MongoDB

Available Procedures

signature
apoc.mongo.aggregate(uri, pipeline, $config) yield value - perform an aggregate operation on mongodb collection
apoc.mongo.count(uri, query, $config) yield value - perform a count operation on mongodb collection
apoc.mongo.find(uri, query, $config) yield value - perform a find operation on mongodb collection
apoc.mongo.delete(uri, query, $config) - delete the given documents from the mongodb collection and returns the number of affected documents
apoc.mongo.insert(uri, documents, $config) yield value - inserts the given documents into the mongodb collection
apoc.mongo.update(uri, query, update, $config) - updates the given documents from the mongodb collection and returns the number of affected documents

signature

apoc.mongo.aggregate(uri, pipeline, $config) yield value - perform an aggregate operation on mongodb collection

apoc.mongo.count(uri, query, $config) yield value - perform a count operation on mongodb collection

apoc.mongo.find(uri, query, $config) yield value - perform a find operation on mongodb collection

apoc.mongo.delete(uri, query, $config) - delete the given documents from the mongodb collection and returns the number of affected documents

apoc.mongo.insert(uri, documents, $config) yield value - inserts the given documents into the mongodb collection

apoc.mongo.update(uri, query, update, $config) - updates the given documents from the mongodb collection and returns the number of affected documents

Install Dependencies

The Mongo procedures have dependencies on a client library that is not included in the APOC Library.

This dependency is included in apoc-mongodb-dependencies-4.3.0.12.jar, which can be downloaded from the releases page. Once that file is downloaded, it should be placed in the plugins directory and the Neo4j Server restarted.

Alternatively, you could copy these jars into the plugins directory:

bson-3.4.2.jar
mongo-java-driver-3.4.2.jar,
mongodb-driver-3.4.2.jar
mongodb-driver-core-3.4.2.jar

You should be able to get them from the following links:

Field description

uri: The connection String URI, with scheme mongodb://[username:password@]host1[:port1][,host2[:port2],…[,hostN[:portN]]]/databaseName.collectionName[?options]. Note that this uri must necessarily have the database name and (if the collection config parameter is not explicit) the collection name (for example mongodb://user:pass@localhost:27017/myDb.myCollection?authSource=admin)
query: query parameter map (can be a map or a json string)
update: update parameter map (only for apoc.mongo.update)
documents: the documents to insert (only for apoc.mongo.insert)
config: see below

Configuration parameters

The procedures support the following config parameters:

Table 1. Config parameters
name	type	default	description
extractReferences	`Boolean`	false	If true and a field contains an `ObjectId` it will include the related document instead of the `ObjectId`
objectIdAsMap	`Boolean`	true	If true extract the `ObjectId` as map
project	`Map<K,V>` OR `String`	empty	The projection parameters (can be a map or a json string)
sort	`Map<K,V>` OR `String`	empty	The sort parameters (can be a map or a json string)
skip	`Long`	0	The number of documents to skip
limit	`Long`	0	The max number of documents to show
collection	`String`	empty	The collection name (takes precedence over the collection passed with `uri` parameter

Examples

Given the following collections:

// Product
...
{"_id": ObjectId("product1"), "name": "Product 1", "price": 100}
{"_id": ObjectId("product3"), "name": "Product 2", "price": 200}
{"_id": ObjectId("product3"), "name": "Product 3", "price": 300}
...

// Person
...
{"_id": ObjectId("personAl"), "name": "Al", expr: BsonRegularExpression("foo*"), "bought": [ObjectId("product1"), ObjectId("product3")]}
{"_id": ObjectId("personJohn"), "name": "John", "age": 40, "foo", "bar"}
{"_id": ObjectId("personJack"), "name": "Jack", "age": 50, "foo", "bar", expr: BsonRegularExpression("bar*"), "bought": [ObjectId("product1"), ObjectId("product2")]}
...

we can run the following procedures.

