+ */
+                       }
+               },
+               {
+                       name: "format_perMill",
+                       runTest: function(t){
+/**
+ * TODO: Failed case 
+ * Refer to ICU4J's NumberFormatTest.TestPerMill()
+ */
+       //print("test_number_format_PerMill() start..............");
+       var pattern;
+       var result;
+       var expectResult;
+       
+    //TODO: !!Failed case - ###.###\u2030(\u2030 is ‰)
+       //Pattern ###.###\u2030 should format 0.4857 as 485.7\u2030,but got 485.700\u2030 instead    
+       pattern = "###.###\u2030";
+       expectResult = "485.7\u2030";
+       result = dojo.number.format(0.4857,{pattern:pattern});
+       t.is(expectResult,result);
+
+    //TODO: !!Failed mile percent case - ###.###m
+       //Pattern "###.###m" should format 0.4857 to 485.7m, but got 0.485m instead
+       /*
+       pattern = "###.###m";
+       expectResult = "485.7m";
+       result = dojo.number.format(0.4857,{pattern:pattern,locale:"en"});
+       t.is(expectResult,result);      
+       */
+       //print("test_number_format_PerMill() end..............\n");
+                       }
+               },
+               {
+                       name: "format_grouping",
+                       runTest: function(t){
+/**
+ * Only test en-us and en-in 
+ * Refer to ICU4J's NumberFormatTest.TestSecondaryGrouping()
+ */
+       //print("test_number_format_Grouping() start..............");
+       //primary grouping
+       var sourceInput = 123456789;
+       var expectResult = "12,34,56,789";
+       var options = {pattern:"#,##,###",locale:"en-us"};      
+
+       //step1: 123456789 formated=> 12,34,56,789
+       //step2:12,34,56,789 parsed=> 123456789 => formated => 12,34,56,789
+       tests.number.checkFormatParseCycle(t, options,sourceInput,expectResult,true);
+                                                 
+       //TODO: sencondary grouping not implemented yet ?
+       //Pattern "#,###" and secondaryGroupingSize=4 should format 123456789 to "12,3456,789"                  
+       
+       //Special case for "en-in" locale
+       //1876543210 should be formated as 1,87,65,43,210 in "en-in" (India)
+/*
+       sourceInput = 1876543210;
+       expectResult = "1,87,65,43,210";
+       var result = dojo.number.format(sourceInput,{locale:"en-in"});
+       t.is(expectResult,result);   
+*/
+       //print("test_number_format_Grouping() end..............\n");    
+                       }
+               },
+               {
+                       name: "format_pad",
+                       runTest: function(t){
+/**
+ * TODO:!!Failed cases:
+ * According to ICU4J test criteria:
+ * 1.with pattern "*^##.##":
+ *      0 should be formatted to "^^^^0",but got "*^0" instead,
+ *   -1.3 should be formatted to "^-1.3",but got "-*^1.3" instead.
+ *   
+ * 2.with pattern "##0.0####*_ 'g-m/s^2'" :
+ *   0 should be formatted to "0.0______ g-m/s^2",but got ":0.0*_ 'g-m/s^2'" instead
+ *   1.0/3 should be formatted to "0.33333__ g-m/s^2",but got "0.33333*_ 'g-m/s^2'" instead
+ *   
+ * 3.with pattern "*x#,###,###,##0.0#;*x(###,###,##0.0#)":
+ *      -10 should be formatted to "xxxxxxxxxx(10.0)",but got "*x(10.0)" instead.
+ *   10 should be formatted to "xxxxxxxxxxxx10.0",but got "*x10.0" instead.
+ *   ......
+ *   -1120456.37 should be formatted to "xx(1,120,456.37)",but got "*x(1,120,456.37)" instead.
+ *   1120456.37 should be formatted to "xxxx1,120,456.37",but got "*x1,120,456.37" instead.
+ *   -1252045600.37 should be formatted to "(1,252,045,600.37)",but got "*x(1,252,045,600.37)" instead.
+ *   1252045600.37 should be formatted to "10,252,045,600.37",but got "*x10,252,045,600.37" instead.
