ArangoDB v3.10 reached End of Life (EOL) and is no longer supported.

This documentation is outdated. Please see the most recent stable version.

HTTP interface for collections

The HTTP API for collections lets you create and delete collections, get information about collections, and modify certain properties of existing collections

Addresses of collections

All collections in ArangoDB have a unique identifier and a unique name. To access a collection, use the collection name to refer to it:

http://server:port/_api/collection/<collection-name>

For example, assume that the collection identifier is 7254820 and the collection name is demo, then the URL of that collection is:

http://localhost:8529/_api/collection/demo

Get information about collections

List all collections

GET /_api/collection

Returns an object with a result attribute containing an array with the descriptions of all collections in the current database.

By providing the optional excludeSystem query parameter with a value of true, all system collections are excluded from the response.

Path Parameters

Query Parameters

excludeSystem boolean
Whether or not system collections should be excluded from the result.

HTTP Headers

Responses

200 OK
The list of collections

Examples

Return information about all collections:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection

Show output

HTTP/1.1 200 OK
content-type: application/json
cache-control: no-cache, no-store, must-revalidate, pre-check=0, post-check=0, max-age=0, s-maxage=0
connection: Keep-Alive
content-length: 1646
content-security-policy: frame-ancestors 'self'; form-action 'self';
expires: 0
pragma: no-cache
server: ArangoDB
strict-transport-security: max-age=31536000 ; includeSubDomains
x-arango-queue-time-seconds: 0.000000
x-content-type-options: nosniff

{ 
  "error" : false, 
  "code" : 200, 
  "result" : [ 
    { 
      "id" : "31", 
      "name" : "_apps", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_apps" 
    }, 
    { 
      "id" : "113", 
      "name" : "demo", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : false, 
      "globallyUniqueId" : "h1617A8BD8EEE/113" 
    }, 
    { 
      "id" : "119", 
      "name" : "animals", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : false, 
      "globallyUniqueId" : "h1617A8BD8EEE/119" 
    }, 
    { 
      "id" : "10", 
      "name" : "_statistics", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_statistics" 
    }, 
    { 
      "id" : "4", 
      "name" : "_users", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_users" 
    }, 
    { 
      "id" : "22", 
      "name" : "_aqlfunctions", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_aqlfunctions" 
    }, 
    { 
      "id" : "16", 
      "name" : "_statisticsRaw", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_statisticsRaw" 
    }, 
    { 
      "id" : "34", 
      "name" : "_appbundles", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_appbundles" 
    }, 
    { 
      "id" : "108", 
      "name" : "ignore", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : false, 
      "globallyUniqueId" : "h1617A8BD8EEE/108" 
    }, 
    { 
      "id" : "28", 
      "name" : "_jobs", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_jobs" 
    }, 
    { 
      "id" : "67258", 
      "name" : "mycollection", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : false, 
      "globallyUniqueId" : "h1617A8BD8EEE/67258" 
    }, 
    { 
      "id" : "13", 
      "name" : "_statistics15", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_statistics15" 
    }, 
    { 
      "id" : "7", 
      "name" : "_graphs", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_graphs" 
    }, 
    { 
      "id" : "25", 
      "name" : "_queues", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_queues" 
    }, 
    { 
      "id" : "19", 
      "name" : "_analyzers", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_analyzers" 
    }, 
    { 
      "id" : "37", 
      "name" : "_frontend", 
      "status" : 3, 
      "type" : 2, 
      "isSystem" : true, 
      "globallyUniqueId" : "_frontend" 
    } 
  ] 
}

Get the collection information

GET /_api/collection/{collection-name}

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

The result is an object describing the collection with the following attributes:

id: The identifier of the collection.
name: The name of the collection.
status: The status of the collection as number.
- 3: loaded
- 5: deleted

Every other status indicates a corrupted collection.

type: The type of the collection as number.
- 2: document collection (normal case)
- 3: edge collection
isSystem: If true then the collection is a system collection.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Responses

404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Get the properties of a collection

GET /_api/collection/{collection-name}/properties

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Returns all properties of the specified collection.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Responses

