Migration to BlogEngine.NET

It was about time I migrated my weblog, based on .Text version 0.95 to a more modern blogging platform. As you will probably know, the codebase of .Text was integrated to Community Server ages ago. I doubt whether much of that original code is still alive in CS, if any at all.

Anyway, both BlogEngine.NET and dasBlog seemed like good alternatives. Subtext too. I decided to go with BlogEngine.NET. It’s not that I did extensive research on all the options, as long I was able to migrate the content and use Live Writer for posting. A post by Erik Lane helped me through it. He also posts instructions on how to keep old links working. I didn’t want to mess in the code of BlogEngine.NET, so after updating the BE.NET database, I just added the following hack to global.asax:

    protected void Application_BeginRequest(object sender, EventArgs e)
    {   
        HttpContext context = HttpContext.Current;
        
        // Handle .Text rss feed requests.
        if (context.Request.Url.PathAndQuery.ToLower().Contains("/rss.aspx"))
        {
            context.Response.Redirect("syndication.axd", false);
            context.Response.StatusCode = 301;
            context.Response.End();
        }
        
        // Handle .Text requests from archived posts
        if (context.Request.Url.PathAndQuery.Contains("/archive/"))
        {
            // lookup post
            int dotTextPostId = GetPostId(context.Request.Url.PathAndQuery);
            System.Collections.Generic.Dictionary<int, string> posts = GetPosts();
 
            string relativeLink = string.Empty;
            if (posts.ContainsKey(dotTextPostId))
                relativeLink = posts[dotTextPostId];
                
            string redirUrl = string.Format("{0}://{1}{2}", 
                  context.Request.Url.Scheme, context.Request.Url.Authority, 
                  relativeLink);            
            context.Response.Redirect(redirUrl, false);
            context.Response.StatusCode = 301;
            context.Response.End();
        }
    }
 
    private static readonly Regex DOTTEXT = 
        new Regex(@"/archive/\d{4}/\d{2}/\d{2}/(?<postId>\d+)\.aspx$", 
        RegexOptions.IgnoreCase | RegexOptions.Compiled);
 
    private int GetPostId(string url)
    {
        Match match = DOTTEXT.Match(url);
        if (match.Groups.Count > 0)
        {
            int postId;
            if (int.TryParse(match.Groups["postId"].ToString(), out postId))
            {
                return postId;
            }
        }
        return 0;
    }
    
    // Retrieve a list of old posts, cache the result
    protected System.Collections.Generic.Dictionary<int, string> GetPosts()
    {
        System.Collections.Generic.Dictionary<int, string> posts = 
            HttpContext.Current.Cache["posts"] as 
            System.Collections.Generic.Dictionary<int, string>;
        if (posts != null) return posts;
        
        posts = new System.Collections.Generic.Dictionary<int, string>();
 
        string query = 
     "SELECT PostID, DottextPostID FROM be_Posts WHERE DottextPostID IS NOT NULL";
        
        System.Data.SqlClient.SqlDataAdapter sqlDa = 
            new System.Data.SqlClient.SqlDataAdapter(query, 
            ConfigurationManager.ConnectionStrings["BlogEngine"].ConnectionString);
 
        System.Data.DataTable postTable = new System.Data.DataTable();
        sqlDa.Fill(postTable);
        foreach (System.Data.DataRow item in postTable.Rows)
        {
            Post p = Post.GetPost((Guid)item["PostID"]);
            posts.Add(int.Parse(item["DottextPostID"].ToString()), p.RelativeLink);
        }
 
        HttpContext.Current.Cache.Add("posts", posts, null, DateTime.MaxValue, 
            Cache.NoSlidingExpiration, CacheItemPriority.Normal, null);
        return posts;
    }

It may not be the most well-designed code, but it does the trick.