+ *   
+ * 4.with pattern "#,###,###,##0.0#*x;(###,###,##0.0#*x)"
+ *      -10 should be formatted to (10.0xxxxxxxxxx),but got "(10.0*x)" instead.
+ *   10 should be formatted to "10.0xxxxxxxxxxxx",but got "10.0*x" instead.
+ *   ......
+ *   -1120456.37 should be formatted to "(1,120,456.37xx)",but got "(1,120,456.37*x)" instead.
+ *   1120456.37 should be formatted to "xxxx1,120,456.37",but got "1,120,456.37*x" instead.
+ *   -1252045600.37 should be formatted to "(1,252,045,600.37)",but got "(1,252,045,600.37*x)" instead.
+ *   1252045600.37 should be formatted to ""10,252,045,600.37"",but got "10,252,045,600.37*x" instead.*   
+ * 
+ * Refer to ICU4J's NumberFormatTest.TestPad()
+ */
+/*
+function test_number_format_pad(){
+       var locale = "en-us";
+       print("test_number_format_Pad() start..............");
+       var options = {pattern:"*^##.##",locale:locale};
+
+       tests.number.check(t, options,0,"^^^^0");
+       tests.number.check(t, options,-1.3,"^-1.3");    
+       
+       
+       options = {pattern:"##0.0####*_ 'g-m/s^2'",locale:locale};
+       tests.number.check(t, options,0,"0.0______ g-m/s^2");   
+       tests.number.checkFormatParseCycle(t, options,1.0/3,"0.33333__ g-m/s^2",true);  
+       
+       //exponent not implemented
+       //options = {pattern:"##0.0####E0*_ 'g-m/s^2'",locale:locale};
+       //tests.number.check(t, options,0,"0.0E0______ g-m/s^2");
+       //tests.number.checkFormatParseCycle(t, options,1.0/3,"333.333E-3_ g-m/s^2",true);
+       
+       // Test padding before a sign
+       options = {pattern:"*x#,###,###,##0.0#;*x(###,###,##0.0#)",locale:locale};
+
+       tests.number.check(t, options,-10,"xxxxxxxxxx(10.0)");
+       tests.number.check(t, options,-1000, "xxxxxxx(1,000.0)");
+       tests.number.check(t, options,-1000000, "xxx(1,000,000.0)");
+       tests.number.check(t, options,-100.37, "xxxxxxxx(100.37)");
+       tests.number.check(t, options,-10456.37, "xxxxx(10,456.37)");
+       tests.number.check(t, options,-1120456.37, "xx(1,120,456.37)");
+       tests.number.check(t, options,-112045600.37, "(112,045,600.37)");    
+       tests.number.check(t, options,-1252045600.37, "(1,252,045,600.37)");
+
+
+       tests.number.check(t, options,10, "xxxxxxxxxxxx10.0");
+       tests.number.check(t, options,1000, "xxxxxxxxx1,000.0");
+       tests.number.check(t, options,1000000, "xxxxx1,000,000.0");
+       tests.number.check(t, options,100.37, "xxxxxxxxxx100.37");
+       tests.number.check(t, options,10456.37, "xxxxxxx10,456.37");        
+       tests.number.check(t, options,1120456.37, "xxxx1,120,456.37");
+       tests.number.check(t, options,112045600.37, "xx112,045,600.37");
+       tests.number.check(t, options,10252045600.37, "10,252,045,600.37");
+
+       // Test padding between a sign and a number
+       options = {pattern:"#,###,###,##0.0#*x;(###,###,##0.0#*x)",locale:locale};
+       tests.number.check(t, options, -10, "(10.0xxxxxxxxxx)");
+       tests.number.check(t, options, -1000, "(1,000.0xxxxxxx)");
+       tests.number.check(t, options, -1000000, "(1,000,000.0xxx)");
+       tests.number.check(t, options, -100.37, "(100.37xxxxxxxx)");
+       tests.number.check(t, options, -10456.37, "(10,456.37xxxxx)");    
+       tests.number.check(t, options, -1120456.37, "(1,120,456.37xx)");
+       tests.number.check(t, options, -112045600.37, "(112,045,600.37)");   
+       tests.number.check(t, options, -1252045600.37, "(1,252,045,600.37)");  
+
+       tests.number.check(t, options, 10, "10.0xxxxxxxxxxxx");
+       tests.number.check(t, options, 1000, "1,000.0xxxxxxxxx");
+       tests.number.check(t, options, 1000000, "1,000,000.0xxxxx");
+       tests.number.check(t, options, 100.37, "100.37xxxxxxxxxx");
+       tests.number.check(t, options, 10456.37, "10,456.37xxxxxxx");
+       tests.number.check(t, options, 1120456.37, "1,120,456.37xxxx");
+       tests.number.check(t, options, 112045600.37, "112,045,600.37xx");
+       tests.number.check(t, options, 10252045600.37, "10,252,045,600.37");    
+       