200 OK
- cacheEnabled* boolean
  Whether the in-memory hash cache for documents is enabled for this collection.
- computedValues array of objects
  A list of objects, each representing a computed value.
  computeOn array of strings
  An array of strings that defines on which write operations the value is computed. The possible values are "insert", "update", and "replace".
  expression* string
  An AQL RETURN operation with an expression that computes the desired value.
  failOnWarning boolean
  Whether the write operation fails if the expression produces a warning.
  keepNull boolean
  Whether the target attribute is set if the expression evaluates to null.
  name* string
  The name of the target attribute.
  overwrite* boolean
  Whether the computed value takes precedence over a user-provided or existing attribute.
- distributeShardsLike string
  The name of another collection. This collection uses the replicationFactor, numberOfShards, shardingStrategy, and writeConcern properties of the other collection and the shards of this collection are distributed in the same way as the shards of the other collection.
- globallyUniqueId string
  A unique identifier of the collection. This is an internal property.
- id string
  A unique identifier of the collection (deprecated).
- isDisjoint boolean
  Whether the SmartGraph or EnterpriseGraph this collection belongs to is disjoint (Enterprise Edition only). This is an internal property. (cluster only)
- isSmart boolean
  Whether the collection is used in a SmartGraph or EnterpriseGraph (Enterprise Edition only). This is an internal property. (cluster only)
- isSystem boolean
  Whether the collection is a system collection. Collection names that starts with an underscore are usually system collections.
- keyOptions* object
  An object which contains key generation options.
  allowUserKeys* boolean
  If set to true, then you are allowed to supply own key values in the _key attribute of a document. If set to false, then the key generator is solely responsible for generating keys and an error is raised if you supply own key values in the _key attribute of documents.
  You should not use both user-specified and automatically generated document keys in the same collection in cluster deployments for collections with more than a single shard. Mixing the two can lead to conflicts because Coordinators that auto-generate keys in this case are not aware of all keys which are already used.
  increment integer
  The increment value for the autoincrement key generator. Not used for other key generator types.
  lastValue* integer
  The current offset value of the autoincrement or padded key generator. This is an internal property for restoring dumps properly.
  offset integer
  The initial offset value for the autoincrement key generator. Not used for other key generator types.
  type* string
  Specifies the type of the key generator. Possible values:
  "traditional"
  "autoincrement"
  "uuid"
  "padded"
- name string
  The name of this collection.
- numberOfShards integer
  The number of shards of the collection. (cluster only)
- replicationFactor integer
  Contains how many copies of each shard are kept on different DB-Servers. It is an integer number in the range of 1-10 or the string "satellite" for SatelliteCollections (Enterprise Edition only). (cluster only)
- schema object
  An object that specifies the collection-level schema for documents.
- shardKeys array of strings
  Contains the names of document attributes that are used to determine the target shard for documents. (cluster only)
- shardingStrategy string
  The sharding strategy selected for the collection. (cluster only)
  Possible values:
  "community-compat"
  "enterprise-compat"
  "enterprise-smart-edge-compat"
  "hash"
  "enterprise-hash-smart-edge"
  "enterprise-hex-smart-vertex"
- smartGraphAttribute string
  The attribute that is used for sharding: vertices with the same value of this attribute are placed in the same shard. All vertices are required to have this attribute set and it has to be a string. Edges derive the attribute from their connected vertices (Enterprise Edition only). (cluster only)
- smartJoinAttribute string
  Determines an attribute of the collection that must contain the shard key value of the referred-to SmartJoin collection (Enterprise Edition only). (cluster only)
- syncByRevision* boolean
  Whether the newer revision-based replication protocol is enabled for this collection. This is an internal property.
- type integer
  The type of the collection:
  0: “unknown”
  2: regular document collection
  3: edge collection
- waitForSync* boolean
  If true, creating, changing, or removing documents waits until the data has been synchronized to disk.
- writeConcern integer
  Determines how many copies of each shard are required to be in-sync on the different DB-Servers. If there are less than these many copies in the cluster, a shard refuses to write. Writes to shards with enough up-to-date copies succeed at the same time, however. The value of writeConcern cannot be greater than replicationFactor.
  If distributeShardsLike is set, the writeConcern is that of the prototype collection. For SatelliteCollections, the writeConcern is automatically controlled to equal the number of DB-Servers and has a value of 0. Otherwise, the default value is controlled by the current database’s default writeConcern, which uses the --cluster.write-concern startup option as default, which defaults to 1. (cluster only)
400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

Using an identifier:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/67601/properties

Show output

Using a name:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/properties

Show output

Get the document count of a collection

GET /_api/collection/{collection-name}/count

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Get the number of documents in a collection.

count: The number of documents stored in the specified collection.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

x-arango-trx-id string
To make this operation a part of a Stream Transaction, set this header to the transaction ID returned by the POST /_api/transaction/begin call.

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

Requesting the number of documents:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/count

Show output

Get the collection statistics

GET /_api/collection/{collection-name}/figures

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

In addition to the above, the result also contains the number of documents and additional statistical information about the collection.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

details boolean
Setting details to true will return extended storage engine-specific details to the figures. The details are intended for debugging ArangoDB itself and their format is subject to change. By default, details is set to false, so no details are returned and the behavior is identical to previous versions of ArangoDB. Please note that requesting details may cause additional load and thus have an impact on performance.

HTTP Headers

Responses

200 OK
Returns information about the collection:
- count* integer
  The number of documents currently present in the collection.
- figures* object
  The metrics of the collection.
  indexes* object
  The index metrics.
  count* integer
  The total number of indexes defined for the collection, including the pre-defined indexes (e.g. primary index).
  size* integer
  The total memory allocated for indexes in bytes.
400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

Using an identifier and requesting the figures of the collection:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/figures

Show output

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/figures?details=true

Show output

Get the responsible shard for a document

PUT /_api/collection/{collection-name}/responsibleShard

Returns the ID of the shard that is responsible for the given document (if the document exists) or that would be responsible if such document existed.

