比较不同长度的2个Word文档

Question

我正在尝试比较 2 个单词的文档。我的目标是获得一个列表，其中包含两个文件中的每一行以及它们是否匹配。我有那部分工作。我的问题是，如果一个文件大于另一个文件，则较大文件中的文本不会添加到上述列表中。在某些情况下，一份文件可能比另一份文件大。这是一个文档修订系统。当前版本中的文本可能比以前的版本多，反之亦然。

到目前为止，我有这个代码。我修改了找到here的例子。

这是我一直在使用的两个示例文件（word 文档）
Test1.docx：

Test

This is a test document. It was created May 31.
The contents of this document are:
Unknown

Test2.docx：

Test

This is a test document. It was created Apr 1.
The contents of this document are:
Test Item 1
Test Item 2

这是我的 Program.cs 文件，这是我编辑的地方。在 CompareDocuments 方法中：

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Xml.Linq;
using DocumentFormat.OpenXml.Packaging;

namespace DocxDiff
{
    public class Program
    {
        private static List<DocumentCompare> _differences = new List<DocumentCompare>();

        public static string GetParagraphText(XElement p)
        {
            return p.Descendants(W.r)
                .Where(e => e.Parent.Name != W.del && e.Parent.Name != W.moveFrom)
                .Descendants(W.t)
                .Select(t => (string) t)
                .StringConcatenate();
        }

        public static List<DocumentCompare> CompareDocuments(WordprocessingDocument doc1, WordprocessingDocument doc2)
        {
            XDocument xDoc1 = doc1.MainDocumentPart.GetXDocument();
            XDocument xDoc2 = doc2.MainDocumentPart.GetXDocument();

            var doc1Elements = xDoc1
                .Descendants()
                .Where(e => e.Name != W.commentRangeStart
                            && e.Name != W.commentRangeEnd
                            && e.Name != W.proofErr
                            && !e.Ancestors(W.p).Any());
            var doc2Elements = xDoc2
                .Descendants()
                .Where(e => e.Name != W.commentRangeStart
                            && e.Name != W.commentRangeEnd
                            && e.Name != W.proofErr
                            && !e.Ancestors(W.p).Any());

            List<DocumentCompare> differences = new List<DocumentCompare>();

            IEnumerable<bool> correspondingElementEquivalency = doc1Elements.Zip(doc2Elements, (e1, e2) =>
            {
                // if the lines are different, set to true
                bool difference = false;
                if (e1.Name != e2.Name)
                {
                    return false;
                }

                if (e1.Name == W.p && e2.Name == W.p)
                {
                    if ((GetParagraphText(e1) != GetParagraphText(e2)))
                    {
                        // there is a difference between the documents
                        difference = true;
                    }

                    // record lines
                    differences.Add(new DocumentCompare() { Document1Text = e1.Value, Document2Text = e2.Value, Difference = difference });
                }

                // this is from the code in the link above
                // this method does not return a bool, it returns the list of differences
                return true;
            });

            // determine if the documents are equivalent
            // this has to be here to run the code above
            bool test = correspondingElementEquivalency.Any(e => e != true);

            return differences;
        }

        public static void Main(string[] args)
        {
            var doc1Path = @"C:\Diff\Test1.docx";
            var doc2Path = @"C:\Diff\Test2.docx";

            using(WordprocessingDocument doc1 = WordprocessingDocument.Open(doc1Path, false))
            using(WordprocessingDocument doc2 = WordprocessingDocument.Open(doc2Path, false))
            {
                _differences = CompareDocuments(doc1, doc2);

                foreach (var t in _differences)
                {
                    Console.WriteLine("Difference: {0}\nDoc 1: {1}\nDoc 2: {2}", t.Difference, t.Document1Text, t.Document2Text);
                }
            }

            Console.Read();
        }
    }
}

这是我用来存储比较文档的类：

public class DocumentCompare
{
    public string Document1Text { get; set; }
    public string Document2Text { get; set; }
    public bool Difference { get; set; }
}

