Facettes et requêtes composées
Les apprenants combineront plusieurs clauses de recherche avec l’opérateur composé et calculeront des nombres par facette pour les filtres de catégorie, en parallèle des résultats de recherche.
Facettes et requêtes composées est une leçon MongoDB Academy gratuite sur CoddyKit. Ceci est la leçon 4 sur 4. Tu peux lire la leçon complète ci-dessous gratuitement — puis la pratiquer en direct dans le navigateur avec un éditeur de code intégré et un tuteur IA 24/7. Elle fait partie du parcours d'apprentissage MongoDB Academy, et ta progression se synchronise sur le web et l'application CoddyKit. Le cours MongoDB Academy comprend 4 leçons au total.
Certaines parties de cette leçon n'ont pas encore été traduites et s'affichent en anglais.
What Are Search Facets?
Facets are aggregated counts of how many search results fall into each category of a field. You have seen facets on e-commerce sites: 'Brand: Nike (142), Adidas (89), Puma (45)' or 'Price: Under $50 (230), $50-$100 (185)'. In Atlas Search, facets are computed server-side alongside the search results in a single query using the $searchMeta stage or the facet collector within $search.
Configuring Fields for Faceting
To facet on a field, it must be indexed with the stringFacet (for string categories) or numberFacet/dateFacet (for range buckets) data type in the Atlas Search index mapping. You can add a facet type alongside a string type on the same field. Standard string-typed fields are not facetable—you must explicitly add the facet type to your index definition.
// Index mapping with facetable fields
{
'mappings': {
'dynamic': false,
'fields': {
'title': { 'type': 'string', 'analyzer': 'lucene.standard' },
'category': [
{ 'type': 'string' }, // for filtering
{ 'type': 'stringFacet' } // for facet counts
],
'brand': { 'type': 'stringFacet' },
'price': [
{ 'type': 'number' },
{ 'type': 'numberFacet' }
]
}
}
}The $searchMeta Stage for Facet Counts
Use $searchMeta (instead of $search) when you need only the metadata (facet counts, total results) without returning the actual documents. This is efficient for re-computing facet counts when a filter changes. The stage returns a single document containing the facet buckets. Combine it with a separate $search query to get both documents and facets in two parallel requests.
// Get facet counts for categories and brands
db.products.aggregate([
{
$searchMeta: {
index: 'default',
facet: {
operator: {
text: { query: 'wireless headphones', path: 'title' }
},
facets: {
categoriesFacet: {
type: 'string',
path: 'category',
numBuckets: 10 // return up to 10 category buckets
},
brandsFacet: {
type: 'string',
path: 'brand',
numBuckets: 20
}
}
}
}
}
])Number Range Facets
For numeric fields, define explicit price range buckets using the numericFacet type with a boundaries array and a default bucket name for values that fall outside all boundaries. The boundaries define bin edges—[0, 50, 100, 500] creates buckets 0–50, 50–100, and 100–500. Each bucket in the result shows the count of matching documents in that range.
db.products.aggregate([
{
$searchMeta: {
facet: {
operator: { text: { query: 'running shoes', path: 'name' } },
facets: {
priceRanges: {
type: 'number',
path: 'price',
boundaries: [0, 50, 100, 200, 500],
// Produces: '$0-$50', '$50-$100', '$100-$200', '$200-$500'
default: 'Other' // for products outside all ranges
}
}
}
}
}
])The compound Operator: Combining Search Clauses
The compound operator is the most powerful Atlas Search operator—it lets you combine multiple search clauses using four clause types: must (all must match, contributes to score), mustNot (must not match), should (optional, but boosts score if they match), and filter (must match but does not affect score). This mirrors Elasticsearch's bool query and enables sophisticated relevance ranking.