The request must body must contain a JSON document with at least the collection’s shard key attributes set to some values.

The response is a JSON object with a shardId attribute, which will contain the ID of the responsible shard.

This method is only available in cluster deployments on Coordinators.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Request Body application/json object

document* object
The request body must be a JSON object with at least the shard key attributes set to some values, but it may also be a full document.

Responses

200 OK
Returns the ID of the responsible shard.
400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned. Additionally, if not all of the collection’s shard key attributes are present in the input document, then a HTTP 400 is returned as well.
404 Not Found
If the collection-name is unknown, then an HTTP 404 is returned.
501 Not Implemented
HTTP 501 is returned if the method is called on a single server.

Examples

curl -X PUT --header 'accept: application/json' --data-binary @- --dump - 'http://localhost:8529/_api/collection/testCollection/responsibleShard' <<'EOF'
{
  "_key": "testkey",
  "value": 23
}
EOF

Show output

Get the shard IDs of a collection

GET /_api/collection/{collection-name}/shards

By default returns a JSON array with the shard IDs of the collection.

If the details parameter is set to true, it will return a JSON object with the shard IDs as object attribute keys, and the responsible servers for each shard mapped to them. In the detailed response, the leader shards will be first in the arrays.

This method is only available in cluster deployments on Coordinators.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

details boolean
If set to true, the return value will also contain the responsible servers for the collections’ shards.

HTTP Headers

Responses

200 OK
Returns the collection’s shards.
400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then an HTTP 404 is returned.
501 Not Implemented
HTTP 501 is returned if the method is called on a single server.

Examples

Retrieves the list of shards:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/testCollection/shards

Show output

Retrieves the list of shards with the responsible servers:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/testCollection/shards?details=true

Show output

Get the collection revision ID

GET /_api/collection/{collection-name}/revision

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

The response will contain the collection’s latest used revision id. The revision id is a server-generated string that clients can use to check whether data in a collection has changed since the last revision check.

revision: The collection revision id as a string.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

Retrieving the revision of a collection

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/revision

Show output

Get the collection checksum

GET /_api/collection/{collection-name}/checksum

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Will calculate a checksum of the meta-data (keys and optionally revision ids) and optionally the document data in the collection.

The checksum can be used to compare if two collections on different ArangoDB instances contain the same contents. The current revision of the collection is returned too so one can make sure the checksums are calculated for the same state of data.

By default, the checksum will only be calculated on the _key system attribute of the documents contained in the collection. For edge collections, the system attributes _from and _to will also be included in the calculation.

By setting the optional query parameter withRevisions to true, then revision ids (_rev system attributes) are included in the checksumming.

By providing the optional query parameter withData with a value of true, the user-defined document attributes will be included in the calculation too.

Including user-defined attributes will make the checksumming slower.

The response is a JSON object with the following attributes:

checksum: The calculated checksum as a number.
revision: The collection revision id as a string.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

withRevisions boolean
Whether or not to include document revision ids in the checksum calculation.
withData boolean
Whether or not to include document body data in the checksum calculation.

HTTP Headers

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

Retrieving the checksum of a collection:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/checksum

Show output

Retrieving the checksum of a collection including the collection data, but not the revisions:

curl --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/checksum?withRevisions=false&withData=true

Show output

Create and delete collections

Create a collection

POST /_api/collection

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Creates a new collection with a given name. The request must contain an object with the following attributes.

Path Parameters

Query Parameters

waitForSyncReplication boolean
The default is true, which means the server only reports success back to the client when all replicas have created the collection. Set it to false if you want faster server responses and don’t care about full replication.
enforceReplicationFactor boolean
The default is true, which means the server checks if there are enough replicas available at creation time and bail out otherwise. Set it to false to disable this extra check.

