Delphi 怎样使用 mshtml 解析 HTML代码(不要依赖 Webbrowse 控件的)

Question

Delphi 怎样使用 mshtml 解析 HTML代码(不要依赖 Webbrowse 控件的)

Answer 1

mshtml是一组由IE提供的DOM接口，无法离开IE/WebBrowser控件使用。
你可以使用一个隐藏的WebBrowser来获得IE的HTML解析功能，或者考虑其他的HTML解析器实现，可以在Google上搜索"HTML解析器"找到一些有用信息。

Answer 2

其接口仍就是IE核心,不过可以用没有UI的TInternetExplorer的

Answer 3

var
   doc:IHTMLDocument2;
   hh:String;begin
   hh := '<font color="red">字体</font>';
   doc := CoHTMLDocument.Create as IHTMLDocument2;
   
   //我创建了 doc 实例,想用 doc 来解析 hh,取出 hh 的innerHTML 和 属性 color
   //怎么做?急用谢谢了

Answer 4

使用mshtml解析HTML源文件，不下载 
http://program.sdoq.com/show/68305

Answer 5

贴一段代码或许对你有用 
void       OnLogin()       
{   
_variant_t       v;   
HRESULT       hr;   
VARIANT       id,       index;   
CComPtr   <IDispatch>       spDispatch;   
CComQIPtr   <IHTMLDocument2,       &IID_IHTMLDocument2>       pDoc2;   
CComQIPtr   <IHTMLElement,       &IID_IHTMLElement>       pElement;   
CComQIPtr   <IHTMLElementCollection,&IID_IHTMLElementCollection>       pElementCol;
CComPtr   <IHTMLInputTextElement>       pInputTextElement;
if       (m_spSHWinds       ==       NULL)       {   
hr       =       m_spSHWinds.CreateInstance(__uuidof(SHDocVw::ShellWindows));   
   if       (FAILED(hr)){   
           MessageBox(   "Failed   ");   
           CoUninitialize();   
    }   
}   
if       (m_spSHWinds)       {   
       long       n=0;   
       m_spSHWinds->   get_Count(&n);       //得到浏览器的个数   
           //遍历每一个浏览器   
         for       (long       i       =       0;       i       <       n;       i++)       {   
         CComPtr   <IDispatch>       spDisp;   
         v       =       (long)i;  
         spDisp       =m_spSHWinds->   Item(&v);   
         SHDocVw::IWebBrowser2Ptr       spBrowser(spDisp);               //生成一个IE窗口的智能指针   
          if       (spBrowser)           { 
    //获取IHTMLDocument2接口   
             if       (SUCCEEDED(spBrowser->   get_Document(       &spDispatch)))   
                 pDoc2       =       spDispatch;   
            if(pDoc2!=NULL)       {   
                 //获取所有元素的集合   
                   if(SUCCEEDED(pDoc2->   get_all(&pElementCol))){   
                 long       p=0;   
                 if(SUCCEEDED(pElementCol->   get_length(&p)))                   {       
                      if(p!=0){
                            //遍历所有元素           
                                for(long       i=0;i   <p;i++){
                                V_VT(&id)       =       VT_I4;   
                                V_I4(&id)       =       i;   
                                V_VT(&index)       =       VT_I4;   
                                V_I4(&index)       =       0;   
                      if(SUCCEEDED(pElementCol->   item(id,index,       &spDispatch)))       {
//应该就这附近的几行错误吧   
if(SUCCEEDED(spDispatch->   QueryInterface(IID_IHTMLElement,(void**)&pElement)))       {           //应该就这附近的几行错误吧   
    