这是我的 Extensions.cs 文件（来自教程，未修改）：

using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Text;
using System.Xml;
using System.Xml.Linq;
using DocumentFormat.OpenXml.Packaging;

namespace DocxDiff
{
    public static class Extensions
    {
        public static XDocument GetXDocument(this OpenXmlPart part)
        {
            XDocument xdoc = part.Annotation<XDocument>();

            if (xdoc != null)
                return xdoc;

            using (StreamReader streamReader = new StreamReader(part.GetStream()))
                xdoc = XDocument.Load(XmlReader.Create(streamReader));

            part.AddAnnotation(xdoc);
            return xdoc;
        }

        public static string StringConcatenate(this IEnumerable<string> source)
        {
            StringBuilder sb = new StringBuilder();
            foreach (var s in source)
                sb.Append(s);
            return sb.ToString();
        }

        public static IEnumerable<TResult> Zip<TFirst, TSecond, TResult>(this IEnumerable<TFirst> first, IEnumerable<TSecond> second, Func<TFirst, TSecond, TResult> func)
        {
            var ie1 = first.GetEnumerator();
            var ie2 = second.GetEnumerator();

            while (ie1.MoveNext() && ie2.MoveNext())
            {
                yield return func(ie1.Current, ie2.Current);
            }
        }
    }
}

这是本教程也给出的 W.cs 类：

public static class W
{
    public static XNamespace w = "http://schemas.openxmlformats.org/wordprocessingml/2006/main";

    public static XName p = w + "p";
    public static XName r = w + "r";
    public static XName t = w + "t";
    public static XName commentRangeStart = w + "commentRangeStart";
    public static XName commentRangeEnd = w + "commentRangeEnd";
    public static XName proofErr = w + "proofErr";
    public static XName del = w + "del";
    public static XName moveFrom = w + "moveFrom";
}

编辑：我很确定我需要修改 Zip 方法以在较大文件中添加行并为另一个文件添加一个空字符串。我尝试（不成功）修改它以使其工作（来自here）：

static void Main() {
    var a = new List<int> { 1, 2, 3 };
    var b = new List<int> { 1, 2, 3, 4, 5 };
    foreach (var c in a.Merge(b, (x, y) => x + y)) {
        Console.WriteLine(c);
    }
}
static IEnumerable<T> Merge<T>(this IEnumerable<T> first,
        IEnumerable<T> second, Func<T, T, T> operation) {
    using (var iter1 = first.GetEnumerator())
    using (var iter2 = second.GetEnumerator()) {
        while (iter1.MoveNext()) {
            if (iter2.MoveNext()) {
                yield return operation(iter1.Current, iter2.Current);
            } else {
                yield return iter1.Current;
            }
        }
        while (iter2.MoveNext()) {
            yield return iter2.Current;
        }
    }
}

我很确定我必须做这样的事情。合并文档。不知何故，我需要向较小的文档添加一个空列表项。

编辑：这是我刚刚想出的，它有点工作，只是它不会显示列表（单词列表）中的文档中的项目。我把它放在 Program.cs 中的 DocumentCompare 方法中，就在bool test... 之后和return differences; 之前：

// get document sizes
        var largerDoc = doc1Elements.Count() > doc2Elements.Count() && doc1Elements.Count() != doc2Elements.Count() ? doc1Elements : doc2Elements;
        var smallerDocCount = doc1Elements.Count() < doc2Elements.Count() && doc1Elements.Count() != doc2Elements.Count() ? doc1Elements.Count() : doc2Elements.Count();
        var doc1Larger = doc1Elements.Count() > doc2Elements.Count() && doc1Elements.Count() != doc2Elements.Count() ? true : false;
        var doc1Arr = doc1Elements.ToArray();
        var doc2Arr = doc2Elements.ToArray();