HTTP Headers

Request Body application/json object

cacheEnabled boolean
Whether the in-memory hash cache for documents should be enabled for this collection (default: false). Can be controlled globally with the --cache.size startup option. The cache can speed up repeated reads of the same documents via their document keys. If the same documents are not fetched often or are modified frequently, then you may disable the cache to avoid the maintenance costs.
computedValues array of objects
An optional list of objects, each representing a computed value.
- computeOn array of strings
  An array of strings to define on which write operations the value shall be computed. The possible values are "insert", "update", and "replace". The default is ["insert", "update", "replace"].
- expression* string
  An AQL RETURN operation with an expression that computes the desired value. See Computed Value Expressions for details.
- failOnWarning boolean
  Whether to let the write operation fail if the expression produces a warning. The default is false.
- keepNull boolean
  Whether the target attribute shall be set if the expression evaluates to null. You can set the option to false to not set (or unset) the target attribute if the expression returns null. The default is true.
- name* string
  The name of the target attribute. Can only be a top-level attribute, but you may return a nested object. Cannot be _key, _id, _rev, _from, _to, or a shard key attribute.
- overwrite* boolean
  Whether the computed value shall take precedence over a user-provided or existing attribute.
distributeShardsLike string
The name of another collection. If this property is set in a cluster, the collection copies the replicationFactor, numberOfShards, shardingStrategy, and writeConcern properties from the specified collection (referred to as the prototype collection) and distributes the shards of this collection in the same way as the shards of the other collection. In an Enterprise Edition cluster, this data co-location is utilized to optimize queries.
You need to use the same number of shardKeys as the prototype collection, but you can use different attributes.
The default is "".
Using this parameter has consequences for the prototype collection. It can no longer be dropped, before the sharding-imitating collections are dropped. Equally, backups and restores of imitating collections alone generate warnings (which can be overridden) about a missing sharding prototype.
isDisjoint boolean
Whether the collection is for a Disjoint SmartGraph (Enterprise Edition only). This is an internal property.
isSmart boolean
Whether the collection is for a SmartGraph or EnterpriseGraph (Enterprise Edition only). This is an internal property.
isSystem boolean
If true, create a system collection. In this case, the collection-name should start with an underscore. End-users should normally create non-system collections only. API implementors may be required to create system collections in very special occasions, but normally a regular collection will do. (The default is false)
keyOptions object
additional options for key generation. If specified, then keyOptions should be a JSON object containing the following attributes:
- allowUserKeys* boolean
  If set to true, then you are allowed to supply own key values in the _key attribute of documents. If set to false, then the key generator is solely be responsible for generating keys and an error is raised if you supply own key values in the _key attribute of documents.
  You should not use both user-specified and automatically generated document keys in the same collection in cluster deployments for collections with more than a single shard. Mixing the two can lead to conflicts because Coordinators that auto-generate keys in this case are not aware of all keys which are already used.
- increment* integer
  increment value for autoincrement key generator. Not used for other key generator types.
- offset* integer
  Initial offset value for autoincrement key generator. Not used for other key generator types.
- type* string
  specifies the type of the key generator. The currently available generators are traditional, autoincrement, uuid and padded.
  The traditional key generator generates numerical keys in ascending order. The sequence of keys is not guaranteed to be gap-free.
  The autoincrement key generator generates numerical keys in ascending order, the initial offset and the spacing can be configured (note: autoincrement is currently only supported for non-sharded collections). The sequence of generated keys is not guaranteed to be gap-free, because a new key will be generated on every document insert attempt, not just for successful inserts.
  The padded key generator generates keys of a fixed length (16 bytes) in ascending lexicographical sort order. This is ideal for the RocksDB storage engine, which will slightly benefit keys that are inserted in lexicographically ascending order. The key generator can be used in a single-server or cluster. The sequence of generated keys is not guaranteed to be gap-free.
  The uuid key generator generates universally unique 128 bit keys, which are stored in hexadecimal human-readable format. This key generator can be used in a single-server or cluster to generate “seemingly random” keys. The keys produced by this key generator are not lexicographically sorted.
  Please note that keys are only guaranteed to be truly ascending in single server deployments and for collections that only have a single shard (that includes collections in a OneShard database). The reason is that for collections with more than a single shard, document keys are generated on Coordinator(s). For collections with a single shard, the document keys are generated on the leader DB-Server, which has full control over the key sequence.
name* string
The name of the collection.
numberOfShards integer
(The default is 1): in a cluster, this value determines the number of shards to create for the collection.
replicationFactor integer
(The default is 1): in a cluster, this attribute determines how many copies of each shard are kept on different DB-Servers. The value 1 means that only one copy (no synchronous replication) is kept. A value of k means that k-1 replicas are kept. For SatelliteCollections, it needs to be the string "satellite", which matches the replication factor to the number of DB-Servers (Enterprise Edition only).
Any two copies reside on different DB-Servers. Replication between them is synchronous, that is, every write operation to the “leader” copy will be replicated to all “follower” replicas, before the write operation is reported successful.
If a server fails, this is detected automatically and one of the servers holding copies take over, usually without an error being reported.
schema object
Optional object that specifies the collection level schema for documents. The attribute keys rule, level and message must follow the rules documented in Document Schema Validation
shardKeys string
(The default is [ "_key" ]): in a cluster, this attribute determines which document attributes are used to determine the target shard for documents. Documents are sent to shards based on the values of their shard key attributes. The values of all shard key attributes in a document are hashed, and the hash value is used to determine the target shard.
Values of shard key attributes cannot be changed once set.
shardingStrategy string
This attribute specifies the name of the sharding strategy to use for the collection. There are different sharding strategies to select from when creating a new collection. The selected shardingStrategy value remains fixed for the collection and cannot be changed afterwards. This is important to make the collection keep its sharding settings and always find documents already distributed to shards using the same initial sharding algorithm.
The available sharding strategies are:
- community-compat: default sharding used by ArangoDB Community Edition before version 3.4
- enterprise-compat: default sharding used by ArangoDB Enterprise Edition before version 3.4
- enterprise-smart-edge-compat: default sharding used by smart edge collections in ArangoDB Enterprise Edition before version 3.4
- hash: default sharding used for new collections starting from version 3.4 (excluding smart edge collections)
- enterprise-hash-smart-edge: default sharding used for new smart edge collections starting from version 3.4
- enterprise-hex-smart-vertex: sharding used for vertex collections of EnterpriseGraphs
If no sharding strategy is specified, the default is hash for all normal collections, enterprise-hash-smart-edge for all smart edge collections, and enterprise-hex-smart-vertex for EnterpriseGraph vertex collections (the latter two require the Enterprise Edition of ArangoDB). Manually overriding the sharding strategy does not yet provide a benefit, but it may later in case other sharding strategies are added.
smartGraphAttribute string
The attribute that is used for sharding: vertices with the same value of this attribute are placed in the same shard. All vertices are required to have this attribute set and it has to be a string. Edges derive the attribute from their connected vertices.
This feature can only be used in the Enterprise Edition.
smartJoinAttribute string
In an Enterprise Edition cluster, this attribute determines an attribute of the collection that must contain the shard key value of the referred-to SmartJoin collection. Additionally, the shard key for a document in this collection must contain the value of this attribute, followed by a colon, followed by the actual primary key of the document.
This feature can only be used in the Enterprise Edition and requires the distributeShardsLike attribute of the collection to be set to the name of another collection. It also requires the shardKeys attribute of the collection to be set to a single shard key attribute, with an additional ‘:’ at the end. A further restriction is that whenever documents are stored or updated in the collection, the value stored in the smartJoinAttribute must be a string.
type integer
(The default is 2): the type of the collection to create. The following values for type are valid:
- 2: document collection
- 3: edge collection
waitForSync boolean
If true then the data is synchronized to disk before returning from a document create, update, replace or removal operation. (Default: false)
writeConcern integer
Write concern for this collection (default: 1). It determines how many copies of each shard are required to be in sync on the different DB-Servers. If there are less than these many copies in the cluster, a shard refuses to write. Writes to shards with enough up-to-date copies succeed at the same time, however. The value of writeConcern cannot be greater than replicationFactor.
If distributeShardsLike is set, the writeConcern is that of the prototype collection. For SatelliteCollections, the writeConcern is automatically controlled to equal the number of DB-Servers and has a value of 0. Otherwise, the default value is controlled by the current database’s default writeConcern, which uses the --cluster.write-concern startup option as default, which defaults to 1. (cluster only)

