/// <summary>
/// 从批量集合中获取数据
/// </summary>
/// <param name="doc"></param>
/// <param name="crawItem"></param>
/// <param name="shortv"></param>
/// <param name="document"></param>
public static void GetDataFromXPath(this HtmlDocument doc, CrawlItem crawItem, IFreeDocument document)
{
var result = doc.DocumentNode.GetDataFromXPath(crawItem.XPath, crawItem.IsHTML);
if (result != null)
{
document.SetValue(crawItem.Name, result);
}
}
/// <summary>
/// 从批量集合中获取数据
/// </summary>
/// <param name="doc"></param>
/// <param name="crawItem"></param>
/// <param name="shortv"></param>
/// <param name="document"></param>
public static void GetDataFromXPath(this HtmlDocument doc, CrawlItem crawItem, string shortv,
IFreeDocument document)
{
var result = doc.DocumentNode.GetDataFromXPath(new XPath(crawItem.XPath).TakeOff(shortv).ToString(),
crawItem.IsHTML);
if (result != null)
{
document.SetValue(crawItem.Name, result);
}
}
/// <summary>
/// 查询XPATH，返回CrawlItem
/// </summary>
/// <param name="doc"></param>
/// <param name="keyword"></param>
/// <param name="name"></param>
/// <returns></returns>
public static CrawlItem SearchXPath(this HtmlDocument doc, string keyword, string name, bool hasAttr = true)
{
var xpath = doc.SearchXPath(keyword, () => hasAttr).FirstOrDefault();
if (xpath == null)
{
return(null);
}
var crawitem = new CrawlItem {
XPath = xpath, SampleData1 = keyword, Name = name
}
;
return(crawitem);
}
private static List <CrawlItem> GetDiffNodes(HtmlDocument doc2, string root, bool isAttrEnabled,
IEnumerable <CrawlItem> exists = null, int minNodeCount = 2)
{
HtmlNodeCollection nodes = null;
var crawlItems = new List <CrawlItem>();
try
{
nodes = doc2.DocumentNode.SelectNodes(root);
}
catch (Exception ex)
{
XLogSys.Print.Error(ex.Message + " 可能XPath表达式有误");
return(new List <CrawlItem>());
}
if (nodes == null || nodes.Count < minNodeCount)
{
return(new List <CrawlItem>());
}
var buffers = new List <List <string> >();
var nodes3 = nodes.ToList(); // .Where(d => d.Name.Contains("#") == false).ToList();
if (nodes3.Count > 1)
{
GetDiffNodes(nodes3, crawlItems, buffers, isAttrEnabled);
}
if (exists != null)
{
var copied = exists.Select(d =>
{
var xp = new CrawlItem();
d.DictCopyTo(xp);
return(xp);
}).ToList();
crawlItems.RemoveElementsNoReturn(d => copied.Any(r => IsSameXPath(d.XPath, r.XPath, root)));
crawlItems.AddRange(copied);
}
return(crawlItems);
}
/// <summary>
/// find different text and return it name and xpath
/// </summary>
/// <param name="isAttrEnabled">是否抓取标签中的数据</param>
/// <returns></returns>
private static bool GetDiffNodes(List <HtmlNode> nodes, List <CrawlItem> result, List <List <string> > buffers,
bool isAttrEnabled)
{
var isChildContainInfo = false;
var node1 = nodes.First();
var node2 = nodes[1];
if (node1.ChildNodes.Count == 1 && node1.ChildNodes[0].NodeType == HtmlNodeType.Text)
{
var row = nodes.Select(d => d.SelectSingleNode(d.XPath))
.Where(d => d != null).Select(d => d.InnerText.Trim()).ToList();
if (row.Any(d => CompareString(d, node1.InnerText) == false))
{
var name = node1.SearchPropertyName(result) ?? "属性" + result.Count;
if (buffers.Any(d => ListEqual(d, row)))
{
return(true);
}
var crawlItem = new CrawlItem
{
Name = name,
SampleData1 = node1.InnerText,
XPath = result.Count % 2 == 0 ? node1.XPath : node2.XPath
};
result.Add(crawlItem);
buffers.Add(row);
return(true);
}
return(false);
}
foreach (var nodechild1 in node1.ChildNodes)
{
if (nodechild1.XPath.Contains("#"))
{
continue;
}
var path = new XPath(nodechild1.XPath).TakeOff(node1.XPath).ToString();
var nodechild2 =
nodes.Select(d => d.SelectSingleNode(d.XPath + path)).Where(d => d != null).ToList();
if (nodechild2.Count == 1)
{
continue;
}
isChildContainInfo |= GetDiffNodes(nodechild2, result, buffers, isAttrEnabled);
}
if (isAttrEnabled == false)
{
return(isChildContainInfo);
}
foreach (var attribute in node1.Attributes)
{
var attr1 = attribute.Value;
;
var row = nodes.Select(d => d.SelectSingleNode(d.XPath))
.Where(d => d != null).Where(d => d.Attributes.Contains(attribute.Name)).Select(d => d.Attributes[attribute.Name].Value).ToList();
