从http网站批量下载最新文件（其中文件名在新版本上更改）

Question

我需要一个批处理来从http网站（http://www.rarlab.com/download.htm）下载文件。 从这个网站我只需要32位和64位英语的最新版本 该程序始终列在本网站的顶部。

问题1：网站上有两个以上的文件供下载

问题2：文件名随每个新版本而改变

如何在不知道确切文件名的情况下下载这两个文件（最新版本） （并且没有首先访问网页以找出文件名）??

也许我可以使用wget，curl或aria2执行该任务，但我不知道参数/选项。

有人可以帮我解决这个问题吗？

（请只批量解决方案 - 没有vbs，java，jscript，powershell等。）

谢谢。

抱歉，我忘了说我使用Windows 7 32bit。而且我更喜欢批处理，因为脚本应该能够在所有Windows版本上运行，而无需为不同的Windows版本下载额外的程序或资源工具包（对于必须为windows xp等下载的powershell） - 并且因为我只了解批处理脚本。

Answer 1

这是一个批处理+ JScript混合脚本。我知道你说没有vbs，java，jscript等等，但是你会很难用纯批处理HTML。但这确实符合您的其他标准 - 在所有Windows版本上运行而不必依赖可选软件（如powershell或.Net）。*使用JScript的XMLHTTP对象，您甚至不需要第三方应用程序获取网页内容。

至于不了解JScript，除了一些专有的ActiveX对象外，它就像JavaScript一样。如果您不熟悉JavaScript或正则表达式，我添加了大量的注释来帮助您。希望无论我做什么都不打扰，评论很明显它的作用。

更新

该脚本现在检测系统区域设置，将其与WinRAR下载页面上的语言进行匹配，然后下载该语言版本。

无论如何，使用.bat扩展名保存它并像运行任何其他批处理脚本一样运行它。

@if (@a==@b) @end /*

:: batch script portion

@echo off
setlocal

set "url=http://www.rarlab.com/download.htm"
set /p "savepath=Location to save? [%cd%] "

if "%savepath%"=="" set "savepath=%cd%"

cscript /nologo /e:jscript "%~f0" "%url%" "%savepath%"

goto :EOF

:: JScript portion */

// populate translation from locale identifier hex value to WinRAR language label
// http://msdn.microsoft.com/en-us/library/dd318693.aspx
var abbrev={}, a=function(arr,val){for(var i=0;i<arr.length;i++)abbrev[arr[i]]=val};
a(['1401','3c01','0c01','0801','2001','4001','2801','1c01','3801','2401'],'Arabic');
a(['042b'],'Armenian');
a(['082c','042c'],'Azerbaijani');
a(['0423'],'Belarusian');
a(['0402'],'Bulgarian');
a(['0403'],'Catalan');
a(['7c04'],'Chinese Traditional');
a(['0c04','1404','1004','0004'],'Chinese Simplified');
a(['101a'],'Croatian');
a(['0405'],'Czech');
a(['0406'],'Danish');
a(['0813','0413'],'Dutch');
a(['0425'],'Estonian');
a(['040b'],'Finnish');
a(['080c','0c0c','040c','140c','180c','100c'],'French');
a(['0437'],'Georgian');
a(['0c07','0407','1407','1007','0807'],'German');
a(['0408'],'Greek');
a(['040d'],'Hebrew');
a(['040e'],'Hungarian');
a(['0421'],'Indonesian');
a(['0410','0810'],'Italian');
a(['0411'],'Japanese');
a(['0412'],'Korean');
a(['0427'],'Lithuanian');
a(['042f'],'Macedonian');
a(['0414','0814'],'Norwegian');
a(['0429'],'Persian');
a(['0415'],'Polish');
a(['0816'],'Portuguese');
a(['0416'],'Portuguese Brazilian');
a(['0418'],'Romanian');
a(['0419'],'Russian');
a(['7c1a','1c1a','0c1a'],'Serbian Cyrillic');
a(['181a','081a'],'Serbian Latin');
a(['041b'],'Slovak');
a(['0424'],'Slovenian');
a(['2c0a','400a','340a','240a','140a','1c0a','300a','440a','100a','480a','080a','4c0a','180a','3c0a','280a','500a','0c0a','040a','540a','380a','200a'],'Spanish');
a(['081d','041d'],'Swedish');
a(['041e'],'Thai');
a(['041f'],'Turkish');
a(['0422'],'Ukranian');
a(['0843','0443'],'Uzbek');
a(['0803'],'Valencian');
a(['042a'],'Vietnamese');

function language() {
    var os = GetObject('winmgmts:').ExecQuery('select Locale from Win32_OperatingSystem');
    var locale = new Enumerator(os).item().Locale;

    // default to English if locale is not in abbrev{}
    return abbrev[locale.toLowerCase()] || 'English';
}

function fetch(url) {
    var xObj = new ActiveXObject("Microsoft.XMLHTTP");
    xObj.open("GET",url,true);
    xObj.setRequestHeader('User-Agent','XMLHTTP/1.0');
    xObj.send('');
    while (xObj.readyState != 4) WSH.Sleep(50);
    return(xObj);
}

function save(xObj, file) {
    var stream = new ActiveXObject("ADODB.Stream");
    with (stream) {
        type = 1; // binary
        open();
        write(xObj.responseBody);
        saveToFile(file, 2); // overwrite
        close();
    }
}

// fetch the initial web page
var x = fetch(WSH.Arguments(0));

// make HTML response all one line
var html = x.responseText.split(/\r?\n/).join('');

// create array of hrefs matching *.exe where the link text contains system language
var r = new RegExp('<a\\s*href="[^"]+\\.exe(?=[^\\/]+' + language() + ')', 'g');
var anchors = html.match(r)

