XMLDiff无法正确识别差异? [英] XMLDiff fails to recognize differences correcly?
问题描述
我在这里想念什么?
一个用于在两个XML文件之间进行比较的Helper类: XMLDiff是否应该考虑元素名称并寻求最佳匹配以正确识别以下更改? p>
a Helper class for making comparisons between two XML files:
public class XMLDiffer
{
public XDocument Diff(string originalXML, string changedXML)
{
//http://msdn2.microsoft.com/en-us/library/aa302294.aspx
XmlDiff xmlDiff = new XmlDiff(XmlDiffOptions.IgnoreChildOrder | XmlDiffOptions.IgnoreComments | XmlDiffOptions.IgnoreWhitespace);
xmlDiff.Algorithm = XmlDiffAlgorithm.Precise;
StringBuilder diffgramStringBuilder = new StringBuilder();
bool xmlComparisonResult = false;
using (StringReader legacySr = new StringReader(originalXML), nextgenSr = new StringReader(changedXML))
{
using (XmlReader legacyReader = XmlReader.Create(legacySr), nextgenReader = XmlReader.Create(nextgenSr))
{
using (StringWriter sw = new StringWriter(diffgramStringBuilder))
{
using (XmlWriter diffgramWriter = XmlWriter.Create(sw))
{
xmlComparisonResult = xmlDiff.Compare(legacyReader, nextgenReader, diffgramWriter);
}
}
}
}
XDocument xdoc = XDocument.Parse(diffgramStringBuilder.ToString());
return xdoc;
}
public string GetChangeHtml(string originalXML, string changedXML)
{
XmlDiffView view = new XmlDiffView();
var diffgram = Diff(originalXML, changedXML);
string ret = "";
using (StringReader legacySr = new StringReader(originalXML), diffGramSr = new StringReader(diffgram.ToString()))
{
using (XmlReader legacyReader = XmlReader.Create(legacySr), diffgramReader = XmlReader.Create(diffGramSr))
{
using (StringWriter sw = new StringWriter())
{
view.Load(legacyReader, diffgramReader);
view.GetHtml(sw);
ret = sw.ToString();
}
}
}
return ret;
}
}
具有以下测试:
With Following test:
[TestMethod]
public void XMLDiff_AreNotSame_GetChangeHtmlAll()
{
//Arrange
string source = "<root><child>some text</child><child>more text</child><child1>REMOVED</child1></root>";
//Ordering of the generic child nodes is not changed, but it might
string target = "<root><child>some text CHANGE</child><child>more text</child><child>ADDITION</child></root>";
XMLDiffer differ = new XMLDiffer();
//Act
var diffview = differ.GetChangeHtml(source, target);
//Assert
Assert.IsNotNull(diffview);
}
产生以下内容(添加了html和table元素):
https://pste.eu/p/Fm7Z.html
Produces following (html and table elements added): https://pste.eu/p/Fm7Z.html
有关库的更多信息: http://msdn2.microsoft .com / en-us / library / aa302294.aspx
参考链接: https://www.nuget.org/packages/XMLDiffPatch/
推荐答案
我最终实现了以下类来进行更改:
I ended up implementing following classes to get changes:
public class XMLComparer : IEqualityComparer<XNode>
{
public bool Equals(XNode e1, XNode e2)
{
if (!(e1 is XElement)) return true;
if (!(e2 is XElement)) return false;
var el1 = e1 as XElement;
var el2 = e2 as XElement;
return Tuple.Create(el1.Name, el1.Value).Equals(Tuple.Create(el2.Name, el2.Value));
}
public int GetHashCode(XNode n)
{
if (!(n is XElement)) return 0;
var el = n as XElement;
return Tuple.Create(el.Name, el.Value).GetHashCode();
}
}
public class XMLDifference
{
public bool IsNew { get; set; }
public XElement Node { get; set; }
}
public class XMLDifferenceComparer
{
public List<XMLDifference> GetDifferences(string original, string changed)
{
List<XMLDifference> ret = new List<XMLDifference>();
var originalDoc = XDocument.Parse(original);
var changedDoc = XDocument.Parse(changed);
//Get differences that are present in new xml version
var differences = changedDoc.Root.Descendants().Except(originalDoc.Root.Descendants(), new XMLComparer());
ret.AddRange(GetList(differences, true));
//Get differences that have changed since the old xml version
var oldValues = originalDoc.Root.Descendants().Except(changedDoc.Root.Descendants(), new XMLComparer());
ret.AddRange(GetList(oldValues, false));
return ret;
}
private List<XMLDifference> GetList(IEnumerable<XNode> nodes, bool isNew)
{
List<XMLDifference> ret = new List<XMLDifference>();
foreach (XNode d in nodes)
{
var diff = new XMLDifference();
diff.IsNew = isNew;
var el = d as XElement;
diff.Node = el;
ret.Add(diff);
}
return ret;
}
}
这可以识别更改,但不是特定于元素的由于缺少每个元素的唯一标识符,因此无法映射确切地更改了哪个元素以及更改的方式。
This can recognize changes but is not element specific, it cannot map which element exactly was changed and how, caused by lack of unique identifiers for each element.
此解决方案的主要思想来自于此: https://gist.github.com/krcourville/6933451
The main idea for this solution came from here: https://gist.github.com/krcourville/6933451
这篇关于XMLDiff无法正确识别差异?的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!