$elemMatch: การจับคู่เอกสารย่อยในอาร์เรย์

It's common in MongoDB to store arrays of embedded sub-documents—objects with multiple fields—inside a parent document. Examples include orders containing line items, users with multiple addresses, or students with per-subject scores. Querying these structures requires care to avoid the spread field problem where conditions are matched across different array elements.

When you filter an array of sub-documents using dot-notation fields directly, MongoDB applies each condition independently to any element in the array. The query { 'scores.subject': 'math', 'scores.grade': 'A' } would match a document if any element has subject='math' AND any (possibly different) element has grade='A'. This false-positive behavior is the spread field problem.

$elemMatch is the solution: it constrains all conditions to match on the same single array element. MongoDB only returns a document if at least one element in the array satisfies every condition inside the $elemMatch block simultaneously. This is the correct way to query arrays of sub-documents with multiple conditions.

You can use any MongoDB query operator inside $elemMatch, including range operators like $gt, $lte, and $in. This lets you express conditions like 'find any element where score is between 80 and 100 AND the subject is math'—conditions that must be true for one specific element.

You can negate an $elemMatch condition using $not to find documents where no array element satisfies all the conditions. For example, 'find students who do NOT have a math A' means 'no element satisfies both subject=math AND grade=A'. This is more precise than checking { 'scores.grade': { $ne: 'A' } } which would exclude students with any A-grade subject.

$elemMatch can also be used in the projection (second argument to find()) to return only the first array element that matches a condition. When used in projection, it's called the $elemMatch projection operator (same name, different context). It returns at most one matching element per document.

There are two ways to project a single matching array element: the $elemMatch projection (in the projection object) lets you specify a different filter than the query filter, while the positional $ operator returns the first element matched by the query filter itself. Use $elemMatch in projection when the query filter and the element you want to project are different.

MongoDB supports querying arrays of arrays and deeply nested sub-documents using chained dot notation. However, $elemMatch only applies at one level deep at a time. For queries on arrays nested inside arrays, you need to chain multiple $elemMatch operators or restructure your schema to avoid excessive nesting.

A multikey index on the array field supports $elemMatch queries. MongoDB uses the index to narrow down candidate documents by the indexed field values, then applies the full $elemMatch condition to confirm each candidate. To maximise index efficiency, include the most selective field of your $elemMatch condition in the index.

You can use $exists and $type inside $elemMatch to find array elements that have optional fields or match a specific BSON type. This is useful for heterogeneous arrays where not all elements share the same shape—common in legacy data migrations or flexible event log schemas.

A practical use of $elemMatch is in e-commerce: finding orders that contain a line item for a specific product with a quantity above a threshold. Without $elemMatch, the conditions would spread across different line items and produce false positives.

In this lesson you learned: $elemMatch in queries requires all conditions to match a single array element, solving the spread field problem, $elemMatch in projection returns only the first matching element, and multikey indexes support $elemMatch queries efficiently. Next up we tackle array update operators.