HTML标签解析

如何解析名称：&用DI HtmlParser从标签中的值文本？我尝试使用来自CLever Components的TCLHtmlParser,但它失败了.第二个问题是可以DIHtmlParser解析个别标签,例如循环通过其子标签.这是一个完全噩梦的这样一个简单的问题.

<div class="tvRow tvFirst hasLabel tvFirst" title="example1">
  <label class="tvLabel">Name:</label>
  <span class="tvValue">Value</span>
<div class="clear"></div></div>

<div class="tvRow tvFirst hasLabel tvFirst" title="example2">
  <label class="tvLabel">Name:</label>
  <span class="tvValue">Value</span>
<div class="clear"></div></div>

解决方法

您可以使用 IHTMLDocument2 DOM从HTML解析您需要的任何元素：

uses ActiveX,MSHTML;

const
  HTML =
  '<div class="tvRow tvFirst hasLabel tvFirst" title="example1">' +
  '<label class="tvLabel">Name:</label>' +
  '<span class="tvValue">Value</span>' +
  '<div class="clear"></div>' +
  '</div>';

procedure TForm1.Button1Click(Sender: TObject);
var
  doc: OleVariant;
  el: OleVariant;
  i: Integer;
begin
  doc := coHTMLDocument.Create as IHTMLDocument2;
  doc.write(HTML);
  doc.close;
  ShowMessage(doc.body.innerHTML);
  for i := 0 to doc.body.all.length - 1 do
  begin
    el := doc.body.all.item(i);
    if (el.tagName = 'LABEL') and (el.className = 'tvLabel') then
      ShowMessage(el.innerText);
    if (el.tagName = 'SPAN') and (el.className = 'tvValue') then
      ShowMessage(el.innerText);
  end;
end;

我想提到我今天发现的另一个非常好的HTML解析器：htmlp(Delphi Dom HTML解析器和转换器).它不像IHTMLDocument2那么灵活,但是很容易使用,快速,免费,并支持旧的Delphi版本的Unicode.

样品用量：

uses HtmlParser,DomCore;

function GetDocBody(HtmlDoc: TDocument): TElement;
var
  i: integer;
  node: TNode;
begin
  Result := nil;
  for i := 0 to HtmlDoc.documentElement.childNodes.length - 1 do
  begin
    node := HtmlDoc.documentElement.childNodes.item(i);
    if node.nodeName = 'body' then
    begin
      Result := node as TElement;
      Break;
    end;
  end;
end;

procedure THTMLForm.Button2Click(Sender: TObject);
var
  HtmlParser: THtmlParser;
  HtmlDoc: TDocument;
  i: Integer;
  body,el: TElement;
  node: TNode;
begin
  HtmlParser := THtmlParser.Create;
  try
    HtmlDoc := HtmlParser.parseString(HTML);
    try
      body := GetDocBody(HtmlDoc);
      if Assigned(body) then
        for i := 0 to body.childNodes.length - 1 do
        begin
          node := body.childNodes.item(i);
          if (node is TElement) then
          begin
            el := node as TElement;
            if (el.tagName = 'div') and (el.GetAttribute('class') = 'tvRow tvFirst hasLabel tvFirst') then
            begin
              // iterate el.childNodes here...
              ShowMessage(IntToStr(el.childNodes.length));
            end;
          end;
        end;
    finally
      HtmlDoc.Free;
    end;
  finally
    HtmlParser.Free
  end;
end;