7

我正在使用 C# 开发一个网络爬虫,其中我有一个加载网页的网络浏览器控件,然后我收集数据并根据数据也作用于它。我的问题是我需要将一些数据写入具有侦听器的输入(文本框)。输入控件从不显示 html 中的值,它由控件上的事件触发的一些 Javascript 处理。OK我可以为控件值参数设置一个值,但是当我单击表单时它会丢失。

我已经在网上搜索了几天,现在我知道如何从 C# 调用 Javascript,但它是我无法理解的输入上的 Javascript 函数。

输入控件如下所示:

<div id="size_1" class="incrementControl" style="width: 68px;">
    <input id="size_1-input" class="hasListeners" maxlength="6" tabindex="103" value="" style="width:52px">
    <a id="size_1-up" class="up hasListeners" style="left:52px" href="javascript:void(0)"> </a>
    <a id="size_1-down" class="down hasListeners" style="left:52px" href="javascript:void(0)"> </a>
</div>

输入控件事件 keyup、keydown、change、keypress 和 blur 都调用这个 Javascript:

function (a) 
{
    return typeof f != "undefined" && (!a || f.event.triggered !== a.type) ? f.event.dispatch.apply(i.elem, arguments) : b;
}

我尝试过调用不同的事件,如 onClick 并以编程方式单击控件,但没有任何效果?!我不知道如何调用这个 Javascript 函数,因为它没有名称以及作为参数传递的内容?

非常感谢任何帮助或想法!

4

2 回答 2

1

您可以使用Watin。它支持Webbrowser 控制和Internet Explorer。

于 2012-10-22T03:08:46.557 回答
0

这里有一些扩展方法(建立在 WatiN 之上),允许您(除其他外)等待 JS 对象或事件:

using System;
using System.Threading;
using FluentSharp.Web35;
using FluentSharp.WinForms;
using FluentSharp.CoreLib;
using FluentSharp.CoreLib.API;

namespace FluentSharp.Watin
{
    public static class WatiN_IE_ExtensionMethods_Javascript
    {

        public static object invokeScript(this WatiN_IE ie, string functionName)
        {
            return ie.invokeScript(functionName, null);
        }

        public static object invokeScript(this WatiN_IE ie, string functionName, params object[] parameters)
        {
            //"[WatiN_IE] invokeScript '{0}' with parameters:{1}".info(functionName ,parameters.size());
            return ie.invokeScript(true, functionName, parameters);
        }   

        public static object invokeScript(this WatiN_IE ie, bool waitForExecutionComplete, string functionName, params object[] parameters)
        {
            var sync = new AutoResetEvent(false);
            object responseValue = null;
            ie.WebBrowser.invokeOnThread(
                ()=>{
                        var document = ie.WebBrowser.Document;
                        if (parameters.isNull())
                            responseValue = document.InvokeScript(functionName); 
                        else
                            responseValue = document.InvokeScript(functionName, parameters); 
                        sync.Set(); 
                });
            if (waitForExecutionComplete)
                sync.WaitOne();
            return responseValue;   
        }

        public static object invokeEval(this WatiN_IE ie, string evalScript)
        {
            var evalParam = "(function() { " + evalScript + "})();";
            //"[WatiN_IE] invokeEval evalParam: {0}".debug(evalParam);
            return ie.invokeScript("eval", evalParam);   
        }
        public static WatiN_IE.ToCSharp injectJavascriptFunctions(this WatiN_IE ie)
        {
            return ie.injectJavascriptFunctions(false);
        }

        public static WatiN_IE.ToCSharp injectJavascriptFunctions(this WatiN_IE ie, bool resetHooks)
        {
            if (ie.WebBrowser.isNull())
                "in InjectJavascriptFunctions, ie.WebBrowser was null".error();
            else
            {
                if (ie.WebBrowser.ObjectForScripting.isNull() || resetHooks)  
                {
                    ie.WebBrowser.ObjectForScripting = new WatiN_IE.ToCSharp();

                    "Injecting Javascript Hooks * Functions for page: {0}".debug(ie.url());
                    ie.eval("var o2Log = function(message) { window.external.write(message) };");
                    ie.invokeScript("o2Log","Test from Javascript (via toCSharp(message) )");
                    ie.eval("$o2 = window.external");
                    "Injection complete (use o2Log(...) or $o2.write(...)  to talk back to O2".info();
                    return (ie.WebBrowser.ObjectForScripting as WatiN_IE.ToCSharp);
                }
                else 
                {
                    if((ie.WebBrowser.ObjectForScripting is WatiN_IE.ToCSharp))
                        return (ie.WebBrowser.ObjectForScripting as WatiN_IE.ToCSharp);
                    else
                        "in WatiN_IE injectJavascriptFunctions, unexpected type in ie.WebBrowser.ObjectForScripting: {0}".error(ie.WebBrowser.ObjectForScripting.typeName());                   
                }