        // add in the remaining text for the larger document
        for (var i = smallerDocCount; i < largerDoc.Count(); i++)
        {
            // if doc1 is larger, add doc 1 and null for doc 2
            if (doc1Larger)
            {
                Console.WriteLine("doc1 Text: {0}", doc1Arr[i].Value);
                differences.Add(new DocumentComparison() { Document1Text = doc1Arr[i].Value, Document2Text = "", Difference = true });
            }
            else if(!doc1Larger) {
                Console.WriteLine("doc2 Text: {0}", doc2Arr[i].Value);
                differences.Add(new DocumentComparison() { Document1Text = "", Document2Text = doc2Arr[i].Value, Difference = true });
            }
        }

Answer 1

好的，这就是我想出的。 （它可能不是最有效的，但我相信它是这个项目的一个很好的解决方案）。这是代码（下面的解释）：

public static List<DocumentCompare> CompareDocuments(WordprocessingDocument doc1, WordprocessingDocument doc2)
    {
        XDocument xDoc1 = doc1.MainDocumentPart.GetXDocument();
        XDocument xDoc2 = doc2.MainDocumentPart.GetXDocument();

        // these queries return the elements that contain text in the word documents
        var doc1Elements = xDoc1
            .Descendants()
            .Where(e => e.Name != W.commentRangeStart
                        && e.Name != W.commentRangeEnd
                        && e.Name != W.proofErr
                        && !e.Ancestors(W.p).Any());
        var doc2Elements = xDoc2
            .Descendants()
            .Where(e => e.Name != W.commentRangeStart
                        && e.Name != W.commentRangeEnd
                        && e.Name != W.proofErr
                        && !e.Ancestors(W.p).Any());

        List<DocumentCompare> differences = new List<DocumentCompare>();

        IEnumerable<bool> correspondingElementEquivalency = doc1Elements.Zip(doc2Elements, (e1, e2) =>
        {
            bool difference = false;
            if (e1.Name != e2.Name)
            {
                return false;
            }

            if (e1.Name == W.p && e2.Name == W.p)
            {
                // e1.Name == W.p && 
                if ((GetParagraphText(e1) != GetParagraphText(e2)))
                {
                    // there is a difference between the documents
                    difference = true;
                }

                // record lines
                differences.Add(new DocumentCompare() { Document1Text = e1.Value, Document2Text = e2.Value, Difference = difference });
            }
            return true;
        });

        // determine if the documents are equivalent
        bool test = correspondingElementEquivalency.Any(e => e != true);

        var doc1Values = (from ie1 in doc1Elements where ie1.Name == W.p select ie1.Value).ToList();
        var doc2Values = (from ie2 in doc2Elements where ie2.Name == W.p select ie2.Value).ToList();

        // determine the larger document to add the remainder of that document to the list
        var largerDoc = doc1Values.Count() > doc2Values.Count() && doc1Values.Count() != doc2Values.Count() ? doc1Values : doc2Values;
        var smallerDocCount = doc1Values.Count() < doc2Values.Count() && doc1Values.Count() != doc2Values.Count() ? doc1Values.Count() : doc2Values.Count();
        var doc1Larger = doc1Values.Count() > doc2Values.Count() && doc1Values.Count() != doc2Values.Count() ? true : false;
        var doc1Arr = doc1Values.ToArray();
        var doc2Arr = doc2Values.ToArray();

        // add in the remaining text for the larger document
        for (var i = smallerDocCount; i < largerDoc.Count(); i++)
        {
            // if doc1 is larger, add doc 1 and null for doc 2
            if (doc1Larger)
            {
                Console.WriteLine("doc1 Text: {0}", doc1Arr[i]);
                differences.Add(new DocumentCompare() { Document1Text = doc1Arr[i], Document2Text = "", Difference = true });
            }
            else if(!doc1Larger) {
                Console.WriteLine("doc2 Text: {0}", doc2Arr[i]);
                differences.Add(new DocumentCompare() { Document1Text = "", Document2Text = doc2Arr[i], Difference = true });
            }
        }

        return differences;
    }

好的，所以，IEnumerable 完成到最小文档结束的位置。然后，我得到较大文档的其余部分（从较小文档的末尾开始）并将其附加到列表的末尾。我仍然需要收紧一些代码，但它可以工作。