if (row.Any(d => CompareString(d, attr1) == false))
{
if (buffers.Any(d => ListEqual(d, row)))
{
return(isChildContainInfo);
}
var name = node1.SearchPropertyName(result);
if (name != null)
{
name += '_' + attribute.Name;
}
else
{
name = "属性" + result.Count;
}
var craw = new CrawlItem {
Name = name, SampleData1 = attr1, XPath = attribute.XPath
};
result.Add(craw);
}
}
return(isChildContainInfo);
}
/// <summary>
/// find different text and return it name and xpath
/// </summary>
/// <param name="isAttrEnabled">是否抓取标签中的数据</param>
/// <returns></returns>
private static bool GetDiffNodes(List <HtmlNode> nodes, List <CrawlItem> result, List <List <string> > buffers,
bool isAttrEnabled)
{
var isChildContainInfo = false;
var len = nodes.Count;
var node1 = nodes[Random.Next(0, len / 2)];
var node2 = nodes[Random.Next(len / 2, len)];
if (node1.ChildNodes.Count == 1 && node1.ChildNodes[0].NodeType == HtmlNodeType.Text)
{
var row = nodes.Select(d => d.SelectSingleNode(d.XPath))
.Where(d => d != null).Select(d => d.InnerText.Trim()).ToList();
if (row.Any(d => CompareString(d, node1.InnerText) == false))
{
var name = node1.SearchPropertyName(result) ?? "属性" + result.Count;
if (buffers.Any(d => ListEqual(d, row)))
{
return(true);
}
var crawlItem = new CrawlItem
{
Name = name,
SampleData1 = node1.InnerText,
XPath = result.Count % 2 == 0 ? node1.XPath : node2.XPath
};
result.Add(crawlItem);
buffers.Add(row);
return(true);
}
return(false);
}
foreach (var nodechild1 in node1.ChildNodes)
{
if (nodechild1.XPath.Contains("#"))
{
continue;
}
var path = XPath.TakeOff(nodechild1.XPath, node1.XPath);
var fail = false;
var nodechild2 =
nodes.Select(d =>
{
if (fail)
{
return(null);
}
try
{
var node = d.SelectSingleNode(d.XPath + path);
return(node);
}
catch (Exception)
{
fail = true;
return(null);
}
}).Where(d => d != null && fail == false).ToList();
if (nodechild2.Count < 2 || fail)
{
continue;
}
isChildContainInfo |= GetDiffNodes(nodechild2, result, buffers, isAttrEnabled);
}
if (isAttrEnabled == false)
{
return(isChildContainInfo);
}
foreach (var attribute in node1.Attributes)
{
var attr1 = attribute.Value;
;
var row = nodes.Select(d =>
{
try
{
return(d.SelectSingleNode(d.XPath));
}
catch (Exception ex)
{
XLogSys.Print.Error("XPath表达式编写错误： " + d.XPath);
return(null);
}
})
.Where(d => d != null)
.Where(d => d.Attributes.Contains(attribute.Name))
.Select(d => d.Attributes[attribute.Name].Value)
.ToList();
if (row.Any(d => CompareString(d, attr1) == false))
{
if (buffers.Any(d => ListEqual(d, row)))
{
return(isChildContainInfo);
}
var name = node1.SearchPropertyName(result);
if (name != null)
{
name += '_' + attribute.Name;
}
else
{
name = "属性" + result.Count;
}
var craw = new CrawlItem {
Name = name, SampleData1 = attr1, XPath = attribute.XPath
};
result.Add(craw);
}
}
return(isChildContainInfo);
}
private void AddNewItem(bool isAlert = true)
{
var path = SelectXPath;
if (!string.IsNullOrEmpty(RootXPath))
{
var root = HtmlDoc.DocumentNode.SelectSingleNode(RootXPath).ParentNode;
var node = HtmlDoc.DocumentNode.SelectSingleNode(path);
if (!node.IsAncestor(root))
{
if (isAlert)
MessageBox.Show("当前XPath所在节点不是父节点的后代，请检查对应的XPath");
return;
}
path = new XPath(node.XPath).TakeOff(root.XPath).ToString();
}
var item = new CrawlItem {XPath = path, Name = SelectName, SampleData1 = SelectText};
if (CrawlItems.Any(d => d.Name == SelectName))
{
SelectName = "属性" + CrawlItems.Count;
if (isAlert)
{
MessageBox.Show($"已存在名称为{SelectName}的属性，不能重复添加");
return;
}
}
CrawlItems.Add(item);
SelectXPath = "";
}
public override void DictDeserialize(IDictionary<string, object> dicts, Scenario scenario = Scenario.Database)
{
base.DictDeserialize(dicts, scenario);
URL = dicts.Set("URL", URL);
RootXPath = dicts.Set("RootXPath", RootXPath);
IsMultiData = dicts.Set("IsMultiData", IsMultiData);
URLFilter = dicts.Set("URLFilter", URLFilter);
Crawler = dicts.Set("Crawler", Crawler);
ContentFilter = dicts.Set("ContentFilter", ContentFilter);
if (dicts.ContainsKey("HttpSet"))
{
var doc2 = dicts["HttpSet"];