+       //Not implemented yet,refer to NumberFormatTest.TestPatterns2()
+       //For future use - maily test pad patterns
+       print("test_number_format_Pad() end..............");
+}
+*/
+                       }
+               },
+               {
+                       name: "parse_icu4j3_6",
+                       runTest: function(t){
+/**
+ * In ICU4J, testing logic for NumberFormat.parse() is seperated into 
+ * differernt single tese cases. So part of these logic are 
+ * collected together in this test case. * 
+ */
+       //print("test_number_parse_icu4j3_6() start..............");
+       //Refer to ICU4J's NumberFormatTest.TestParse() which is only a rudimentary version
+       var pattern = "00";
+       var str = "0.0";        
+       var result = dojo.number.parse(str,{pattern:pattern});
+       //TODO: add more locales
+//FIXME: is this a valid test?
+//     t.is(0,result);
+
+       /**************************************** tolerant parse *****************************************
+        * refers to ICU4J's NumberFormatTest.TestStrictParse()??
+        * TODO: Seems dojo.number parses string in a tolerant way.  
+        */
+        var options = {locale:"en-us"};
+       /*
+        * TODO: !!Failed case,Should all pass,
+        * but the following elements failed (all parsed to NaN):
+        * [1]-"0 ",[2]-"0.",[3]-"0,",[5]-"0. ",[6]-"0.100,5",
+        * [7]-".00",[9]-"12345, ",[10]-"1,234, ",[12]-"0E" 
+        */
+       var passData = ([
+               "0",           //[0] single zero before end of text is not leading
+        //"0 ",        //[1] single zero at end of number is not leading
+        //"0.",        //[2] single zero before period (or decimal, it's ambiguous) is not leading
+        //"0,",          //[3] single zero before comma (not group separator) is not leading
+        "0.0",         //[4] single zero before decimal followed by digit is not leading
+        //"0. ",       //[5] same as above before period (or decimal) is not leading
+        //"0.100,5",   //[6] comma stops parse of decimal (no grouping)
+        //".00",       //[7] leading decimal is ok, even with zeros
+        "1234567",     //[8] group separators are not required
+        //"12345, ",   //[9] comma not followed by digit is not a group separator, but end of number
+        //"1,234, ",   //[10] if group separator is present, group sizes must be appropriate
+        "1,234,567"   //[11] ...secondary too
+        //,"0E"         //[12]not implemented yet,an exponnent not followed by zero or digits is not an exponent 
+         ]);
+       runBatchParse(options,passData,true/*tolerant parse*/); 
+       
+       /* 
+        * TODO:!!Failed case,should all pass,
+        * but the following failed,
+        * [10]-"1,45 that" implies that we partially parse input
+        */
+       var failData = ([            
+               "00",          //[0] leading zero before zero
+        "012",         //[1] leading zero before digit
+        "0,456",       //[2] leading zero before group separator
+        "1,2",         //[3] wrong number of digits after group separator
+        ",0",          //[4] leading group separator before zero
+        ",1",          //[5] leading group separator before digit
+        ",.02",        //[6] leading group separator before decimal
+        "1,.02",       //[7] group separator before decimal
+        "1,,200",      //[8] multiple group separators
+        "1,45",        //[9] wrong number of digits in primary group
+        //"1,45 that",   //[10] wrong number of digits in primary group
+        "1,45.34",     //[11] wrong number of digits in primary group
+        "1234,567",    //[12] wrong number of digits in secondary group
+        "12,34,567",   //[13] wrong number of digits in secondary group
+        "1,23,456,7890" //[14] wrong number of digits in primary and secondary groups
+               ]);
+       runBatchParse(options,failData,false);
+       
+        options = {pattern:"#,##,##0.#",locale:"en-us"};
+       /*
+        * TODO:!!Failed case,shoudl all pass.
