Code Summary: Using MongoDB Aggregation Stages with PHP: Match and Group

Review the following code, which demonstrates how to use the $match and $group stages in a MongoDB aggregation pipeline by using the MongoDB PHP library.

Using $match

Use the $match stage to select documents that match the specified query condition(s) and pass the matching documents to the next stage. $match takes a document that specifies the query.

$match should be placed early in a pipeline to reduce the number of documents that will be processed later in the pipeline.

Here's an example of the $match stage:

// Select accounts with balances of less than $1000


{ "$match": 
  { "balance": 
    { "$lt": 1000 }
  }
}

Using $group

Use the $group stage to separate documents into groups. The $group stage must have an _id field that specifies the group key. The group key is preceded by a $ and enclosed in quotation marks.

A $group stage can include additional fields that are computed by using accumulator operators, such as $avg.

Here's an example of the $group stage:

// Separate documents by account type and calculate the average balance for 
// each account type
{ "$group": 
  { "_id": "$account_type", "avg_balance": { "$avg": "$balance" } }
}

Aggregation Example That Uses $match and $group

The following is an example of an aggregation pipeline that uses $match and $group in PHP.

<?php

require_once __DIR__ . '/vendor/autoload.php';

$client = new MongoDB\Client('<Atlas-connection-string>');

$accountsCollection = $client->bank->accounts;

$pipeline = [
  ['$match' => ['balance' => ['$lt' => 1000]]],
  ['$group' => [
    '_id' => '$account_type',
    'avg_balance' => ['$avg' => '$balance'],
  ]],
];

$cursor = $accountsCollection->aggregate($pipeline);

foreach ($cursor as $result){
  var_dump($result);
}