Responses

200 OK
- cacheEnabled* boolean
  Whether the in-memory hash cache for documents is enabled for this collection.
- computedValues array of objects
  A list of objects, each representing a computed value.
  computeOn array of strings
  An array of strings that defines on which write operations the value is computed. The possible values are "insert", "update", and "replace".
  expression* string
  An AQL RETURN operation with an expression that computes the desired value.
  failOnWarning boolean
  Whether the write operation fails if the expression produces a warning.
  keepNull boolean
  Whether the target attribute is set if the expression evaluates to null.
  name* string
  The name of the target attribute.
  overwrite* boolean
  Whether the computed value takes precedence over a user-provided or existing attribute.
- distributeShardsLike string
  The name of another collection. This collection uses the replicationFactor, numberOfShards, shardingStrategy, and writeConcern properties of the other collection and the shards of this collection are distributed in the same way as the shards of the other collection.
- globallyUniqueId string
  A unique identifier of the collection. This is an internal property.
- id string
  A unique identifier of the collection (deprecated).
- isDisjoint boolean
  Whether the SmartGraph or EnterpriseGraph this collection belongs to is disjoint (Enterprise Edition only). This is an internal property. (cluster only)
- isSmart boolean
  Whether the collection is used in a SmartGraph or EnterpriseGraph (Enterprise Edition only). This is an internal property. (cluster only)
- isSystem boolean
  Whether the collection is a system collection. Collection names that starts with an underscore are usually system collections.
- keyOptions* object
  An object which contains key generation options.
  allowUserKeys* boolean
  If set to true, then you are allowed to supply own key values in the _key attribute of a document. If set to false, then the key generator is solely responsible for generating keys and an error is raised if you supply own key values in the _key attribute of documents.
  You should not use both user-specified and automatically generated document keys in the same collection in cluster deployments for collections with more than a single shard. Mixing the two can lead to conflicts because Coordinators that auto-generate keys in this case are not aware of all keys which are already used.
  increment integer
  The increment value for the autoincrement key generator. Not used for other key generator types.
  lastValue* integer
  The current offset value of the autoincrement or padded key generator. This is an internal property for restoring dumps properly.
  offset integer
  The initial offset value for the autoincrement key generator. Not used for other key generator types.
  type* string
  Specifies the type of the key generator. Possible values:
  "traditional"
  "autoincrement"
  "uuid"
  "padded"
- name string
  The name of this collection.
- numberOfShards integer
  The number of shards of the collection. (cluster only)
- replicationFactor integer
  Contains how many copies of each shard are kept on different DB-Servers. It is an integer number in the range of 1-10 or the string "satellite" for SatelliteCollections (Enterprise Edition only). (cluster only)
- schema object
  An object that specifies the collection-level schema for documents.
- shardKeys array of strings
  Contains the names of document attributes that are used to determine the target shard for documents. (cluster only)
- shardingStrategy string
  The sharding strategy selected for the collection. (cluster only)
  Possible values:
  "community-compat"
  "enterprise-compat"
  "enterprise-smart-edge-compat"
  "hash"
  "enterprise-hash-smart-edge"
  "enterprise-hex-smart-vertex"
- smartGraphAttribute string
  The attribute that is used for sharding: vertices with the same value of this attribute are placed in the same shard. All vertices are required to have this attribute set and it has to be a string. Edges derive the attribute from their connected vertices (Enterprise Edition only). (cluster only)
- smartJoinAttribute string
  Determines an attribute of the collection that must contain the shard key value of the referred-to SmartJoin collection (Enterprise Edition only). (cluster only)
- syncByRevision* boolean
  Whether the newer revision-based replication protocol is enabled for this collection. This is an internal property.
- type integer
  The type of the collection:
  0: “unknown”
  2: regular document collection
  3: edge collection
- waitForSync* boolean
  If true, creating, changing, or removing documents waits until the data has been synchronized to disk.
- writeConcern integer
  Determines how many copies of each shard are required to be in-sync on the different DB-Servers. If there are less than these many copies in the cluster, a shard refuses to write. Writes to shards with enough up-to-date copies succeed at the same time, however. The value of writeConcern cannot be greater than replicationFactor.
  If distributeShardsLike is set, the writeConcern is that of the prototype collection. For SatelliteCollections, the writeConcern is automatically controlled to equal the number of DB-Servers and has a value of 0. Otherwise, the default value is controlled by the current database’s default writeConcern, which uses the --cluster.write-concern startup option as default, which defaults to 1. (cluster only)
400 Bad Request
If the collection-name is missing, then an HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then an HTTP 404 is returned.