// use only the first two
for (var i=0; i<2; i++) {

    // use only the stuff after the quotation mark to the end
    var dl = '' + /[^"]+$/.exec(anchors[i]);

    // if the location is a relative path, prepend the domain
    if (dl.substring(0,1) == '/') dl = /.+:\/\/[^\/]+/.exec(WSH.Arguments(0)) + dl;

    // target is path\filename
    var target=WSH.Arguments(1) + '\\' + /[^\/]+$/.exec(dl)

    // echo without a new line
    WSH.StdOut.Write('Saving ' + target + '... ');

    // fetch file and save it
    save(fetch(dl), target);

    WSH.Echo('Done.');
}

更新2

这是一个相同的脚本，只有一些小的调整，它还可以检测Windows的体系结构（32/64位），只下载一个安装程序而不是两个：

@if (@a==@b) @end /*

:: batch script portion

@echo off
setlocal

set "url=http://www.rarlab.com/download.htm"
set /p "savepath=Location to save? [%cd%] "

if "%savepath%"=="" set "savepath=%cd%"

cscript /nologo /e:jscript "%~f0" "%url%" "%savepath%"

goto :EOF

:: JScript portion */

// populate translation from locale identifier hex value to WinRAR language label
// http://msdn.microsoft.com/en-us/library/dd318693.aspx
var abbrev={}, a=function(arr,val){for(var i=0;i<arr.length;i++)abbrev[arr[i]]=val};
a(['1401','3c01','0c01','0801','2001','4001','2801','1c01','3801','2401'],'Arabic');
a(['042b'],'Armenian');
a(['082c','042c'],'Azerbaijani');
a(['0423'],'Belarusian');
a(['0402'],'Bulgarian');
a(['0403'],'Catalan');
a(['7c04'],'Chinese Traditional');
a(['0c04','1404','1004','0004'],'Chinese Simplified');
a(['101a'],'Croatian');
a(['0405'],'Czech');
a(['0406'],'Danish');
a(['0813','0413'],'Dutch');
a(['0425'],'Estonian');
a(['040b'],'Finnish');
a(['080c','0c0c','040c','140c','180c','100c'],'French');
a(['0437'],'Georgian');
a(['0c07','0407','1407','1007','0807'],'German');
a(['0408'],'Greek');
a(['040d'],'Hebrew');
a(['040e'],'Hungarian');
a(['0421'],'Indonesian');
a(['0410','0810'],'Italian');
a(['0411'],'Japanese');
a(['0412'],'Korean');
a(['0427'],'Lithuanian');
a(['042f'],'Macedonian');
a(['0414','0814'],'Norwegian');
a(['0429'],'Persian');
a(['0415'],'Polish');
a(['0816'],'Portuguese');
a(['0416'],'Portuguese Brazilian');
a(['0418'],'Romanian');
a(['0419'],'Russian');
a(['7c1a','1c1a','0c1a'],'Serbian Cyrillic');
a(['181a','081a'],'Serbian Latin');
a(['041b'],'Slovak');
a(['0424'],'Slovenian');
a(['2c0a','400a','340a','240a','140a','1c0a','300a','440a','100a','480a','080a','4c0a','180a','3c0a','280a','500a','0c0a','040a','540a','380a','200a'],'Spanish');
a(['081d','041d'],'Swedish');
a(['041e'],'Thai');
a(['041f'],'Turkish');
a(['0422'],'Ukranian');
a(['0843','0443'],'Uzbek');
a(['0803'],'Valencian');
a(['042a'],'Vietnamese');

function language() {
    var os = GetObject('winmgmts:').ExecQuery('select Locale from Win32_OperatingSystem');
    var locale = new Enumerator(os).item().Locale;

    // default to English if locale is not in abbrev{}
    return abbrev[locale.toLowerCase()] || 'English';
}

function fetch(url) {
    var xObj = new ActiveXObject("Microsoft.XMLHTTP");
    xObj.open("GET",url,true);
    xObj.setRequestHeader('User-Agent','XMLHTTP/1.0');
    xObj.send('');
    while (xObj.readyState != 4) WSH.Sleep(50);
    return(xObj);
}

function save(xObj, file) {
    var stream = new ActiveXObject("ADODB.Stream");
    with (stream) {
        type = 1; // binary
        open();
        write(xObj.responseBody);
        saveToFile(file, 2); // overwrite
        close();
    }
}

// fetch the initial web page
var x = fetch(WSH.Arguments(0));

// make HTML response all one line
var html = x.responseText.split(/\r?\n/).join('');

// get OS architecture (This method is much faster than the Win32_Processor.AddressWidth method)
var os = GetObject('winmgmts:').ExecQuery('select OSArchitecture from Win32_OperatingSystem');
var arch = /\d+/.exec(new Enumerator(os).item().OSArchitecture) * 1;

// get link matching *.exe where the link text contains system language and architecture
var r = new RegExp('<a\\s*href="[^"]+\\.exe(?=[^\\/]+' + language() + '[^<]+' + arch + '\\Wbit)');
var link = r.exec(html)

// use only the stuff after the quotation mark to the end
var dl = '' + /[^"]+$/.exec(link);

// if the location is a relative path, prepend the domain
if (dl.substring(0,1) == '/') dl = /.+:\/\/[^\/]+/.exec(WSH.Arguments(0)) + dl;

// target is path\filename
var target=WSH.Arguments(1) + '\\' + /[^\/]+$/.exec(dl)

// echo without a new line
WSH.StdOut.Write('Saving ' + target + '... ');

// fetch file and save it
save(fetch(dl), target);

WSH.Echo('Done.');