MapReduce输出C＃读取mongoDB

Question

public class myType
{
public int key1;
public int key2;
public int val1;
public int val2;
public int val3;
};

When I insert myType objects to collection there's no problems with reading. Just:

collection.FindAs<myType>(query);

get objects:

"key1":key1, "key2":key2, "val1":val1, "val2":val2, "val3":val3

But after mapreduce (key is constructed as combination of two fields) output collection has a slightly modified structure. Something like

"_id" {"key1" : key1, "key2": key2}, "value" : {"val1":val1, "val2":val2, "val3":val3}

What is the way to read object from output collection?

Answer 1

Once you have your output collection, you can query it like any other collection. However, because the structure is different, you won't be able to use the same C# class to do this. So, you'll either need to read this as a BsonDocument, or create a couple of new classes like below:

public class Keys
{
  [BsonElement("key1")]
  public int Key1;
  [BsonElement("key2")]
  public int key2;
}

public class values
{
  [BsonElement("val1")]
  public int Val1;
  [BsonElement("val1")]
  public int Val2;
  [BsonElement("val1")]
  public int Val3;
}

public class MapReduceOutput
{
  [BsonId]
  public Keys Keys;
  [BsonElement("values")]
  public Values Values;
}

Then you can do this:

collection.FindAs<MapReduceOutput>(query);

Hope this helps.

Answer 2

Simple for each through the m/r results supposed to work fast (at least faster than automatic deserializtion):

var result = RunMapReduce(..);
foreach (var item in result.InlineResults)
{
   var id = item["_id"].AsBsonDocument;
   var value = item["value"].AsBsonDocument;

   var obj = new myType();
   obj.key1 = id["key1"].AsInt32;
   //and so on
}