Google DiffMatchPatch to NodeView

Question

I been struggling for some time with a XmlComparsion functionality. I have to compare two Xml for differences and then display them in a TreeView way.

I've done some research and found out Google Diff Match Patch library that looks exactly what I need for my comparison part.

https://github.com/google/diff-match-patch

I've implemented it as following

public List<Diff> DiffLineMode(string text1, string text2)
        {
            var a = base.diff_linesToChars(text1, text2);
            var lineText1 = a[0];
            var lineText2 = a[1];
            var lineArray = (IList<string>)a[2];
            var diffs = base.diff_main(lineText1.ToString(), lineText2.ToString(), false);
            base.diff_charsToLines(diffs, lineArray);

                  return diffs;
        }

Now my big challenge which I would need some help with is getting this difference into a treeview like object.

I've done a tree view class in which I parse both my Xml files used in comparison

public class TreeViewModel
    {
        public long ID { get; set; }
        public string NodeName { get; set; }
        public string AbsolutPath { get; set; }
        public string Value { get; set; }
        public string Xml { get; set; }
        public int Operation { get; set; } = 2;
        public List<TreeViewModel> Children { get; set; }
        public Dictionary<string, string> Attributes { get; set; }
    }

And now what I would like to do is make sure the Operation from TreeView matches the Operation result from DIffMatchPatch, but my problem is that DiffMatchPatch compares Xml as string and does not consider at all Xml as a node tree. As a result for 2 simple Xml book files.

<?xml version="1.0"?>
<catalog>
   <book id="bk101">
      <author>Gambardella, Matthew</author>
      <title>XML Developer's Guide</title>
      <genre>Computer</genre>
      <price>44.95</price>
      <publish_date>2000-10-01</publish_date>
      <description>An in-depth look at creating applications 
      with XML.</description>
   </book>
   <book id="bk112">
      <author>Ralls, Kim</author>
      <title>Midnight Rain Update but should not reflect</title>
      <genre>Fantasy</genre>
      <price>5.95</price>
      <publish_date>2000-12-16</publish_date>
      <description>A former architect battles corporate zombies, 
      an evil sorceress, and her own childhood to become queen 
      of the world.</description>
   </book>
   <book id="bk999">
      <author>O'Brien, Tim</author>
      <title>Microsoft .NET: The Programming Bible</title>
      <genre>Computer</genre>
      <price>36.95</price>
      <publish_date>2000-12-09</publish_date>
      <description>Microsoft's .NET initiative is explored in 
      detail in this deep programmer's reference.</description>
   </book>  
   <book id="bk998">
      <author>Chris Tucker</author>
      <title>This Is my Churck</title>
      <genre>Music</genre>
      <price>49.99</price>
      <publish_date>2021-11-03</publish_date>
      <description>This is a custom description from Sebi Ciuca</description>
      <coauthor>Sebi</coauthor>
   </book>    
   <book id="bk103">
      <author>Corets, Eva</author>
      <title>Maeve Ascendant</title>
      <genre>Fantasy</genre>
      <price>5.95</price>
      <publish_date>2000-11-17</publish_date>
      <description>After the collapse of a nanotechnology 
      society in England, the young survivors lay the 
      foundation for a new society.</description>
   </book>
</catalog>

and

<?xml version="1.0"?>
<catalog>
   <book id="bk101">
      <author>Gambardella, Matthew</author>
      <title>XML Developer's Guide</title>
      <genre>Computer</genre>
      <price>44.95</price>
      <publish_date>2000-10-01</publish_date>
      <description>An in-depth look at creating applications 
      with XML.</description>
   </book>
   <book id="bk102">
      <author>Ralls, Kim</author>
      <title>Midnight Rain</title>
      <genre>Fantasy</genre>
      <price>5.95</price>
      <publish_date>2000-12-16</publish_date>
      <description>A former architect battles corporate zombies, 
      an evil sorceress, and her own childhood to become queen 
      of the world.</description>
   </book>
   <book id="bk103">
      <author>Corets, Eva</author>
      <title>Maeve Ascendant</title>
      <genre>Fantasy</genre>
      <price>5.95</price>
      <publish_date>2000-11-17</publish_date>
      <description>After the collapse of a nanotechnology 
      society in England, the young survivors lay the 
      foundation for a new society.</description>
   </book>
   <book id="bk997">
      <author>Corets, Eva</author>
      <title>Maeve Ascendant Part II</title>
      <genre>Fantasy</genre>
      <price>15.95</price>
      <publish_date>2022-01-17</publish_date>
      <description>After one collapse of a nanotechnology 
      society in England, the young survivors lay the 
      foundation for a new society.</description>
   </book>
</catalog>

