RavenDB inner collection paging

4
votes

I am new to RavenDB and from my understanding, when you ask for the document, you will get the entire document (unless you use some sort of index, etc).

Example Scenario

Take for example the Blog document scenario, where the document looks like this:

public class Blog
{
   public string Id { get; set; }
   public string AuthorId { get; set; }
   public DateTime PublishedUTC { get; set; }
   public string Title { get; set; }
   public string Content { get; set; }
   public Comment[] Comments { get; set; }
}

public class Comment
{
   public string Id { get; set; }
   public string AuthorId { get; set; }
   public DateTime PublishedUTC { get; set; }
   public string Content { get; set; }
}

Say we have a webpage /blogs/posts/. The page displays a paged set of the Blog Posts and the Comments for each of the blogs. I understand how to use the paging on the Blog documents with Skip() and Take() methods. I'd like to apply paging logic to the inner Comments collection for each of the Blog documents.

My Questions

  1. How would I get a paged set of Blogs and a paged set of each of their Comments?

  2. Given the paging requirements, would you change the given Blog document scenario so that the comments don't live within the Blog document?

1
c#nosqlravendb

1 Answers

3
votes

  1. If you were using .Skip() and .Take() to get a paged list of blog posts you were already doing it right. In order to get a paged list on the comments too, you can use the same methods on the in-memory list (it will be linq-to-objects then). So I suggest to change your Comment[] array into a List<Comment>, then you have these two linq methods available.

  2. If I'd change the post document so that it doesn't contain the comments, I would not have done that because of paging (there's no disadvantage, see my first point) but because of saved bandwidth in database-request. I would have 2 documents, one containing the post another containing all the comments. That way you don't need to load all the comments each time you want to display a list of posts but still have the document-database advantage when you actually need them. Using RavenDB it is also very easy to define indexes on your comment items, in case you need them somewhere else (sidebar or so). You can find exactly this implementation in RaccoonBlog.