var p = doc2 as IDictionary<string, object>;
Http.UnsafeDictDeserialize(p);
}
if (dicts.ContainsKey("Login"))
{
var doc2 = dicts["Login"];
var p = doc2 as IDictionary<string, object>;
var item = new HttpItem();
item.DictDeserialize(p);
Documents.Add(item);
}
if (dicts.ContainsKey("Generator"))
{
var doc2 = dicts["Generator"];
var p = doc2 as IDictionary<string, object>;
}
var doc = dicts as FreeDocument;
if (doc?.Children != null)
{
foreach (var child in doc.Children)
{
var item = new CrawlItem();
item.DictDeserialize(child);
CrawlItems.Add(item);
}
}
}
/// <summary>
/// find different text and return it name and xpath
/// </summary>
/// <param name="isAttrEnabled">是否抓取标签中的数据</param>
/// <returns></returns>
private static bool GetDiffNodes(List<HtmlNode> nodes, List<CrawlItem> result, List<List<string>> buffers,
bool isAttrEnabled)
{
var isChildContainInfo = false;
var node1 = nodes.First();
var node2 = nodes[1];
if (node1.ChildNodes.Count == 1 && node1.ChildNodes[0].NodeType == HtmlNodeType.Text)
{
var row = nodes.Select(d => d.SelectSingleNode(d.XPath))
.Where(d => d != null).Select(d => d.InnerText.Trim()).ToList();
if (row.Any(d => CompareString(d, node1.InnerText) == false))
{
var name = node1.SearchPropertyName(result) ?? "属性" + result.Count;
if (buffers.Any(d => ListEqual(d, row)))
return true;
var crawlItem = new CrawlItem
{
Name = name,
SampleData1 = node1.InnerText,
XPath = result.Count%2 == 0 ? node1.XPath : node2.XPath
};
result.Add(crawlItem);
buffers.Add(row);
return true;
}
return false;
}
foreach (var nodechild1 in node1.ChildNodes)
{
if (nodechild1.XPath.Contains("#"))
continue;
var path = new XPath(nodechild1.XPath).TakeOff(node1.XPath).ToString();
var nodechild2 =
nodes.Select(d => d.SelectSingleNode(d.XPath + path)).Where(d => d != null).ToList();
if (nodechild2.Count == 1)
continue;
isChildContainInfo |= GetDiffNodes(nodechild2, result, buffers, isAttrEnabled);
}
if (isAttrEnabled == false)
{
return isChildContainInfo;
}
foreach (var attribute in node1.Attributes)
{
var attr1 = attribute.Value;
;
var row = nodes.Select(d => d.SelectSingleNode(d.XPath))
.Where(d => d != null).Where(d => d.Attributes.Contains(attribute.Name)).Select(d=>d.Attributes[attribute.Name].Value).ToList();
if (row.Any(d => CompareString(d, attr1) == false))
{
if (buffers.Any(d => ListEqual(d, row)))
return isChildContainInfo;
var name = node1.SearchPropertyName(result);
if (name != null)
name += '_' + attribute.Name;
else
{
name = "属性" + result.Count;
}
var craw = new CrawlItem {Name = name, SampleData1 = attr1, XPath = attribute.XPath};
result.Add(craw);
}
}
return isChildContainInfo;
}
/// <summary>
/// 查询XPATH，返回CrawlItem
/// </summary>
/// <param name="doc"></param>
/// <param name="keyword"></param>
/// <param name="name"></param>
/// <returns></returns>
public static CrawlItem SearchXPath(this HtmlDocument doc, string keyword, string name, bool hasAttr = true)
{
var xpath = doc.SearchXPath(keyword, () => hasAttr).FirstOrDefault();
if (xpath == null)
return null;
var crawitem = new CrawlItem {XPath = xpath, SampleData1 = keyword, Name = name}
;
return crawitem;
}
/// <summary>
/// 从批量集合中获取数据
/// </summary>
/// <param name="doc"></param>
/// <param name="crawItem"></param>
/// <param name="shortv"></param>
/// <param name="document"></param>
public static void GetDataFromXPath(this HtmlDocument doc, CrawlItem crawItem, IFreeDocument document)
{
var result = doc.DocumentNode.GetDataFromXPath(crawItem.XPath, crawItem.IsHTML);
if (result != null)
document.SetValue(crawItem.Name, result);
}
/// <summary>
/// 从批量集合中获取数据
/// </summary>
/// <param name="doc"></param>
/// <param name="crawItem"></param>
/// <param name="shortv"></param>
/// <param name="document"></param>
public static void GetDataFromXPath(this HtmlDocument doc, CrawlItem crawItem, string shortv,
IFreeDocument document)
{
var result = doc.DocumentNode.GetDataFromXPath(new XPath(crawItem.XPath).TakeOff(shortv).ToString(),
crawItem.IsHTML);
if (result != null)
document.SetValue(crawItem.Name, result);
}