+        
+        * but [1] [2] and [3] failed
+        * should be parsed to 1234567,but NaN instead 
+        */
+       var mixedPassData = ([            
+               "12,34,567"     //[0]
+        //,"12,34,567,"                //[1]
+        //"12,34,567, that",//[2]
+        //"12,34,567 that"     //[3]
+               ]);     
+       runBatchParse(options,mixedPassData,true/*tolerant parse*/);
+       
+       /*
+        * TODO:!!Failed case,should all pass,
+        * but actually mixedFailData[2] and mixedFailData[3] passed.
+        * "12,34,56, that " and [3]-"12,34,56 that" should be parsed to 123456,but NaN instead
+        */
+       var mixedFailData = ([
+        "12,34,56",                    //[0]
+        "12,34,56,"            //[1]
+        //,"12,34,56, that ",//[2]
+        //"12,34,56 that",     //[3]
+               ]);                     
+       runBatchParse(options,mixedFailData,false);
+       
+
+       /**************************************** strict parse ******************************************
+        * TODO:May need to test strict parsing in the future?
+        * e.g. A strict parsing like (with pattern "#,##0.#") 
+        * 1.Leading zeros
+        *              '00', '0123' fail the parse, but '0' and '0.001' pass
+        * 2.Leading or doubled grouping separators
+        *              ',123' and '1,,234" fail
+        * 3.Groups of incorrect length when grouping is used
+        *              '1,23' and '1234,567' fail, but '1234' passes
+        * 4.Grouping separators used in numbers followed by exponents
+        *              '1,234E5' fails, but '1234E5' and '1,234E' pass
+        */     
+       //options={locale:"en",strict:true};
+       //runBatchParse(options,passData,false/*strict parse*/);
+       //runBatchParse(options,failData,false/*strict parse*/);        
+       
+       //options = {pattern:"#,##,##0.#",locale:"en-us",strict:true};
+       //runBatchParse(options,mixedPassData,false/*strict parse*/);
+       //runBatchParse(options,mixedFailData,false/*strict parse*/);           
+
+       //print("test_number_parse_icu4j3_6() end..............\n");
+                       }
+               },
+               {
+                       name: "parse_whitespace",
+                       runTest: function(t){
+/**
+ * TODO:!!Failed case
+ * With pattern "a  b#0c  ",both "a b3456c " and and "a   b1234c   " should be parsed to 3456,but got NaN instead.
+ * 
+ * Refer to ICU4J's NumberFormatTest.TestWhiteSpaceParsing
+ */
+    /*
+       print("test_number_parse_WhiteSpace() start..............");
+       var pattern = "a  b#0c  ";
+       var expectResult = 3456;   
+       result =  dojo.number.parse("a b3456c ",{pattern:pattern,locale:"en-us"});
+       t.is(expectResult,result);      
+       result =  dojo.number.parse("a   b3456c   ",{pattern:pattern,locale:"en-us"});
+       t.is(expectResult,result);
+       print("test_number_parse_WhiteSpace() end..............\n");
+       */   
+                       }
+               },
+/*************************************************************************************************
+ *                            Regression test cases
+ * These test cases are referred to ICU4J's NumberFormatRegressionTest and NumberFormatRegression.
+ * The regression cases in ICU4J are used as unit test cases for bug fixing, 
+ * They are inluced here so that dojo.number may avoid those similar bugs. 
+ *************************************************************************************************/
+               {
+                       name: "number_regression_1",
+                       runTest: function(t){
+/**
+ * Refer to ICU4J's NumberFormatRegressionTest.Test4161100()
+ */
+       tests.number.checkFormatParseCycle(t, {pattern:"#0.#"},-0.09,"-0.1",false);
+                       }
+               },
+               {
+                       name: "number_regression_2",
+                       runTest: function(t){
+/**
+ * !!Failed case,rounding hasn't been implemented yet.