Examples

curl -X POST --header 'accept: application/json' --data-binary @- --dump - http://localhost:8529/_api/collection
{
  "name": "testCollectionBasics"
}

curl -X POST --header 'accept: application/json' --data-binary @- --dump - http://localhost:8529/_api/collection
{
  "name": "testCollectionEdges",
  "type": 3
}

Show output

HTTP/1.1 200 OK
content-type: application/json
cache-control: no-cache, no-store, must-revalidate, pre-check=0, post-check=0, max-age=0, s-maxage=0
connection: Keep-Alive
content-length: 458
content-security-policy: frame-ancestors 'self'; form-action 'self';
expires: 0
pragma: no-cache
server: ArangoDB
strict-transport-security: max-age=31536000 ; includeSubDomains
x-arango-queue-time-seconds: 0.000000
x-content-type-options: nosniff

{ 
  "error" : false, 
  "code" : 200, 
  "writeConcern" : 1, 
  "waitForSync" : false, 
  "usesRevisionsAsDocumentIds" : true, 
  "syncByRevision" : true, 
  "cacheEnabled" : false, 
  "isSystem" : false, 
  "internalValidatorType" : 0, 
  "globallyUniqueId" : "h1617A8BD8EEE/67865", 
  "id" : "67865", 
  "isSmartChild" : false, 
  "schema" : null, 
  "keyOptions" : { 
    "allowUserKeys" : true, 
    "type" : "traditional", 
    "lastValue" : 0 
  }, 
  "name" : "testCollectionBasics", 
  "computedValues" : null, 
  "objectId" : "67864", 
  "type" : 2, 
  "status" : 3, 
  "statusString" : "loaded" 
}


HTTP/1.1 200 OK
content-type: application/json
cache-control: no-cache, no-store, must-revalidate, pre-check=0, post-check=0, max-age=0, s-maxage=0
connection: Keep-Alive
content-length: 457
content-security-policy: frame-ancestors 'self'; form-action 'self';
expires: 0
pragma: no-cache
server: ArangoDB
strict-transport-security: max-age=31536000 ; includeSubDomains
x-arango-queue-time-seconds: 0.000000
x-content-type-options: nosniff

{ 
  "error" : false, 
  "code" : 200, 
  "writeConcern" : 1, 
  "waitForSync" : false, 
  "usesRevisionsAsDocumentIds" : true, 
  "syncByRevision" : true, 
  "cacheEnabled" : false, 
  "isSystem" : false, 
  "internalValidatorType" : 0, 
  "globallyUniqueId" : "h1617A8BD8EEE/67870", 
  "id" : "67870", 
  "isSmartChild" : false, 
  "schema" : null, 
  "keyOptions" : { 
    "allowUserKeys" : true, 
    "type" : "traditional", 
    "lastValue" : 0 
  }, 
  "name" : "testCollectionEdges", 
  "computedValues" : null, 
  "objectId" : "67869", 
  "type" : 3, 
  "status" : 3, 
  "statusString" : "loaded" 
}

curl -X POST --header 'accept: application/json' --data-binary @- --dump - http://localhost:8529/_api/collection
{
  "name": "testCollectionUsers",
  "keyOptions": {
    "type": "autoincrement",
    "increment": 5,
    "allowUserKeys": true
  }
}

Show output

Drop a collection

DELETE /_api/collection/{collection-name}

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Drops the collection identified by collection-name.

