/// <summary>
/// 段落分析
/// </summary>
/// <param name="paragraph"></param>
/// <param name="root"></param>
/// <param name="subTitle"></param>
void AnlayzeParagraph(HtmlNode paragraph, MyHtmlNode root, String subTitle = "")
{
//原始HTML的第二阶层无法保证嵌套结构是正确的，
//所以决定第二阶层不分层
if (paragraph.ChildNodes.Count == 1)
{
if (paragraph.ChildNodes[0].Name == "#text")
{
if (subTitle != String.Empty)
{
var txtnode = new MyHtmlNode();
txtnode.Content = subTitle;
root.Children.Add(txtnode);
return;
}
}
}
foreach (var node in paragraph.ChildNodes)
{
if (node.Name == "div")
{
if (node.Attributes.Contains("type"))
{
if (node.Attributes["type"].Value == "content")
{
foreach (var child in node.ChildNodes)
{
if (child.Name == "table")
{
var tablenode = new MyHtmlNode();
tablenode.Content = String.Empty;
TableId++;
tablenode.TableId = TableId;
var tablecontentlist = HTMLTable.GetTable(child, TableId);
TableList.Add(TableId, tablecontentlist);
root.Children.Add(tablenode);
continue;
}
if (child.Name == "hidden")
{
continue;
}
var content = Normalizer.Normalize(child.InnerText);
if (!String.IsNullOrEmpty(content))
{
var contentnode = new MyHtmlNode();
contentnode.Content = subTitle + content;
root.Children.Add(contentnode);
}
else
{
if (subTitle != String.Empty)
{
var contentnode = new MyHtmlNode();
contentnode.Content = subTitle;
root.Children.Add(contentnode);
}
}
subTitle = String.Empty;
}
}
if (node.Attributes["type"].Value == "paragraph")
{
var title = String.Empty;
if (node.Attributes.Contains("title"))
{
title = node.Attributes["title"].Value;
title = Normalizer.Normalize(title);
}
AnlayzeParagraph(node, root, title);
}
}
}
if (node.Name == "table")
{
var tablenode = new MyHtmlNode();
tablenode.Content = String.Empty;
TableId++;
tablenode.TableId = TableId;
var tablecontentlist = HTMLTable.GetTable(node, TableId);
TableList.Add(TableId, tablecontentlist);
root.Children.Add(tablenode);
}
}
}