it will output this

 "difference": [
    {
      "operation": 2,
      "text": "<catalog>\r\n  <book id=\"bk101\">\r\n    <author>Gambardella, Matthew</author>\r\n    <title>XML Developer's Guide</title>\r\n    <genre>Computer</genre>\r\n    <price>44.95</price>\r\n    <publish_date>2000-10-01</publish_date>\r\n    <description>An in-depth look at creating applications \r\n      with XML.</description>\r\n  </book>\r\n"
    },
    {
      "operation": 0,
      "text": "  <book id=\"bk112\">"
    },
    {
      "operation": 1,
      "text": "  <book id=\"bk102\">"
    },
    {
      "operation": 2,
      "text": "    <author>Ralls, Kim</author>\r\n    <genre>Fantasy</genre>\r\n    <price>5.95</price>\r\n    <publish_date>2000-12-16</publish_date>\r\n    <description>A former architect battles corporate zombies, \r\n      an evil sorceress, and her own childhood to become queen \r\n      of the world.</description>\r\n  </book>\r\n"
    },
    {
      "operation": 0,
      "text": "  <book id=\"bk999\">    <author>O'Brien, Tim</author>    <title>Microsoft .NET: The Programming Bible</title>    <genre>Computer</genre>    <price>36.95</price>    <publish_date>2000-12-09</publish_date>    <description>Microsoft's .NET initiative is explored in       detail in this deep programmer's reference.</description>  </book>  <book id=\"bk998\">    <author>Chris Tucker</author>    <title>This Is my Churck</title>    <genre>Music</genre>    <price>49.99</price>    <publish_date>2021-11-03</publish_date>    <description>This is a custom description from Sebi Ciuca</description>    <coauthor>Sebi</coauthor>  </book>"
    },
    {
      "operation": 2,
      "text": "  <book id=\"bk103\">\r\n    <author>Corets, Eva</author>\r\n    <title>Maeve Ascendant</title>\r\n    <genre>Fantasy</genre>\r\n    <price>5.95</price>\r\n    <publish_date>2000-11-17</publish_date>\r\n    <description>After the collapse of a nanotechnology \r\n"
    },
    {
      "operation": 1,
      "text": "      society in England, the young survivors lay the       foundation for a new society.</description>  </book>  <book id=\"bk997\">    <author>Corets, Eva</author>    <title>Maeve Ascendant Part II</title>    <genre>Fantasy</genre>    <price>15.95</price>    <publish_date>2022-01-17</publish_date>    <description>After one collapse of a nanotechnology "
    },
    {
      "operation": 2,
      "text": "      society in England, the young survivors lay the \r\n      foundation for a new society.</description>\r\n  </book>\r\n</catalog>"
    }
  ]

which looks something like this in a more "readable" view

Now as you can see if you nave 2 or more nodes difference, they are "put" together inside as a "hole difference" this makes it very tricky for me to parse it into a treeview like object.

Here you can also see for book997 which is a hole new book but with the same description as book103, it will consider the "added" text beging from description of book 103 and ending before the description of book997. This in a way I would like to translte into Node that both nodes have changes, but I don't know what I can use to match a specific TreeView Node with a part or hole of a difference.

Hope I made myself clear. I start to feel I started very wrong with GoogleDiffMatchPatch for what I need, but at this point it would be even harder for me to create a custom xml comparer based on nodes.

