Word OpenXML . Traversing OpenXmlElements between bookmarks

问题

I need to traverse nodes between a bookmark start and a bookmark end tag. The problem appears to break down into a tree traversal but I am having trouble pinning down the correct algorithm. The bookmark start and end elements are non-composite nodes (no children) and may appear at an arbitrary depth in the tree. Bookmark start are also not guaranteed to be a at the same depth.

If you draw the tree structure for the document I would want to examine all nodes between the start and end bookmark. I think an algorithm to traverse an unbalanced tree starting at node x and ending at node y would work. Does this sounds feasible or am I missing something.

If this is feasible could you point me in the direction of a tree traversal that could accomplish returning the nodes?

回答1:

This depends on what you want to do, however, if you are primarily interested in the text between two bookmarks, then this is one of those cases where XmlDocument / XPath semantics are easier to use than LINQ to XML or the strongly-typed object model of the Open XML SDK V2. The semantics of the 'following::*' axis of XPath is what you want. The following example uses XmlDocument and XPath to print the names of the nodes between the start and end of a bookmark.

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

class Program
{
    public static XmlDocument GetXmlDocument(OpenXmlPart part)
    {
        XmlDocument xmlDoc = new XmlDocument();
        using (Stream partStream = part.GetStream())
        using (XmlReader partXmlReader = XmlReader.Create(partStream))
            xmlDoc.Load(partXmlReader);
        return xmlDoc;
    }

    static void Main(string[] args)
    {
        using (WordprocessingDocument doc =
            WordprocessingDocument.Open("Test.docx", false))
        {
            XmlDocument xmlDoc = GetXmlDocument(doc.MainDocumentPart);
            string wordNamespace =
                "http://schemas.openxmlformats.org/wordprocessingml/2006/main";
            XmlNamespaceManager nsmgr =
                new XmlNamespaceManager(xmlDoc.NameTable);
            nsmgr.AddNamespace("w", wordNamespace);
            XmlElement bookmarkStart = (XmlElement)xmlDoc.SelectSingleNode("descendant::w:bookmarkStart[@w:id='0']", nsmgr);
            XmlNodeList nodesFollowing = bookmarkStart.SelectNodes("following::*", nsmgr);
            var nodesBetween = nodesFollowing
                .Cast<XmlNode>()
                .TakeWhile(n =>
                    {
                        if (n.Name != "w:bookmarkEnd")
                            return true;
                        if (n.Attributes.Cast<XmlAttribute>().Any(a => a.Name == "w:id" && a.Value == "0"))
                            return false;
                        return true;
                    });
            foreach (XmlElement item in nodesBetween)
            {
                Console.WriteLine(item.Name);
                if (item.Name == "w:bookmarkStart" || item.Name == "w:bookmarkEnd")
                    foreach (XmlAttribute att in item.Attributes)
                        Console.WriteLine("{0}:{1}", att.Name, att.Value);
            }
        }
    }
}

回答2:

I've put together an algorithm that can easily retrieve the text of a bookmark.

How to Retrieve the Text of a Bookmark from an OpenXML WordprocessingML Document

I've also written code to replace the text of a bookmark:

Replacing Text of a Bookmark in an OpenXML WordprocessingML Document

-Eric

来源：https://stackoverflow.com/questions/1718663/word-openxml-traversing-openxmlelements-between-bookmarks