db.jobs.aggregate([
{
$search: {
compound: {
must: [
// Required: title must contain 'senior engineer'
{ text: { query: 'senior engineer', path: 'title' } }
],
should: [
// Optional boost: prefer remote jobs
{ equals: { path: 'remote', value: true } },
// Optional boost: prefer jobs with high salary
{ range: { path: 'salary', gte: 150000 } }
],
filter: [
// Must match but does NOT affect score
{ equals: { path: 'active', value: true } }
],
mustNot: [
// Exclude contract roles
{ equals: { path: 'type', value: 'contract' } }
]
}
}
}
])minimumShouldMatch: Controlling OR Logic
By default, should clauses are entirely optional—a document that matches none of them still appears in results if must clauses match. Set minimumShouldMatch to require that at least N should clauses match. This converts the behavior from pure OR to a 'match at least N' semantic, letting you express queries like 'must have title match AND at least one of: location, skills, or salary match'.
db.candidates.aggregate([
{
$search: {
compound: {
must: [
{ text: { query: 'python developer', path: 'title' } }
],
should: [
{ text: { query: 'machine learning', path: 'skills' } },
{ text: { query: 'tensorflow pytorch', path: 'skills' } },
{ range: { path: 'experienceYears', gte: 3 } }
],
minimumShouldMatch: 1 // must match at least 1 should clause
}
}
}
])Score Boosting in compound Queries
You can control relevance score contributions from each clause using the score option. Boosting multiplies the base score from that clause by a factor—useful when a match in the title is more important than a match in the description. constant scoring replaces the calculated score with a fixed value, useful for filter-like clauses you want to affect ranking but not dominate it.
db.articles.aggregate([
{
$search: {
compound: {
should: [
{
text: {
query: 'mongodb performance',
path: 'title',
score: { boost: { value: 3.0 } } // title matches worth 3x more
}
},
{
text: {
query: 'mongodb performance',
path: 'body',
score: { boost: { value: 1.0 } } // body matches at normal weight
}
}
]
}
}
}
])Getting Documents and Facets Together
A common pattern in search UIs is to return both the result documents and the facet counts in parallel. The most efficient approach uses the Atlas Search facet collector: pair $searchMeta for facet metadata with a separate $search query for documents, running both in parallel from your application. Alternatively, use $search with a $facet aggregation stage after it (this performs two passes but is simpler).
// Run both queries in parallel
const searchQuery = 'wireless headphones';
const [documents, facets] = await Promise.all([
// Query 1: get matching documents
db.collection('products').aggregate([
{ $search: { text: { query: searchQuery, path: 'name' } } },
{ $project: { name: 1, price: 1, category: 1, score: { $meta: 'searchScore' } } },
{ $limit: 20 }
]).toArray(),
// Query 2: get facet counts for the same query
db.collection('products').aggregate([
{ $searchMeta: { facet: { operator: { text: { query: searchQuery, path: 'name' } },
facets: { cats: { type: 'string', path: 'category', numBuckets: 10 } } } } }
]).toArray()
]);
console.log('Results:', documents.length, 'Facets:', facets[0].facet.cats.buckets);Applying Active Facet Filters
When a user clicks a facet to filter results (e.g., 'Category: Electronics'), you need to add that selection as a filter clause in the compound operator. This narrows both the result documents and the remaining facet counts to only matching items. It is important to apply facet filters in the filter (not must) clause so they do not affect relevance scoring.
async function searchWithFacetFilter(query, selectedCategory) {
const filterClauses = [];
if (selectedCategory) {
filterClauses.push({ equals: { path: 'category', value: selectedCategory } });
}
return db.collection('products').aggregate([
{
$search: {
compound: {
must: [{ text: { query, path: 'name' } }],
filter: filterClauses // apply active facet selections
}
}
},
{ $project: { name: 1, price: 1, category: 1 } },
{ $limit: 20 }
]).toArray();
}Facet Result Structure
The result of a $searchMeta with facets is a single document containing a facet key. Each facet you defined is a sub-key containing a buckets array. Each bucket has an _id (the facet value) and a count. For number facets, each bucket also has a lowerBound and upperBound. Parse this structure in your application to render the facet filter sidebar.