EDIT : Added also the Parsing from XML To TreeView

 public TreeViewModel ToTreeView(XDocument xmlDocument)
        {
            var rootNode = xmlDocument.Root;

            return ParseNode(rootNode, 1);
        }

  private TreeViewModel ParseNode(XElement node, int parentId)
        {
            if (node.HasElements)
            {
                var index = 0;
                var children = new List<TreeViewModel>();
                bool hasChild = false;
                foreach (var element in node.Elements())
                {
                    children.Add(ParseNode(element, parentId * 10 + index, panoramaId));
                    index++;
                }

                var treeNode = new TreeViewModel
                {
                    Children = children,
                    Value = hasChild ? string.Empty : node.Value,
                    AbsolutPath = node.GetAbsoluteXPath(),
                    ID = parentId,
                    Xml = node.ToString(),
                    NodeName = node.Name.ToString(),
                    Attributes = node.GetAttributes()
                };
            }
            else
            {
                var treeNode = new TreeViewModel
                {
                    ID = parentId,
                    Value = node.Value,
                    AbsolutPath = node.GetAbsoluteXPath(),
                    NodeName = node.Name.ToString(),
                    Attributes = node.GetAttributes(),
                    Xml = node.ToString(),
                };
            }
        }

Added git repository : https://github.com/SebiCiuca/DiffMatchPatchToNodeView

score -1 · Answer 1 · answered Jan 07 '22 at 13:31

-1

See if following helps :

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Xml;
using System.Xml.Serialization;

namespace ConsoleApplication11
{
    class Program
    {
        const string FILENAME1 = @"c:\temp\test.xml";
        const string FILENAME2 = @"c:\temp\test1.xml";
        static void Main(string[] args)
        {
            TreeViewModel model = new TreeViewModel(FILENAME1, FILENAME2);
        }
    }
    public class TreeViewModel
    {
        public string AbsolutPath { get; set; }
        public string Value { get; set; }
        public int Operation { get; set; }
        public Catalog catalog1 { get;set;}
        public Catalog catalog2 { get;set;}
        public Dictionary<string, Book> catalog1Dict { get; set; }
        public Dictionary<string, Book> catalog2Dict { get; set; }

        public TreeViewModel(string filename1, string filename2)
        {
            XmlSerializer serializer = new XmlSerializer(typeof(Catalog));
            XmlReader reader = XmlReader.Create(filename1);

            catalog1 = (Catalog)serializer.Deserialize(reader);
            catalog1Dict = catalog1.books.GroupBy(x => x.id).ToDictionary(x => x.Key, y => y.FirstOrDefault());

            reader = XmlReader.Create(filename2);
            catalog2 = (Catalog)serializer.Deserialize(reader);
            catalog2Dict = catalog2.books.GroupBy(x => x.id).ToDictionary(x => x.Key, y => y.FirstOrDefault());

        }
    } 
    [XmlRoot("catalog")]
    public class Catalog
    {
        [XmlElement("book")]
        public List<Book> books { get;set;}
    }
    public class Book
    {
        [XmlAttribute()]
        public string id { get;set;}
        public string author { get;set;}
        public string title { get;set;}
        public string genre { get;set;}
        public decimal price { get;set;}
        public DateTime publish_date { get;set;}
        public string description { get;set;}
    }
}