If the collection was successfully dropped, an object is returned with the following attributes:

error: false
id: The identifier of the dropped collection.

Path Parameters

collection-name* string
The name of the collection to drop.

Query Parameters

isSystem boolean
Whether or not the collection to drop is a system collection. This parameter must be set to true in order to drop a system collection.

HTTP Headers

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

Using an identifier:

curl -X DELETE --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/67885

Show output

Using a name:

curl -X DELETE --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products1

Show output

Dropping a system collection

curl -X DELETE --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/_example?isSystem=true

Show output

Truncate a collection

PUT /_api/collection/{collection-name}/truncate

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Removes all documents from the collection, but leaves the indexes intact.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

waitForSync boolean
If true then the data is synchronized to disk before returning from the truncate operation (default: false)
compact boolean
If true (default) then the storage engine is told to start a compaction in order to free up disk space. This can be resource intensive. If the only intention is to start over with an empty collection, specify false.

HTTP Headers

x-arango-trx-id string
To make this operation a part of a Stream Transaction, set this header to the transaction ID returned by the POST /_api/transaction/begin call.

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

curl -X PUT --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/truncate

Show output

Modify collections

Load a collection

PUT /_api/collection/{collection-name}/load

The load function is deprecated from version 3.8.0 onwards and is a no-op from version 3.9.0 onwards. It should no longer be used, as it may be removed in a future version of ArangoDB.

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Since ArangoDB version 3.9.0 this API does nothing. Previously it used to load a collection into memory.

The request body object might optionally contain the following attribute:

count: If set, this controls whether the return value should include the number of documents in the collection. Setting count to false may speed up loading a collection. The default value for count is true.

A call to this API returns an object with the following attributes for compatibility reasons:

id: The identifier of the collection.
name: The name of the collection.
count: The number of documents inside the collection. This is only returned if the count input parameters is set to true or has not been specified.
status: The status of the collection as number.
type: The collection type. Valid types are:
- 2: document collection
- 3: edge collection
isSystem: If true then the collection is a system collection.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

curl -X PUT --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/load

Show output

Unload a collection

PUT /_api/collection/{collection-name}/unload

The unload function is deprecated from version 3.8.0 onwards and is a no-op from version 3.9.0 onwards. It should no longer be used, as it may be removed in a future version of ArangoDB.

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Since ArangoDB version 3.9.0 this API does nothing. Previously it used to unload a collection from memory, while preserving all documents. When calling the API an object with the following attributes is returned for compatibility reasons:

id: The identifier of the collection.
name: The name of the collection.
status: The status of the collection as number.
type: The collection type. Valid types are:
- 2: document collection
- 3: edges collection
isSystem: If true then the collection is a system collection.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

curl -X PUT --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/unload

Show output

Load collection indexes into memory

PUT /_api/collection/{collection-name}/loadIndexesIntoMemory

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

You can call this endpoint to try to cache this collection’s index entries in the main memory. Index lookups served from the memory cache can be much faster than lookups not stored in the cache, resulting in a performance boost.

The endpoint iterates over suitable indexes of the collection and stores the indexed values (not the entire document data) in memory. This is implemented for edge indexes only.

The endpoint returns as soon as the index warmup has been scheduled. The index warmup may still be ongoing in the background, even after the return value has already been sent. As all suitable indexes are scanned, it may cause significant I/O activity and background load.

This feature honors memory limits. If the indexes you want to load are smaller than your memory limit, this feature guarantees that most index values are cached. If the index is greater than your memory limit, this feature fills up values up to this limit. You cannot control which indexes of the collection should have priority over others.

It is guaranteed that the in-memory cache data is consistent with the stored index data at all times.

On success, this endpoint returns an object with attribute result set to true.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Responses

200 OK
If the index loading has been scheduled for all suitable indexes.
400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

curl -X PUT --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/products/loadIndexesIntoMemory

Show output

Change the properties of a collection

PUT /_api/collection/{collection-name}/properties

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Changes the properties of a collection. Only the provided attributes are updated. Collection properties cannot be changed once a collection is created except for the listed properties, as well as the collection name via the rename endpoint (but not in clusters).