+ * Refer to ICU4J's NumberFormatRegressionTest.Test4408066()
+ */
+       /*
+       var data =   ([-3.75, -2.5, -1.5, 
+                   -1.25, 0,    1.0, 
+                   1.25,  1.5,  2.5, 
+                   3.75,  10.0, 255.5]);
+       var expected = (["-4", "-2", "-2",
+                    "-1", "0",  "1",
+                       "1",  "2",  "2",
+                       "4",  "10", "256"]);
+       var options = {locale:"zh-cn",round:true};
+       for(var i =0; i < data.length; i++){
+               tests.number.checkFormatParseCycle(t, options,data[i],expected[i],false);
+       }       
+
+       data = ([       "-3.75", "-2.5", "-1.5", 
+               "-1.25", "0",    "1.0", 
+               "1.25",  "1.5",  "2.5", 
+               "3.75",  "10.0", "255.5"]);
+       expected =([ -3, -2, -1,
+                 -1, 0,  1,
+                 1,  1,  2,
+                 3,  10, 255]);
+       
+       for(var i =0; i < data.length; i++){
+               tests.number.checkParse(t, options,data[i],expected[i]);
+       }
+       */
+                       }
+               },
+               {
+                       name: "number_regression_3",
+                       runTest: function(t){
+/**
+ * Refer to ICU4J's NumberRegression.Test4087535() and Test4243108()
+ */
+       tests.number.checkFormatParseCycle(t, {places:0},0,"0",false);
+       //TODO:in icu4j,0.1 should be formatted to ".1" when minimumIntegerDigits=0
+       tests.number.checkFormatParseCycle(t, {places:0},0.1,"0",false);        
+       tests.number.checkParse(t, {pattern:"#0.#####"},123.55456,123.55456);
+//!! fails because default pattern only has 3 decimal places
+//     tests.number.checkParse(t, null,123.55456,123.55456);
+       
+       //See whether it fails first format 0.0 ,parse "99.99",and then reformat 0.0 
+       tests.number.checkFormatParseCycle(t, {pattern:"#.#"},0.0,"0",false);
+       tests.number.checkParse(t, null,"99.99",99.99);
+       tests.number.checkFormatParseCycle(t, {pattern:"#.#"},0.0,"0",false);
+                       }
+               },
+               {
+                       name: "number_regression_4",
+                       runTest: function(t){
+/**
+ * TODO:
+ * In ICU -0.0 and -0.0001 should be formatted to "-0" with FieldPosition(0)
+ * dojo.i18n.number format -0.0 to "-0"; -0.0001 to "-0.000100" 
+ * 
+ * Refer to ICU4J's NumberRegression.Test4088503() and Test4106658()
+ */
+       tests.number.checkFormatParseCycle(t, {places:0},123,"123",false);
+       
+       //TODO: differernt from ICU where -0.0 is formatted to "-0"
+       tests.number.checkFormatParseCycle(t, {locale:"en-us"},-0.0,"0",false);
+       
+       //TODO: differernt from ICU where -0.0001 is formatted to "-0"
+       tests.number.checkFormatParseCycle(t, {locale:"en-us",places:6},-0.0001,"-0.000100",false);
+                       }
+               },
+               {
+                       name: "number_regression_5",
+                       runTest: function(t){
+/**
+ * !!Failed case,rounding has not implemented yet.
+ * 0.00159999 should be formatted as 0.0016 but got 0.0015 instead.