`apoc.mongo.aggregate`

CALL apoc.mongodb.aggregate('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', [{`$match`: {foo: 'bar'}}, {`$set`: {aggrField: 'Y'} }])

Table 2. Results
value
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "John", "foo": "bar", "age": 40L, "aggrField": "Y", }`
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "Jack", "age": 50L, "foo": "bar", "expr": "bar*", "bought": ["product1", "product2"], "aggrField": "Y", }`

`apoc.mongo.count`

CALL apoc.mongodb.count('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin')

Table 3. Results
value
3

We can also pass the collection name through the config parameter:

CALL apoc.mongodb.count('mongodb://user:pass@localhost:27017/myDb?authSource=admin', {collection: 'Person'})

Table 4. Results
value
3

`apoc.mongo.find`

If we want to extract the all `Person`s with default parameter:

CALL apoc.mongodb.find('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin')

Table 5. Results
value
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "Al", "expr": "foo*", "bought": ["product1", "product3"] }`
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "John", "foo": "bar", "age": 40L }`
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "Jack", "age": 50L, "foo": "bar", "expr": "bar*", "bought": ["product1", "product2"] }`

In addition, we can pass the query param like:

CALL apoc.mongodb.first('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', {expr: {`$regex`: 'bar*', `$options`: ''}})

Table 6. Results
value
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "Jack", "foo": "bar", "expr": "bar*", "bought": ["product1", "product2"] }`

If we want to extract bought references, through config parameter:

CALL apoc.mongodb.first('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', {expr: {`$regex`: 'foo*', `$options`: ''}}, {extractReferences: true})

Table 7. Results
value
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "Al", "expr": "foo*", "bought": [ { "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "Product 1", "price": 100 }, { "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "name": "Product 3", "price": 300 }, ] }`

Moreover, we can retrieve the ObjectId s with theirs HexString representation through objectIdAsMap config:

CALL apoc.mongodb.first('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', {expr: {`$regex`: 'foo*', `$options`: ''}}, {objectIdAsMap: false, extractReferences: true})

Table 8. Results
value
`{ "_id": "personAl", "name": "Al", "expr": "foo*", "bought": [ {"_id": "product1", "name": "Product 1", "price": 100}, {"_id": "product3", "name": "Product 3", "price": 300} ] }`

Furthermore, we can skip n values and pass a project parameter:

CALL apoc.mongo.first('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', null, {skip: 2, project: {age: 1}})

Table 9. Results
value
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "age": 50L, }`

We can pass query, skip and sort parameter as stringified values, for example:

CALL apoc.mongo.first('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', '{foo: "bar"}', {sort: '{name: -1}', project: '{age: 1}'})

Table 10. Results
value
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "age": 40L, }`
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "age": 50L, }`

Furthermore, we can use the limit parameter, for example:

CALL apoc.mongodb.find('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', null, {skip: 1, limit: 1, project: {age: 1}})

Table 11. Results
value
`{ "_id": { "timestamp": <…>, "machineIdentifier": <…>, "processIdentifier": <…>, "counter": <…>, }, "age": 40, }`

Furthermore, we can pass a sort parameter, for example:

CALL apoc.mongodb.find('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', null, {sort: {name: -1}, objectIdAsMap: false, project: {name: 1}})

Table 12. Results
value
`` { "_id": "personJohn", "name": "John", }
`` { "_id": "personJack", "name": "Jack", }
`{ "_id": "personAl", "name": "Al", }`

`apoc.mongo.update`

To update the age property of the John document:

CALL apoc.mongodb.update('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', {name: "John"}, {`$set`: {age:99}})

with the number of row affected as result:

Table 13. Results
value
1

`apoc.mongo.delete`

To delete the John document:

CALL apoc.mongodb.update('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', {name: "John"})

with the number of row affected as result:

Table 14. Results
value
1

`apoc.mongo.insert`

To insert 2 document {"secondId": ObjectId("507f191e811c19729de860ea"), "baz": 1} and {"secondId": ObjectId("507f191e821c19729de860ef"), "baz": 1} in a Person collection (in this case the procedure return void):

CALL apoc.mongo.insert('mongodb://user:pass@localhost:27017/myDb.Person?authSource=admin', [{secondId: {`$oid`: '507f191e811c19729de860ea'}, baz: 1}, {secondId: {`$oid`: '507f191e821c19729de860ef'}, baz: 1}])