Any optimizations tips for this tiny c# function?

Question

I got an XML file that is 9MB large. Apparently, it is broken.

I want to check if on any level 2 sibling elements have an attribute "Id" with same value.

Currently it goes too slow. What kind of optimizations I could make to this code?

Edited to include some tips

namespace ConsoleApplication1{
  using System;
  using System.Collections.Generic;
  using System.Diagnostics;
  using System.IO;
  using System.Linq;
  using System.Xml.Linq;
  internal class Program{
    private const string _pathToXml=@"C:\4\4";
    private static readonly List<object> _duplicateLeafs=new List<object>();
    private static void Main(){
      var xml=ReadXml();
      var elements=xml.Descendants();
      foreach(var element in elements)
        FindDupes(element);
      Console.ReadLine();
      Debugger.Break();
    }
    private static XDocument ReadXml(){
      return XDocument.Parse(File.ReadAllText(_pathToXml));
    }
    private static void FindDupes(XElement element){
      var elements=element.Descendants();
      var elementsWithIds=elements.Where(x=>x.Attribute("Id")!=null);
      var ids=elementsWithIds.Select(x=>x.Attribute("Id")).ToList();
      for(var i=0;i<ids.Count;i++)
        for(var j=i+1;j<ids.Count;j++)
          if(i!=j&&ids[i]==ids[j])
            _duplicateLeafs.Add(elementsWithIds.First(x=>x.Attribute("Id")==ids[i]));
      foreach(var subElement in elements)
        FindDupes(subElement);
    }
  }
}

Answer 1

您说要检查2级后代，但是FindDupes是递归的，因此您要在每个调用的foreach循环中递归检查两个级别。

Answer 2

you are repeat checking things, so you could do int j = i+1 instead of starting at 0.

you wouldn't have to check i != j then

Answer 3

1. In your for loops store the list count to a variable rather than accessing the Count property.

   for (int i = 0, idCount = ids.Count; i++) { }

2. Store ids[i] to a local variable instead of looking it up in the collection more than once.

3. The biggest change would be to avoid nested for loops. Consider rewriting.

Edit: Made the following changes.

        private const string _pathToXml = @"C:\test.xml";
        private static readonly List<object> _duplicateLeafs = new List<object>();
        private static void Main()
        {
            var xml = ReadXml();
            var elements = xml.Descendants();
            FindDupes(elements);
        }

        private static void FindDupes(IEnumerable<XElement> elements)
        {
            foreach (var element in elements)
            {
                var subElements = element.Descendants();
                var subElementsWithIds = subElements.Where(x => x.Attribute("Id") != null).ToList();
                var ids = subElementsWithIds.Select(x => x.Attribute("Id")).ToList();
                var duplicates = ids.GroupBy(s => s.Value).SelectMany(grp => grp.Skip(1)).Distinct().ToList();

                if (duplicates != null)
                {
                    _duplicateLeafs.AddRange(duplicates);
                }
                FindDupes(subElements);
            }
        }

Using the following xml file:

<?xml version="1.0" encoding="utf-8" ?>
<persons>
  <person Id="1">
    <name>Michael</name>
    <age>29</age>
  </person>
  <person Id="1">
    <name>Rebecca</name>
    <age>29</age>
  </person>
  <person Id="2">
    <name>Matthew</name>
    <age>29</age>
  </person>
  <person Id="2">
    <name>Paul</name>
    <age>29</age>
  </person>
</persons>

After testing your latest version and the code I provided, while loading a 16 MB file here are the times:

Time: 2.8704708 seconds Lambda solution
Time: 692.043006 seconds Nested for loops