+ * Refer to ICU4J's NumberRegression.Test4071492()
+ */
+       //tests.number.checkFormatParseCycle(t, {places:4,round:true},0.00159999,"0.0016",false);       
+                       }
+               },
+               {
+                       name: "number_regression_6",
+                       runTest: function(t){
+/**
+ * Refer to ICU4J's NumberRegression.Test4086575()
+ */
+       var pattern = "###.00;(###.00)";
+       var locale = "fr";
+       var options = {pattern:pattern,locale:locale};
+       
+       //no group separator
+       tests.number.checkFormatParseCycle(t, options,1234,"1234,00",false);
+       tests.number.checkFormatParseCycle(t, options,-1234,"(1234,00)",false); 
+       
+       //space as group separator
+       pattern = "#,###.00;(#,###.00)";
+       options = {pattern:pattern,locale:locale};
+       tests.number.checkFormatParseCycle(t, options,1234,"1\u00a0234,00",false);// Expect 1 234,00
+       tests.number.checkFormatParseCycle(t, options,-1234,"(1\u00a0234,00)",false);  // Expect (1 234,00)
+                       }
+               },
+               {
+                       name: "number_regression_7",
+                       runTest: function(t){
+/**
+ * !!Failed case - expontent has not implemented yet 
+ * shuold format 1.000000000000001E7 to 10000000.00000001, but got 10,000,000.000 instead
+ * Refer to ICU4J's NumberRegression.Test4090489() - loses precision
+ */
+       //tests.number.checkFormatParseCycle(t, null,1.000000000000001E7,"10000000.00000001",false);    
+                       }
+               },
+               {
+                       name: "number_regression_8",
+                       runTest: function(t){
+/** 
+ * !!Failed case
+ * 1.with pattern "#,#00.00 p''ieces;-#,#00.00 p''ieces"
+ *   3456.78 should be formated to "3,456.78 p'ieces",
+ *   but got "3,456.78 p''ieces","''" should be replaced with "'"
+ * 2.with illegal pattern "000.0#0"
+ *      no error for the illegal pattern, and 3456.78 is formatted to 456.780
+ * 3.with illegal pattern "0#0.000"
+ *      no error for the illegal pattern, and 3456.78 is formatted to 3456.780
+ * 
+ * Refer to ICU4J's NumberRegression.Test4092480(),Test4074454() 
+ */
+       var patterns = (["#0000","#000","#00","#0","#"]);
+       var expect = (["0042","042","42","42","42"]);
+       
+       for(var i =0; i < patterns.length; i ++){
+               tests.number.checkFormatParseCycle(t, {pattern:patterns[i]},42,expect[i],false);
+               tests.number.checkFormatParseCycle(t, {pattern:patterns[i]},-42,"-"+expect[i],false);   
+       }
+       
+       tests.number.checkFormatParseCycle(t, {pattern:"#,#00.00;-#.#"},3456.78,"3,456.78",false);
+       //!!Failed case
+       //tests.number.checkFormatParseCycle(t, {pattern:"#,#00.00 p''ieces;-#,#00.00 p''ieces"},3456.78,"3,456.78 p'ieces",false);     
+       //tests.number.checkFormatParseCycle(t, {pattern:"000.0#0"},3456.78,null,false);
+       //tests.number.checkFormatParseCycle(t, {pattern:"0#0.000"},3456.78,null,false);        
+                       }
+               },
+               {
+                       name: "number_regression_9",
+                       runTest: function(t){
+/**
+ * TODO
+ * Refer to ICU4J's NumberRegression.Test4052223()
+ */
+       //TODO:only got NaN,need an illegal pattern exception? 
+       tests.number.checkParse(t, {pattern:"#,#00.00"},"abc3");        
+       
+       //TODO: got NaN instead of 1.222, is it ok?