            }
            return null;
        }

        public static object downloadAndExecJavascriptFile(this WatiN_IE ie, string url)
        {
            "[WatiN_IE] downloadAndExecJavascriptFile: {0}".info(url);
            var javascriptCode = url.uri().getHtml();
            if (javascriptCode.valid())
                ie.eval(javascriptCode);
            return ie;
        }

        public static WatiN_IE injectJavascriptFunctions_onNavigate(this WatiN_IE ie)
        {

            ie.onNavigate((url)=> ie.injectJavascriptFunctions());
            return ie;
        }

        public static WatiN_IE setOnAjaxLog(this WatiN_IE ie, Action<string, string,string,string> onAjaxLog)
        {
            (ie.WebBrowser.ObjectForScripting as WatiN_IE.ToCSharp).OnAjaxLog = onAjaxLog;
            return ie;
        }

        public static WatiN_IE eval_ASync(this WatiN_IE ie, string script)
        {
            O2Thread.mtaThread(()=> ie.eval(script));
            return ie;
        }

        public static WatiN_IE eval(this WatiN_IE ie, string script)
        {
            return ie.eval(script, true);
        }

        public static WatiN_IE eval(this WatiN_IE ie, string script, bool waitForExecutionComplete)
        {
            var executionThread = O2Thread.staThread(()=> ie.IE.RunScript(script));         
            if (waitForExecutionComplete)
                executionThread.Join();
            return ie;  
        }

        public static WatiN_IE alert(this WatiN_IE ie, string alertScript)
        {
            return ie.eval("alert({0});".format(alertScript));
        }

        public static object getJsObject(this WatiN_IE ie)
        {
            var toCSharpProxy = ie.injectJavascriptFunctions();
            if (toCSharpProxy.notNull())
                return toCSharpProxy.getJsObject();
            return null;        
        }

        public static T getJsObject<T>(this WatiN_IE ie, string jsCommand)
        {
            var jsObject = ie.getJsObject(jsCommand);
            if (jsObject is T)
                return (T)jsObject;
            return default(T);
        }

        public static bool doesJsObjectExists(this WatiN_IE ie, string jsCommand)
        {
            var toCSharpProxy = ie.injectJavascriptFunctions();
            if (toCSharpProxy.notNull())
            {
                var command = "window.external.setJsObject(typeof({0}))".format(jsCommand);
                ie.invokeEval(command);
                ie.remapInternalJsObject();             
                return toCSharpProxy.getJsObject().str()!="undefined";
            }
            return false;
        }

        public static object getJsVariable(this WatiN_IE ie, string jsCommand)
        {
            return ie.getJsObject(jsCommand);
        }

        public static object getJsObject(this WatiN_IE ie, string jsCommand)
        {
            var toCSharpProxy = ie.injectJavascriptFunctions();
            if (toCSharpProxy.notNull())
            {
                var command = "window.external.setJsObject({0})".format(jsCommand);
                ie.invokeEval(command);
                ie.remapInternalJsObject();             
                return toCSharpProxy.getJsObject();
            }
            return null;
        }               

        public static WatiN_IE remapInternalJsObject(this WatiN_IE ie)
        {       
            //"setting JS _jsObject variable to getJsObject()".info();
            ie.invokeEval("_jsObject = window.external.getJsObject()"); // creates JS variable to be used from JS
            return ie;
        }

        public static WatiN_IE setJsObject(this WatiN_IE ie, object jsObject)
        {
            var toCSharpProxy = ie.injectJavascriptFunctions();
            if (toCSharpProxy.notNull())            
            {
                toCSharpProxy.setJsObject(jsObject);
                ie.remapInternalJsObject();
            }
            return ie;
        }

        public static object waitForJsObject(this WatiN_IE watinIe)
        {
            return watinIe.waitForJsObject(500, 20);
        }

        public static object waitForJsObject(this WatiN_IE watinIe, int sleepMiliseconds, int maxSleepTimes)
        {                   
            "[WatiN_IE][waitForJsObject] trying to find jsObject for {0} x {1} ms".info(maxSleepTimes, sleepMiliseconds);
            watinIe.setJsObject(null);
            for(var i = 0; i < maxSleepTimes ; i++)
            {
                var jsObject = watinIe.getJsObject();
                if(jsObject.notNull())
                {
                    "[watinIe][waitForJsObject] got value: {0} (n tries)".info(jsObject, i);
                    return jsObject;
                }

                watinIe.sleep(500, false);
            }
            "[WatiN_IE][waitForJsObject] didn't find jsObject after {0} sleeps of {1} ms".error(maxSleepTimes, sleepMiliseconds);
            return null;
        }

        public static object waitForJsVariable(this WatiN_IE watinIe, string jsCommand)
        {
            return watinIe.waitForJsVariable(jsCommand,  500, WatiN_IE_ExtensionMethods.WAITFORJSVARIABLE_MAXSLEEPTIMES);
        }

        public static object waitForJsVariable(this WatiN_IE watinIe, string jsCommand, int sleepMiliseconds, int maxSleepTimes)
        {   
            "[WatiN_IE][waitForJsVariable] trying to find jsObject called '{0}' for {1} x {2} ms".info(jsCommand, maxSleepTimes, sleepMiliseconds);         
            watinIe.setJsObject(null);
            for(var i = 0; i < maxSleepTimes ; i++)
            {
                if (watinIe.doesJsObjectExists(jsCommand))
                {
                    var jsObject = watinIe.getJsObject(jsCommand);
                    "[watinIe][waitForJsVariable] got value: {0} ({1} tries)".info(jsObject, i);
                    return jsObject;
                }                   
                watinIe.sleep(500, false);
            }
            "[WatiN_IE][waitForJsVariable] didn't find jsObject called '{0}' after {1} sleeps of {2} ms".error(jsCommand, maxSleepTimes, sleepMiliseconds);
            return null;
        }

        public static WatiN_IE deleteJsVariable(this WatiN_IE watinIe, string jsVariable)
        {
            var evalString = "try { delete " + jsVariable + " } catch(exception) { }";
            watinIe.eval(evalString);
            return watinIe;
        }


    }
}

有关更多 WatiN/WebBroser 扩展方法,请参阅https://github.com/o2platform/FluentSharp_Fork.WatiN/tree/master/FluentSharp.WatiN/ExtensionMethods

于 2014-12-14T09:34:43.730 回答