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Request Body application/json object

cacheEnabled boolean
Whether the in-memory hash cache for documents should be enabled for this collection (default: false). Can be controlled globally with the --cache.size startup option. The cache can speed up repeated reads of the same documents via their document keys. If the same documents are not fetched often or are modified frequently, then you may disable the cache to avoid the maintenance costs.
computedValues array of objects
An optional list of objects, each representing a computed value.
- computeOn array of strings
  An array of strings to define on which write operations the value shall be computed. The possible values are "insert", "update", and "replace". The default is ["insert", "update", "replace"].
- expression* string
  An AQL RETURN operation with an expression that computes the desired value. See Computed Value Expressions for details.
- failOnWarning boolean
  Whether to let the write operation fail if the expression produces a warning. The default is false.
- keepNull boolean
  Whether the target attribute shall be set if the expression evaluates to null. You can set the option to false to not set (or unset) the target attribute if the expression returns null. The default is true.
- name* string
  The name of the target attribute. Can only be a top-level attribute, but you may return a nested object. Cannot be _key, _id, _rev, _from, _to, or a shard key attribute.
- overwrite* boolean
  Whether the computed value shall take precedence over a user-provided or existing attribute.
replicationFactor integer
(The default is 1): in a cluster, this attribute determines how many copies of each shard are kept on different DB-Servers. The value 1 means that only one copy (no synchronous replication) is kept. A value of k means that k-1 replicas are kept. For SatelliteCollections, it needs to be the string "satellite", which matches the replication factor to the number of DB-Servers (Enterprise Edition only).
Any two copies reside on different DB-Servers. Replication between them is synchronous, that is, every write operation to the “leader” copy will be replicated to all “follower” replicas, before the write operation is reported successful.
If a server fails, this is detected automatically and one of the servers holding copies take over, usually without an error being reported.
schema object
Optional object that specifies the collection level schema for documents. The attribute keys rule, level and message must follow the rules documented in Document Schema Validation
waitForSync boolean
If true then the data is synchronized to disk before returning from a document create, update, replace or removal operation. (default: false)
writeConcern integer
Write concern for this collection (default: 1). It determines how many copies of each shard are required to be in sync on the different DB-Servers. If there are less than these many copies in the cluster, a shard refuses to write. Writes to shards with enough up-to-date copies succeed at the same time, however. The value of writeConcern cannot be greater than replicationFactor.
If distributeShardsLike is set, the writeConcern is that of the prototype collection. For SatelliteCollections, the writeConcern is automatically controlled to equal the number of DB-Servers and has a value of 0. Otherwise, the default value is controlled by the current database’s default writeConcern, which uses the --cluster.write-concern startup option as default, which defaults to 1. (cluster only)

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

curl -X PUT --header 'accept: application/json' --data-binary @- --dump - http://localhost:8529/_api/collection/products/properties
{
  "waitForSync": true
}

Show output

Rename a collection

PUT /_api/collection/{collection-name}/rename

Accessing collections by their numeric ID is deprecated from version 3.4.0 on. You should reference them via their names instead.

Renames a collection. Expects an object with the attribute(s)

name: The new name.

It returns an object with the attributes

id: The identifier of the collection.
name: The new name of the collection.
status: The status of the collection as number.
type: The collection type. Valid types are:
- 2: document collection
- 3: edges collection
isSystem: If true then the collection is a system collection.

If renaming the collection succeeds, then the collection is also renamed in all graph definitions inside the _graphs collection in the current database.

Renaming collections is not supported in cluster deployments.

Path Parameters

collection-name* string
The name of the collection to rename.

Query Parameters

HTTP Headers

Responses

400 Bad Request
If the collection-name is missing, then a HTTP 400 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Examples

curl -X PUT --header 'accept: application/json' --data-binary @- --dump - http://localhost:8529/_api/collection/products1/rename
{
  "name": "newname"
}

Show output

Recalculate the document count of a collection

PUT /_api/collection/{collection-name}/recalculateCount

Recalculates the document count of a collection, if it ever becomes inconsistent.

It returns an object with the attributes

result: will be true if recalculating the document count succeeded.

Path Parameters

collection-name* string
The name of the collection.

Query Parameters

HTTP Headers

Responses

200 OK
If the document count was recalculated successfully, HTTP 200 is returned.
404 Not Found
If the collection-name is unknown, then a HTTP 404 is returned.

Compact a collection

PUT /_api/collection/{collection-name}/compact

Compacts the data of a collection in order to reclaim disk space. The operation will compact the document and index data by rewriting the underlying .sst files and only keeping the relevant entries.

Under normal circumstances, running a compact operation is not necessary, as the collection data will eventually get compacted anyway. However, in some situations, e.g. after running lots of update/replace or remove operations, the disk data for a collection may contain a lot of outdated data for which the space shall be reclaimed. In this case the compaction operation can be used.

Path Parameters

collection-name* string
Name of the collection to compact

Query Parameters

HTTP Headers

Responses

200 OK
Compaction started successfully
401 Unauthorized
if the request was not authenticated as a user with sufficient rights

Examples

curl -X PUT --header 'accept: application/json' --dump - http://localhost:8529/_api/collection/testCollection/compact

Show output