DoxoEater2

A little C# application for extracting your documents from your Windows SharePoint Service 3.0 (WSS3) Database.  Builds upon the previous work of Plaxo, with some improvement needed for larger Document Library databases.  Works very well on our primarily Document Library WSS3 site which contains ~85,000 files in ~23,000 folders, totalling over 40 GB in size.

As part of our SharePoint Sever backup routine, I like to run this periodically to ensure we can extract our documents from SharePoint Document Library backup, even in a scenario when a functional SharePoint application can not be restored.

DoxoEater 2 - C# Source Code

---

using System;

using System.Collections.Generic;
using System.Text;
using System.Data;
using System.Data.SqlClient;
using System.IO;
 
namespace DoxoEater2
{
    class Program
    {
        static void Main(string[] args)
        {
 
            // WSS 3.0 default DB via SQL 2005 embedded edition.
            //string DBConnString = "Data Source=\\\\.\\pipe\\MSSQL$MICROSOFT##SSEE\\sql\\query;Database=WSS_Content;Trusted_Connection=yes;Timeout=2000;MultipleActiveResultSets=True";
            
            // Or localhost, WSS_Content, typical SQL connection style, TCP/IP listener.
            string DBConnString = "Data Source=localhost;Database=WSS_Content;Trusted_Connection=yes;Timeout=2000;MultipleActiveResultSets=True";
            string DocsQuery = "select [AllDocs].[Id], [AllDocs].[DirName], [AllDocs].[LeafName] from [AllDocs] where dirname <> '' order by dirname;";
 
            Console.WriteLine("START: " + DateTime.Now.ToUniversalTime() + " UTC");
 
            SqlConnection con;
 
            try // create a DB connection
            {   
                con = new SqlConnection(DBConnString);
                con.Open();
            }
            catch (Exception e) {
                Console.WriteLine(e.Message);
                return;
            }
 
            SqlCommand com = con.CreateCommand();
            com.CommandText = DocsQuery;
 
            // readers for SQL queries: 1) which docs? 2) featch binary data per DocID
            SqlDataReader reader, reader2;
 
            try {   // execute query
                reader = com.ExecuteReader();
            }
            catch (Exception e) {
                if (con != null) { con.Close(); }
                Console.WriteLine(e.Message); 
                return;
            }
 
            while (reader.Read()) { // rows of document ID, name and directory nodes
 
                // grab the file's directory and name
                Guid FileId = (Guid)reader["Id"];
                string DirName = (string)reader["DirName"];
                string LeafName = (string)reader["LeafName"];
           
                // create directory for the file if it doesn't yet exist
                if (!Directory.Exists(DirName))
                {
                    Directory.CreateDirectory(DirName);
                    Console.WriteLine("DIR: " + DirName);
                }
 
                // check if file already exists or not
                if (File.Exists(DirName + "/" + LeafName)) {
                    Console.WriteLine("ERROR: File Already Exists: " + DirName + "/" + LeafName);
                    continue;
                }
 
                SqlCommand com2 = con.CreateCommand();
                com2.CommandText = "select content from [AllDocStreams]  where  [AllDocStreams].[Id] = '" + FileId.ToString() + "';";
                Console.WriteLine("SQL: " + com2.CommandText.ToString());
 
                try
                {   // execute file fetch query
                    reader2 = com2.ExecuteReader(CommandBehavior.SequentialAccess);
                }
                catch (Exception e)
                {
                    if (con != null) { con.Close(); }
                    Console.WriteLine(e.Message);
                    return;
                }
 
                while(reader2.Read()) {
             
                    // create a filestream to spit out the file
                    FileStream fs = new FileStream(DirName + "/" + LeafName, FileMode.Create, FileAccess.Write);
                    BinaryWriter writer = new BinaryWriter(fs);
 
 
                    // depending on the speed of your network, you may want to change the buffer size (it's in bytes)
                    int bufferSize = 1048576;
                    long startIndex = 0;
                    long retval = 0;
                    byte[] outByte = new byte[bufferSize];
 
                    // grab the file out of the db one chunk (of size bufferSize) at a time
                    do
                    {
                        retval = reader2.GetBytes(0, startIndex, outByte, 0, bufferSize);
                        startIndex += bufferSize;
 
                        writer.Write(outByte, 0, (int)retval);
                        writer.Flush();
                    } while (retval == bufferSize);
 
                    // finish writing the file
                    Console.WriteLine("FILE: " + LeafName);
                    writer.Close();
                    fs.Close();
             
                }
                reader2.Close();
 
            }
 
            // close the DB connection and whatnots
            reader.Close();
            con.Close();
 
            Console.WriteLine("DONE: " + DateTime.Now.ToUniversalTime() + " UTC");
        }
    }
}