// $searchMeta result structure:
// [
// {
// 'facet': {
// 'categoriesFacet': {
// 'buckets': [
// { '_id': 'Electronics', 'count': 142 },
// { '_id': 'Audio', 'count': 89 },
// { '_id': 'Accessories', 'count': 45 }
// ]
// },
// 'priceRanges': {
// 'buckets': [
// { '_id': '0.0', 'count': 23 }, // 0 to 50
// { '_id': '50.0', 'count': 67 }, // 50 to 100
// { '_id': '100.0', 'count': 101 } // 100 to 200
// ]
// }
// }
// }
// ]Date Facets for Time-Based Navigation
Date facets work similarly to number facets—you define explicit bucket boundaries as Date objects. This enables time-based navigation like 'Published this week (12)', 'Published this month (47)', 'Published this year (183)'. Define the boundaries as an array of dates and Atlas Search counts documents whose date field falls within each range. Date facets are useful for filtering blog posts, job listings, events, and any time-sensitive content.
db.articles.aggregate([
{
$searchMeta: {
facet: {
operator: { text: { query: 'mongodb', path: 'title' } },
facets: {
publishedDate: {
type: 'date',
path: 'publishedAt',
boundaries: [
new Date('2024-01-01'),
new Date('2024-04-01'),
new Date('2024-07-01'),
new Date('2024-10-01'),
new Date('2025-01-01')
],
default: 'Other'
}
}
}
}
}
])Quick Check
Test your understanding of MongoDB & NoSQL Databases concepts from this lesson.
Lesson Recap
In this lesson you learned: facets require stringFacet/numberFacet types in the index mapping and are computed with $searchMeta, the compound operator combines must, mustNot, should, and filter clauses for sophisticated relevance queries, and filter clauses narrow results without affecting the relevance score, while must clauses do both. Next up we explore connecting to MongoDB with the official Node.js driver.
Questions Fréquemment Posées
La leçon « Facettes et requêtes composées » est-elle gratuite ?
Oui — le texte complet de « Facettes et requêtes composées » est gratuit à lire ici sur le web. Pour la pratiquer de manière interactive (un éditeur de code intégré et un tuteur IA 24/7) et déverrouiller le reste du cours MongoDB Academy, passe à CoddyKit PRO. Le cours MongoDB Academy comprend 4 leçons au total.
Qu'est-ce que j'apprendrai dans « Facettes et requêtes composées » ?
Les apprenants combineront plusieurs clauses de recherche avec l’opérateur composé et calculeront des nombres par facette pour les filtres de catégorie, en parallèle des résultats de recherche. Tu pratiques MongoDB Academy avec du code pratique que tu exécutes directement dans le navigateur, et un tuteur IA 24/7 répond à tes questions au fur et à mesure que tu avances dans la leçon.
Dois-je avoir de l'expérience pour commencer MongoDB Academy ?
Aucune expérience préalable n'est requise. MongoDB Academy sur CoddyKit est structuré pour les débutants jusqu'aux apprenants avancés, donc tu peux commencer ici ou depuis le début et avancer à ton rythme. Ceci est la leçon 4 sur 4.
Combien de temps prend la leçon « Facettes et requêtes composées » ?
La plupart des leçons CoddyKit prennent environ 5–10 minutes. Chacune est courte et interactive, tu progresses régulièrement et tu repiques exactement où tu t'es arrêté sur le web et l'app.
Peux-tu écrire et exécuter du code dans cette leçon MongoDB Academy ?
Oui. Chaque leçon MongoDB Academy inclut un éditeur de code intégré, tu écris et exécutes du vrai code directement dans ton navigateur et tu reçois des retours IA instantanés — aucune configuration locale requise.
Toutes les leçons de ce cours
- Créer un index Atlas Search
- Écrire des requêtes $search : texte, expression et générique
- Saisie semi-automatique et correspondance approximative
- Facettes et requêtes composées