$push accumulator for building arrays in MongoDB - Step-by-Step Execution

Concept Flow - $push accumulator for building arrays

Start aggregation

↓

Group documents by key

↓

For each group

↓

$push adds each value to array

↓

Build array of values

↓

Output grouped document with array

The $push accumulator collects values from grouped documents into an array for each group.

Execution Sample

MongoDB

db.sales.aggregate([
  { $group: {
      _id: "$item",
      allPrices: { $push: "$price" }
  }}
])

Groups sales by item and collects all prices into an array for each item.

Execution Table

Step	Input Document	Group Key (_id)	Value to Push	Array State	Action
1	{"item": "apple", "price": 10}	apple	10	[10]	Start new group 'apple', push 10
2	{"item": "banana", "price": 5}	banana	5	[5]	Start new group 'banana', push 5
3	{"item": "apple", "price": 15}	apple	15	[10, 15]	Push 15 to 'apple' array
4	{"item": "banana", "price": 7}	banana	7	[5, 7]	Push 7 to 'banana' array
5	{"item": "apple", "price": 8}	apple	8	[10, 15, 8]	Push 8 to 'apple' array
6	No more documents	-	-	-	Aggregation ends, output groups with arrays

💡 All documents processed, arrays built for each group.

Variable Tracker

Variable	Start	After 1	After 2	After 3	After 4	After 5	Final
apple array	[]	[10]	[10]	[10, 15]	[10, 15]	[10, 15, 8]	[10, 15, 8]
banana array	[]	[]	[5]	[5]	[5, 7]	[5, 7]	[5, 7]

Key Moments - 3 Insights

Why does $push create an array even if there is only one document in the group?

$push always collects values into an array for consistency, so even one value is inside an array as shown in step 1 and 2 of the execution_table.

What happens if the group key (_id) changes between documents?

A new group is started for each unique _id, so separate arrays are built independently as seen for 'apple' and 'banana' groups in the execution_table.

Can $push add duplicate values to the array?

Yes, $push adds every value it sees without removing duplicates, so arrays may contain repeated values if input documents have them.

Visual Quiz - 3 Questions

Look at the execution_table, what is the 'apple array' after step 3?

A [15]

B [10]

C [10, 15]

D [10, 15, 8]

💡 Hint

Check the 'Array State' column for step 3 in the execution_table.

At which step does the 'banana array' first get a value pushed?

A Step 1

B Step 2

C Step 3

D Step 4

💡 Hint

Look at the 'Group Key (_id)' and 'Array State' columns for 'banana' in the execution_table.

If a new document {"item": "apple", "price": 20} is added after step 5, what will be the final 'apple array'?

A [10, 15, 8, 20]

B [10, 15, 8]

C [20]

D [10, 15]

💡 Hint

Consider how $push adds each new value to the existing array as shown in the execution_table.

Concept Snapshot

$push accumulator in MongoDB aggregation
- Used inside $group stage
- Collects values from documents into an array
- Creates array even if one value
- Arrays can have duplicates
- Useful to gather all values per group

Full Transcript

The $push accumulator in MongoDB aggregation collects values from grouped documents into arrays. When grouping documents by a key, $push adds each value to an array for that group. For example, grouping sales by item and pushing prices builds arrays of prices per item. Each input document is processed: if the group key is new, a new array starts; if existing, the value is appended. This continues until all documents are processed, resulting in grouped documents with arrays of values. $push always creates arrays, even if only one value exists, and it includes duplicates if present. This behavior is shown step-by-step in the execution table and variable tracker, helping beginners understand how arrays build up during aggregation.