answered Jan 07 '22 at 13:31

jdweng

33,250
2
15
20

I don't see how this helps me with my problem, I don't have a problem serializing/deserializing my example Xml to a TreeView, I've done that using a recursive function. My problem is with matching my nodes xml with output from DiffMatchPatch. – CiucaS Jan 07 '22 at 13:51
You said "Xml as string and does not consider at all Xml as a node tree". You never said you deserialized the xml. I provided the code to deserialize. Also you only have one list of books and you need two lists of books and then compare the two lists of books using two dictionaries. – jdweng Jan 07 '22 at 14:57
There is no point in arguing, I detailed my problem at the end of the post. It's about matching the serialized XML with result string from GoogleDiffMatchPatch to be short. Also your solution provides a very "static" way of creating a treeview. My XML is way more complex and not the same always, I don't a class to serialize into. I have to use recursive and create nodes. This book was mainly an example to state the problem. – CiucaS Jan 07 '22 at 15:07
Where are you deserializing the data? You are trying to compare xml string with xml string. Not deserialized elements. I just provided the code to deserialized. You now had to match id or one list to id in 2nd list. XML is not like java deserialization. Java you do not need to create the classes. In xml you do need to create classes using Net library methods. The library you are trying to use is for text and not xml. – jdweng Jan 07 '22 at 15:21
I've added the tree view parsing. I think you missed my point totally. I know I am comparing string with string in DiffMatchPatch, that is exactly what is my problem, that I have in a way or an other get that "diff string" to match my nodes. – CiucaS Jan 10 '22 at 07:41
You are getting the MODEL of the data and the TreeView mixed up. This statement does not make any sense "public List Children { get; set; }". You first have to deserialize two xml files into two models (deserialize). Then compare the two models and put results into the TreeView. You will compare the two models by ID and have three conditions 1) IDs that are in both models 2) IDs that are in first model and not seconds model 3) IDs that are not in first model and are in second model. So you will use a OUTER JOIN to compare the ids of the two models. – jdweng Jan 10 '22 at 10:19
You keep focusing on this particular XML, like I said this is an example. I can't compare XML using id's, they are GENERIC xml which means it can be any kind of XML in this hole world. Children make sense as in a TreeView you have parent->children relationship. – CiucaS Jan 10 '22 at 10:32
The XML library in Net is not like JSON that you can do a generic deserizlize. So it looks like you are thinking in the right direction by doing " public Dictionary Attributes". The dictionary is one book in one input file. So you should have List< Dictionary> model1 and List< Dictionary> model2. – jdweng Jan 10 '22 at 10:39
Yes you can use generic deserialize by using recursive methods. Please check the update answer, I've created a GitRepo for it. – CiucaS Jan 10 '22 at 10:50
Yes treeveiw has children. But the structure of each node/child is the data. You have two xml nodes in each model that you are comparing, not one. – jdweng Jan 10 '22 at 10:53
You need have two input xmls and need compare the two deserizlized structures. So at each recursive level you need to compare three cases 1) both have same ID 2) ID is in first but not second 3) ID is in second but not first. In your case you have two recursive levels. A) List books B) The properties of each book where one property is the ID. So first task is to join the TWO XML (lists of BOOKS) files by ID. – jdweng Jan 10 '22 at 12:21
Please move on to a other questions, I don't want to be rude, but it clear you don't get what I want, and you keep telling me to focus on "id" and use for comparison, when I am telling you that the real XML data I have to compare have no ID's and no attributes, this is just an example. – CiucaS Jan 10 '22 at 13:36
I know exactly what you are doing. You need two lists (one for each xml input) and need to Outer Join the two list do do what you want. You can't compare two lists without a join. What happens if one list contains A,C,D,E, F and the other A,B,F. If you just compare in sequential order you end up comparing 1) A - A 2) C - B 3) D - F. – jdweng Jan 10 '22 at 14:33
that is already done by GoogleDiffMatchPatch. – CiucaS Jan 10 '22 at 14:40
You cannot compare books 102 and 103 as a pair. You have to compare the dictionaries of the two xml for 102 with each other, then compare the dictionaries of the two xml for 103 with each other. – jdweng Jan 10 '22 at 17:56
I understand what you are saying, but we do diff by lines not by "nodes". The XML that we compare is generated by an other app, this means that I don't actually care a lot of the order, 103 will always be in the "same" place. – CiucaS Jan 11 '22 at 08:45
What is highlighted in GREEN on the right side is TWO different books. Why? – jdweng Jan 11 '22 at 10:11
You have to do compare by nodes not lines. You have different number of lines in the two xml sources and the comparison has to be by books and the books do not start on the same row. So to do book comparisons you have to first parse the xml into books and then compare the books. – jdweng Jan 12 '22 at 09:56

Google DiffMatchPatch to NodeView

1 Answers1