        if(SUCCEEDED(pElement->   QueryInterface(IID_IHTMLInputTextElement,(void**)&pInputTextElement)))       {               
CComBSTR       strName(   "csdn   ");           //用户名   
CComBSTR       strPwd(   "123456   ");               //密码   
CComBSTR       type;                                                       //输入框类型   
pInputTextElement->   get_type(&type);//获取输入框类型   
CString       strtype(type);   
strtype.MakeUpper();   
                                                                                        if       (lstrcmpi(strtype,       _T(   "TEXT   "))       ==       0)       {   
                                                                                                                                                                                                        pInputTextElement->   put_value(strName);//设置文本框的值(用户名)       
                            }   
                                                            if       (lstrcmpi(strtype,       _T(   "PASSWORD   "))       ==       0)       {   
                                                                                                                                                                                                        pInputTextElement->   put_value(strPwd);//设置文本框的值(密码)   
                                                                    }               }                                                                       }   
                    }   
}   
            }   
}   
                }   
            }   
            
                    }   
            }   
            }   
}       the       following       code       will       click       submit       button       of       html       page       in       IE,       modify       it       to       click       other       button.       
    ///////////////////////////////////////////////////////////////////////////////////////       
    //click       submit       button       of       IE       window       
    //If       it       works,       it       is       written       by       masterz,otherwise       I       don't       
    //know       who       writes       it^_^       
    ///////////////////////////////////////////////////////////////////////////////////////       
    void       CGetIESrcDlg::NavigateToUrl()       
    {       
    //       Import       the       following       files       in       your       stdafx.h       
    //   #import       <mshtml.tlb>       //       Internet       Explorer       5       
    //   #import       <shdocvw.dll>       
    //           Refer       to       "Connect       to       Internet       Explorer       Instances,       From       your       own       Process.       "       in       www.codeguru.com       
    SHDocVw::IShellWindowsPtr       m_spSHWinds;       
    CoInitialize(NULL);       
    if(m_spSHWinds.CreateInstance(__uuidof(SHDocVw::ShellWindows))       ==       S_OK)       
    {       
    IDispatchPtr       spDisp;       
    long       nCount       =       m_spSHWinds-> GetCount();       
    for       (long       i       =       0;       i       <       nCount;       i++)       
    {       
    _variant_t       va(i,       VT_I4);       
    spDisp       =       m_spSHWinds-> Item(va);       
    SHDocVw::IWebBrowser2Ptr       spBrowser(spDisp);       
    if       (spBrowser       !=       NULL)       
    {       
    IDispatchPtr       spDisp;       
    if(spBrowser-> get_Document(&spDisp)       ==       S_OK       &&       spDisp!=       0       )       
    {       
    MSHTML::IHTMLDocument2Ptr       spHtmlDocument(spDisp);       
    MSHTML::IHTMLElementPtr       spHtmlElement;       
    if(spHtmlDocument==NULL)       
    continue;       
    spHtmlDocument-> get_body(&spHtmlElement);       
    if(spHtmlDocument==NULL)       
    continue;       
    HRESULT       hr;       
    MSHTML::IHTMLElementCollection*       pColl=NULL;       
    hr=spHtmlDocument-> get_all(&pColl);       
    if(pColl!=NULL&&SUCCEEDED(hr))       
    {       
    MSHTML::IHTMLElement*       pElem=NULL;       
    _variant_t       index;       
    index.vt=VT_I4;       
    index.intVal=0;       
    _variant_t       name("Submit");       
    IDispatchPtr       disp;       
    disp=pColl-> item(name,index);       
    if(disp==NULL)       
    hr=E_FAIL;       
    else       
    {       
    hr=disp-> QueryInterface(&pElem);       
    }       
    if       (SUCCEEDED(hr)&&       pElem       !=       NULL)       
    {       
    //       
    BSTR       bstrhtml;       
    pElem-> get_outerHTML(&bstrhtml);       
    CString       str(bstrhtml);       
    AfxMessageBox(str);       
    pElem-> click();       
    pElem-> Release();       
    }       
    pColl-> Release();       
    }       
    }       
        
    }       
    }       
        
    }       
    else       {       
    AfxMessageBox("Shell       Windows       interface       is       not       avilable");       
    }       
    CoUninitialize();       
    }