IHTMLDocument2可以打开一个本地HTML文件, 并取得所有图片SRC属性吗?

[yyyyy_3]

IHTMLDocument2可以打开一个本地HTML文件, 并取得所有图片SRC属性吗?

BSTR url = SysAllocString(L"c:\\zzz.html");
VARIANT name;
BSTR bstr_name = SysAllocString(OLESTR("_blank"));
name.vt = VT_BSTR;
name.bstrVal = bstr_name;
VARIANT feature;
BSTR str_feature = SysAllocString(OLESTR("channelmode = {yes|1}, directories = {yes|1},height = 150, left = 800, location = {no}, menubar = {no|0}, resizable = {no|0}, scrollbars = {no}, status = {no}, titlebar = {no}, toolbar = {no|0}, top = 500, width = 150"));
feature.vt = VT_BSTR;
feature.bstrVal = str_feature;
VARIANT replaces;
BSTR str_rep = SysAllocString(OLESTR("true"));
replaces.vt = VT_BSTR;
replaces.bstrVal = str_rep;
IDispatch *newDispatch;

hr = pDoc->open(url, name, feature, replaces, &newDispatch);
if (hr != S_OK)
AfxMessageBox("error hr");

执行显示: error hr.

能打开本地html文件吗?

如果此技术不行, 有本地html的解析类吗?

[蒋晟]

CComPtr<IPersistFile> pFile;
pFile.CoCreateInstance(CLSID_HTMLDocument)
if(pFile)
{
    pFile->Load(L"c:\\test1.htm",STGM_READ);
    CComQIPtr<IHtmlDocument2> pDoc(pFile);
    ......
}

[breaksoftware]

这个和本地与否没有关系，主要是你获取接口指针可能不对，你参考下蒋晟给你的方案。

[Dobzhansky]

  // 取得对所有元素集合的引用
  if (SUCCEEDED(hr = m_pMSHTML->get_all( &pColl ))) {
    long cElems;

    assert(pColl);

    // 个数
    if (SUCCEEDED(hr = pColl->get_length( &cElems ))) {
      std::cout << "element counts: " << cElems << std::endl;
      // 访问每个元素的属性, 比如 TAGNAME and HREF
      for ( int i = 0; i < cElems; i++ ) {
        VARIANT vIndex;
        vIndex.vt = VT_UINT;
        vIndex.lVal = i;
        VARIANT var2 = { 0 };
        LPDISPATCH pDisp;

        if (SUCCEEDED(hr = pColl->item( vIndex, var2, &pDisp ))) {
          IHTMLElement* pElem = NULL;
          if (SUCCEEDED(hr = pDisp->QueryInterface( IID_IHTMLElement, (LPVOID*)&pElem ))) {
            _bstr_t bstrTag;
            BSTR bstr;

            assert(pElem);

            hr = pElem->get_tagName(&bstr);
            if (bstr) {
              bstrTag = bstr;
              SysFreeString(bstr);
            }

            // 如果是 <IMG> 元素, 获取其 SRC 属性
            IHTMLImgElement* pImage = NULL;
            if (SUCCEEDED(hr = pDisp->QueryInterface( IID_IHTMLImgElement, (LPVOID*)&pImage ))) {
              assert(pImage);

              pImage->get_src(&bstr);
              if (bstr) {
                bstrTag += " - ";
                bstrTag += bstr;
                SysFreeString(bstr);
              }
              pImage->Release();
            }

            PrintBSTR(bstrTag);

            pElem->Release();
          } // QI(IHTMLElement)
          pDisp->Release();
        } // item
      } // for
    } // get_length

    pColl->Release();
  } // get_all

[Dobzhansky]

mshtml 的几种 load

HRESULT
CHtmlDOM::loadFromMoniker(LPCTSTR szUrl)
{
  if (!m_pDisp)
    return S_FALSE;

  HRESULT hr;

  CSmartPtr<IMoniker> pMk;
  if (SUCCEEDED(hr = CreateURLMoniker(NULL, szUrl, &pMk))) {
    CSmartPtr<IBindCtx> pBCtx;
    if (SUCCEEDED(hr = CreateBindCtx(0, &pBCtx))) {
      // Use MSHTML moniker services to load the specified document
      CSmartPtr<IPersistMoniker> pPMk;
      if (SUCCEEDED(hr = m_pDisp->QueryInterface(IID_IPersistMoniker, (LPVOID*)&pPMk))) {
        // Call Load on the IPersistMoniker
        // 这里调用后立马返回, 文档此时仍未加载完毕
        // 需要等到 MSHTML 报告 READYSTATE_COMPLETE 时刻.
        // 查看 Site 实现上的 IPropertyNotifySink::OnChanged 函数
        hr = pPMk->Load(FALSE, pMk, pBCtx, STGM_READ);
      }
    }
  }
  return hr;
}

// 传统的持久化方式.
// MSHTML 也是使用异步加载的方式.

HRESULT
CHtmlDOM::loadFromFile(LPCTSTR szUrl)
{
  if (!m_pDisp)
    return S_FALSE;

  HRESULT hr;
  CSmartPtr<IPersistFile> pPF;
  // MSHTML 支持常规文件持久化.
  if ( SUCCEEDED(hr = m_pDisp->QueryInterface(IID_IPersistFile, (LPVOID*) &pPF))) {
    hr = pPF->Load(szUrl, 0);
  }

  return hr;
}

// 流接口的加载
//

HRESULT
CHtmlDOM::loadFromStream(IStream* pStream)
{
  if (!m_pDisp)
    return S_FALSE;

  HRESULT hr = S_OK;
  CSmartPtr<IPersistStreamInit> pPSI;
  //The IPersistStreamInit interface has InitNew and Load methods
  //that are used to initialize and load an HTML document from a stream.
  //The InitNew method initializes the stream to a known state
  //and the Load method loads the HTML content from the stream.
  if ( SUCCEEDED( hr = m_pDisp->QueryInterface(IID_IPersistStreamInit, (LPVOID*)&pPSI) )) {
    hr = pPSI->InitNew();
    hr = pPSI->Load(pStream);
  }
  return hr;
}

[yyyyy_3]

void CgfgfDlg::OnBnClickedButton2()
{
// TODO: 在此添加控件通知处理程序代码
HRESULT hr;
hr = CoInitialize(NULL);

CComPtr<IPersistFile> pFile;
pFile.CoCreateInstance(CLSID_HTMLDocument);
if(pFile)
{
pFile->Load(L"c:\\zzz2.html",STGM_READ);
CComQIPtr<IHTMLDocument2> m_pMSHTML(pFile);

IHTMLElementCollection* pColl = NULL;
if (SUCCEEDED(hr = m_pMSHTML->get_all( &pColl )))
{
long cElems;

ASSERT(pColl);

// retrieve the count of elements in the collection
if (SUCCEEDED(hr = pColl->get_length( &cElems )))
{
}
} // get_all
}
}


我把程序放在了一个OnButton中
但if (SUCCEEDED(hr = pColl->get_length( &cElems )))
虽然成功了, 但cElems 总为0, 知道为什么?

用的vs2008.