Built-In Reduce Functions

Currently (CouchDB 0.11.0) there are three built-in reduce functions. Built-in reduce functions are performed right inside CouchDB implemented in Erlang. In most cases it is very fast because they are way more efficient.

Usage

To use built-in reduce functions, you simply need to replace your reduce function with a string containing _count, _sum or _stats (without any preceding or tailing whitespace).

Here is an example design document using build-in reduce functions:

{
  "_id":"_design/company",
  "_rev":"12345",
  "language": "javascript",
  "views":
  {
    "all_customers": {
      "map": "function(doc) { if (doc.type == 'customer')  emit(doc.id, 1) }",
      "reduce" : "_count"
    },
    "total_purchases_by_customer": {
      "map": "function(doc) { if (doc.type == 'purchase')  emit(doc.customer_id, doc.amount) }",
      "reduce": "_sum"
    }
  }
}

Available Build-In Functions

_sum

_sum just sums up the emitted values. Therefore the mapped values need to be numbers.

The JavaScript equivalent is:

function(keys, values, rereduce) {
  return sum(values);
}

_count

_count counts the emitted values. It's like _sum for emit(foo, 1). The map function may emit anything you want.

The JavaScript equivalent is:

function(keys, values, rereduce) {
  if (rereduce) {
    return sum(values);
  }
  else {
    return values.length;
  }
}

_stats

_stats calculates various numerical statistics on your emitted data. _stats requires, just like _sum, your emitted values to be numbers.

The reduce output looks like this:

{"update_seq":6,"rows":[
{"key":null,"value":{"sum":2,"count":2,"min":1,"max":1,"sumsqr":2}}
]}

sum and count is the equiverlent to _sum and _count. min and max are the minimum and maximum values of the emitted values. sumsqr is the sum over all squares of the emitted values (useful for statistics calculations).