+       //tests.number.checkParse(t, {pattern:"#,##0.###",locale:"en-us"},"1.222,111",1.222);
+       //tests.number.checkParse(t, {pattern:"#,##0.###",locale:"en-us"},"1.222x111",1.222);
+       
+       //got NaN for illeal input,ok
+       tests.number.checkParse(t, null,"hello: ,.#$@^&**10x"); 
+                       }
+               },
+               {
+                       name: "number_regression_10",
+                       runTest: function(t){
+/**
+ * Refer to ICU4J's NumberRegression.Test4125885()
+ */
+       tests.number.checkFormatParseCycle(t, {pattern:"000.00"},12.34,"012.34",false);
+       tests.number.checkFormatParseCycle(t, {pattern:"+000.00%;-000.00%"},0.1234,"+012.34%",false);
+       tests.number.checkFormatParseCycle(t, {pattern:"##,###,###.00"},9.02,"9.02",false);
+       
+       var patterns =(["#.00", "0.00", "00.00", "#0.0#", "#0.00"]);
+       var expect =  (["1.20", "1.20", "01.20", "1.2",   "1.20" ]);
+       for(var i =0 ; i < patterns.length; i ++){
+               tests.number.checkFormatParseCycle(t, {pattern:patterns[i]},1.2,expect[i],false);
+       }
+                       }
+               },
+               {
+                       name: "number_regression_11",
+                       runTest: function(t){
+/**
+ * TODO:!!Failed case
+ * Make sure that all special characters, when quoted in a suffix or prefix, lose their special meaning.
+ * The detail error info :  
+ * for input 123
+ * pattern:'0'#0'0'; expect:"01230"; but got "'3'#0'0'" instead
+ * pattern:','#0','; expect:",123,"; but got "','123','" instead
+ * pattern:'.'#0'.'; expect:".123."; but got "'.'123'.'" instead
+ * pattern:'‰'#0'‰'; expect:"‰123‰"; but got "'‰'123000'‰'" instead
+ * pattern:'%'#0'%'; expect:"%123%"; but got "'%'12300'%'" instead
+ * pattern:'#'#0'#'; expect:"#123#"; but got "'123'#0'#'" instead
+ * pattern:';'#0';'; expect:";123;"; but got "[dojo-test] FATAL exception raised: 
+ *                                                                                       unable to find a number expression in pattern: '"
+ * pattern:'E'#0'E'; expect:"E123E"; not implemeted yet
+ * pattern:'*'#0'*'; expect:"*123*"; but got "'*'123'*'" instead
+ * pattern:'+'#0'+'; expect:"+123+"; but got "'+'123'+'" instead
+ * pattern:'-'#0'-'; expect:"-123-"; but got "'-'123'-'" instead
+ * 
+ * TODO: is it ok to remain "'" in the formatted result as above??
+ * 
+ * Refer to ICU4J's NumberRegression.Test4212072()
+ */
+/*
+       var specials = ([ '0', ',', '.', '\u2030', '%', '#',';', 'E', '*', '+', '-']);
+       var pattern; 
+       var expect;
+       
+       for(var i=0; i < specials.length; i ++){
+               pattern = "'" + specials[i] + "'#0'" + specials[i] + "'";
+               expect = "" +  specials[i] + "123" +  specials[i];
+               tests.number.checkFormatParseCycle(t, {pattern:pattern,locale:"en-us"},123,expect,false);
+       }
+*/
+                       }
+               },
+               {
+                       name: "number_regression_12",
+                       runTest: function(t){
+/**
+ * TODO: add more rounding test cases, refer to ICU4J's NumberRegression.Test4071005(),Test4071014() etc.. 
+ */
+
+/**
+ * TODO:Decimal format doesnt round a double properly when the number is less than 1
+ * 
+ * Refer to ICU4J's NumberRegression.test4241880()
+ */
+/*
+       var input = ([ .019, .009, .015, .016, .014,
+                       .004, .005, .006, .007, .008,
+                       .5, 1.5, .05, .15, .005,
+                       .015, .0005, .0015]);
+       var patterns = (["##0%", "##0%", "##0%", "##0%", "##0%",
+                        "##0%", "##0%", "##0%", "##0%", "##0%",
+                        "#,##0", "#,##0", "#,##0.0", "#,##0.0", "#,##0.00",
+                        "#,##0.00", "#,##0.000", "#,##0.000"]);
+       var expect =([   "2%", "1%", "2%", "2%", "1%",
+                        "0%", "0%", "1%", "1%", "1%",
+                        "0", "2", "0.0", "0.2", "0.00",
+                        "0.02", "0.000", "0.002",]);
+       for(var i = 0; i <input.length; i ++){
+               tests.number.checkFormatParseCycle(t, {pattern:patterns[i],round:true},input[i],expect[i],false);
+       }
+*/
+                       }
+               }
+       ]
+);
+
+}