/// <summary>
/// 检查元素是否为自结束标签
/// </summary>
/// <param name="tag">元素开始标签</param>
/// <returns>是否为自结束标签</returns>
protected virtual bool IsSelfCloseElement( HtmlBeginTag tag )
{
return HtmlSpecification.IsForbiddenEndTag( tag.TagName );
}
/// <summary>
/// 处理元素开始标签
/// </summary>
/// <param name="beginTag">开始标签信息</param>
/// <returns>处理过程中所创建的元素对象，若不支持则返回 null</returns>
protected virtual IHtmlElement ProcessBeginTag( HtmlBeginTag beginTag )
{
if ( HtmlSpecification == null )//当解析第一个开始标签时还没有找到 DTD 声明，则强行使用默认的 HTML 规范。
SetHtmlSpecification( (HtmlDoctypeDeclaration) null );
string tagName = beginTag.TagName;
bool selfClosed = beginTag.SelfClosed;
//检查是否为自结束标签，并作相应处理
if ( IsSelfCloseElement( beginTag ) )
selfClosed = true;
//检查是否为CData标签，并作相应处理
if ( IsCDataElement( beginTag ) )
Reader.EnterCDataMode( tagName.ToLowerInvariant() );
//检查父级是否可选结束标记，并作相应处理
{
var element = CurrentContainer as IHtmlElement;
if ( element != null && HtmlSpecification.IsOptionalEndTag( element.Name ) )
{
if ( ImmediatelyClose( tagName, element ) )
ContainerStack.Pop();
}
}
//处理所有属性
var attributes = new Dictionary<string, string>( StringComparer.OrdinalIgnoreCase );
foreach ( var a in beginTag.Attributes )
{
string name = a.Name;
string value = a.Value;
if ( value != null )
value = HtmlEncoding.HtmlDecode( value );
if ( attributes.ContainsKey( name ) )//重复的属性名，只取第一个
continue;
attributes.Add( name, value );
}
//创建元素
{
var element = CreateElement( tagName, attributes );
//加入容器堆栈
if ( !selfClosed )
ContainerStack.Push( element );
return element;
}
}
/// <summary>
/// 检查元素是否为CDATA标签
/// </summary>
/// <param name="tag">元素开始标签</param>
/// <returns>是否为CDATA标签</returns>
protected virtual bool IsCDataElement( HtmlBeginTag tag )
{
return HtmlSpecification.IsCDataElement( tag.TagName );
}
