Repository navigation
Search Indexes
A collection definition can also manage a collection's Atlas Search and Vector Search indexes, alongside its regular indexes.
These need MongoDB Atlas, or a local Atlas deployment (mongodb/mongodb-atlas-local). A plain server doesn't run the
search process, so up stops with an error if a definition declares one there.
[ManagedCollection("products")]
public sealed class Products : ICollectionDefinition<BsonDocument>
{
public void Define(CollectionBuilder<BsonDocument> collection)
{
collection.SearchIndexes
.Add("by_text", """{ "mappings": { "dynamic": true } }""")
.AddVector("by_embedding", """
{ "fields": [ { "type": "vector", "path": "embedding", "numDimensions": 1536, "similarity": "cosine" } ] }
""")
.Ignore("adhoc_*");
}
}| Method | Declares |
|---|---|
Add(name, definition) |
An Atlas Search index. |
AddVector(name, definition) |
A Vector Search index. |
Ignore(pattern) |
Leaves matching names alone (* matches anything). |
The definition is the index's JSON definition document, as the Atlas API takes it: a raw JSON string, a
BsonDocument, or an object. A search index is identified by its name, unique per collection.
| Change | What up does |
|---|---|
| Declared, not on the server | Creates it. |
| Definition changed | Updates it in place with updateSearchIndex, so it stays queryable while it rebuilds. |
type changed (search ↔ vectorSearch) |
Drops it and creates it again: updateSearchIndex can't change the type. |
| On the server, not declared | Depends on searchIndexes.prune: report (default) leaves it and lists it as keep, drop removes it. |
A search index needs its collection to exist, so a definition that declares one on a collection that doesn't exist yet creates the collection first.
The server fills a definition in with defaults: dynamic: false and an empty fields on a mapping, each field's
default options for its type (e.g. a string field's store, norms and indexOptions, a number field's
representation), and it leaves out ones such as storedSource: false. Only those default values are allowed for, so an
unchanged definition stays in sync whether or not it spells them out. Any other difference is a change, e.g. a field or
an analyzer removed. The types a field is indexed as can be listed in any order.
-
searchIndexes.pruneisreportby default.collection.SearchIndexes.Prune = SearchIndexPrune.Dropoverrides it for one collection. - A declared search index never changes an ignored one: if the server's has the same name but a different type or definition, the collection is blocked until one of them changes. One that's exactly the same is fine, and one that isn't on the server yet is created.
- Search indexes are only managed when a definition declares at least one. A collection that declares none never has its search indexes listed or pruned, so you can adopt them gradually.
Building a search index is asynchronous: by default up issues the change and returns without waiting. Set
searchIndexes.waitForReady: true to wait for each created or updated
index to become queryable, and each dropped one to be gone, up to searchIndexes.waitTimeout seconds, before the run
returns. That's useful in a test or a deploy step that seeds data straight after. It also waits for a declared index
that's already in line but still building, e.g. after an earlier run that didn't wait or timed out.
docker run -d -p 27017:27017 mongodb/mongodb-atlas-local
The Getting Started tutorial uses this image, so search indexes work out of the box.
Kevsoft.MigrateMongo ·
NuGet ·
Releases ·
Issues ·
Found a mistake? These pages live in
wiki/ on main; pull requests welcome.
Get started
Migrations
- Writing Migrations
- Transactions
- Recording Results
- Dependency Injection
- Custom Templates
- Adopting an Existing Database
Collections
- Managing Collections
- Indexes
- Search Indexes
- Collection Options
- Validators
- Views
- Undeclared Collections
Deploying
Connecting
Reference
Concepts