Created
February 19, 2015 17:33
-
-
Save RobThree/6eb86b8d732b39cc6284 to your computer and use it in GitHub Desktop.
Word table extractor
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
using System; | |
using System.Collections.Generic; | |
using System.IO; | |
using System.IO.Compression; | |
using System.Linq; | |
using System.Xml; | |
//Usage: | |
// var tables = new WordTableExtractor().ExtractTables(@"path\to\my\file.docx"); | |
public class WordTableExtractor | |
{ | |
public IEnumerable<IEnumerable<string[]>> ExtractTables(string filename) | |
{ | |
return ExtractTables(ReadDocxFile(filename, "word/document.xml")); | |
} | |
public IEnumerable<IEnumerable<string[]>> ExtractTables(string filename, string entry) | |
{ | |
return ExtractTables(ReadDocxFile(filename, entry)); | |
} | |
private static IEnumerable<IEnumerable<string[]>> ExtractTables(XmlDocument doc) | |
{ | |
var nsmgr = new XmlNamespaceManager(doc.NameTable); | |
nsmgr.AddNamespace("w", "http://schemas.openxmlformats.org/wordprocessingml/2006/main"); | |
return SelectNodes(doc.DocumentElement, "//w:tbl", nsmgr) //Find all tables | |
.Select(t => SelectNodes(t, "w:tr", nsmgr) //Find their rows | |
.Select(r => SelectNodes(r, "w:tc", nsmgr) //Find their cells | |
.Select(c => c.InnerText.Trim()).ToArray() //Get their textcontent | |
)); | |
} | |
private static IEnumerable<XmlElement> SelectNodes(XmlElement element, string xpath, XmlNamespaceManager nsmgr) | |
{ | |
return element.SelectNodes(xpath, nsmgr).Cast<XmlElement>(); | |
} | |
private static XmlDocument ReadDocxFile(string path, string entry) | |
{ | |
using (var s = File.OpenRead(path)) | |
{ | |
using (var z = new ZipArchive(s, ZipArchiveMode.Read)) | |
{ | |
var f = z.Entries.FirstOrDefault(e => e.FullName.Equals(entry, StringComparison.OrdinalIgnoreCase)); | |
if (f == null) | |
throw new FileNotFoundException(string.Format("Entry '{0}' was not found in '{1}'", entry, path)); | |
using (var sr = f.Open()) | |
{ | |
var x = new XmlDocument(); | |
x.Load(sr); | |
return x; | |
} | |
} | |
} | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment