Workspace API 2.0

The Workspace API allows you to list, import, export, and delete notebooks and folders. The maximum allowed size of a request to the Workspace API is 10MB. See Cluster log delivery examples for a how to guide on this API.

Important

To access Databricks REST APIs, you must authenticate.

Delete

Endpoint	HTTP Method
`2.0/workspace/delete`	`POST`

Delete an object or a directory (and optionally recursively deletes all objects in the directory). If path does not exist, this call returns an error RESOURCE_DOES_NOT_EXIST. If path is a non-empty directory and recursive is set to false, this call returns an error DIRECTORY_NOT_EMPTY. Object deletion cannot be undone and deleting a directory recursively is not atomic.

Example

Request:

curl --netrc --request POST \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/delete \
  --header 'Accept: application/json' \
  --data '{ "path": "/Users/me@example.com/MyFolder", "recursive": true }'

If successful, this endpoint returns no response.

Request structure

Field Name	Type	Description
path	`STRING`	The absolute path of the notebook or directory. This field is required.
recursive	`BOOL`	The flag that specifies whether to delete the object recursively. It is `false` by default. Please note this deleting directory is not atomic. If it fails in the middle, some of objects under this directory may be deleted and cannot be undone.

Export

Endpoint	HTTP Method
`2.0/workspace/export`	`GET`

Export a notebook or contents of an entire directory. You can also export a Databricks Repo, or a notebook or directory from a Databricks Repo. If path does not exist, this call returns an error RESOURCE_DOES_NOT_EXIST. You can export a directory only in DBC format. If the exported data exceeds the size limit, this call returns an error MAX_NOTEBOOK_SIZE_EXCEEDED or BAD_REQUEST. This API does not support exporting a library.

Examples

Export a notebook

Request:

curl --netrc --request GET \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/export \
  --header 'Accept: application/json' \
  --data '{ "path": "/Users/me@example.com/MyFolder/MyNotebook", "format": "SOURCE", "direct_download": true }'

curl --netrc --request GET \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/export \
  --header 'Accept: application/json' \
  --data '{ "path": "/Repos/me@example.com/MyFolder/MyNotebook", "format": "SOURCE", "direct_download": true }'

Response:

If the direct_download field was set to false or was omitted from the request, a base64-encoded version of the content is returned, for example:

{
  "content": "Ly8gRGF0YWJyaWNrcyBub3RlYm9vayBzb3VyY2UKMSsx",
}

Otherwise, if direct_download was set to true in the request, the content is downloaded.

Export a workspace file

This functionality is in Public Preview.

Request:

curl --netrc --request GET \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/export \
  --header 'Accept: application/json' \
  --data '{ "path": "/Users/me@example.com/MyRepo/my_file.txt", "format": "AUTO", “direct_download”: true }'

Response:

If the direct_download field was set to false or was omitted from the request, a base64-encoded version of the content is returned, for example:

{
  "content": "Ly8gRGF0YWJyaWNrcyBub3RlYm9vayBzb3VyY2UKMSsx",
}

Otherwise, if direct_download was set to true in the request, the content is downloaded.

Request structure

Field Name	Type	Description
path	`STRING`	The absolute path of the notebook or directory. Exporting a directory is supported only for `DBC`. This field is required.
format	ExportFormat	This specifies the format of the exported file. By default, this is `SOURCE`. The value is case sensitive.
direct_download	`BOOL`	Flag to enable direct download. If it is `true`, the response will be the exported file itself. Otherwise, the response contains content as base64 encoded string. See Export a notebook or folder for more information about how to use it.

Response structure

Field Name	Type	Description
content	`BYTES`	The base64-encoded content. If the limit (10MB) is exceeded, exception with error code `MAX_NOTEBOOK_SIZE_EXCEEDED` is thrown.

Get status

Endpoint	HTTP Method
`2.0/workspace/get-status`	`GET`

Gets the status of an object or a directory. If path does not exist, this call returns an error RESOURCE_DOES_NOT_EXIST.

Example

Request:

curl --netrc --request GET \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/get-status \
  --header 'Accept: application/json' \
  --data '{ "path": "/Users/me@example.com/MyFolder/MyNotebook" }'

Response:

{
  "object_type": "NOTEBOOK",
  "path": "/Users/me@example.com/MyFolder/MyNotebook",
  "language": "PYTHON",
  "object_id": 123456789012345
}

Request structure

Field Name	Type	Description
path	`STRING`	The absolute path of the notebook or directory. This field is required.

Response structure

Field Name	Type	Description
object_type	ObjectType	The type of the object.
object_id	`INT64`	Unique identifier for the object.
path	`STRING`	The absolute path of the object.
language	Language	The language of the object. This value is set only if the object type is `NOTEBOOK`.

Import

Endpoint	HTTP Method
`2.0/workspace/import`	`POST`

Import a notebook, workspace file, or the contents of an entire directory. If path already exists and overwrite is set to false, this call returns an error RESOURCE_ALREADY_EXISTS. You can use only DBC format to import a directory.

Example

Import a base64-encoded string:

curl --netrc --request POST \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/import \
  --header 'Accept: application/json' \
  --data '{ "path": "/Users/me@example.com/MyFolder/MyNotebook", "content": "Ly8gRGF0YWJyaWNrcyBub3RlYm9vayBzb3VyY2UKMSsx", "language": "PYTHON", "overwrite": true, "format": "SOURCE" }'

Import a local notebook:

curl --netrc --request POST \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/import \
  --header 'Content-Type: multipart/form-data' \
  --form path=/Users/me@example.com/MyFolder/MyNotebook \
  --form content=@myCode.py.zip

Import a local file (Public Preview):

curl --netrc --request POST \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/import \
  --header 'Content-Type: multipart/form-data' \
  --form path=/Repos/me@example.com/MyRepo/my_file.py \
  --form format=AUTO \
  --form content=@non-notebook.py

If successful, this endpoint returns no response.

Request structure

Field Name	Type	Description
path	`STRING`	The absolute path of the notebook or directory. Importing directory is only support for `DBC` format. This field is required.
format	ExportFormat	This specifies the format of the file to be imported. By default, this is `SOURCE`. The value is case sensitive.
language	Language	The language. If format is set to `SOURCE`, this field is required; otherwise, it will be ignored.
content	`BYTES`	The base64-encoded content. This has a limit of 10 MB. If the limit (10MB) is exceeded, exception with error code `MAX_NOTEBOOK_SIZE_EXCEEDED` is thrown. This parameter might be absent, and instead a posted file will be used. See Import a notebook or directory for more information about how to use it.
overwrite	`BOOL`	The flag that specifies whether to overwrite existing object. It is `false` by default. For `DBC` format, overwrite is not supported since it may contain a directory.

List

Endpoint	HTTP Method
`2.0/workspace/list`	`GET`

List the contents of a directory, or the object if it is not a directory. If the input path does not exist, this call returns an error RESOURCE_DOES_NOT_EXIST.

Example

List directories and their contents:

Request:

curl --netrc --request GET \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/list \
  --header 'Accept: application/json' \
  --data '{ "path": "/Users/me@example.com" }'

Response:

{
  "objects": [
    {
      "path": "/Users/me@example.com/MyFolder",
      "object_type": "DIRECTORY",
      "object_id": 234567890123456
    },
    {
      "path": "/Users/me@example.com/MyFolder/MyNotebook",
      "object_type": "NOTEBOOK",
      "language": "PYTHON",
      "object_id": 123456789012345
    },
    {
      "..."
    }
  ]
}

List repos:

curl --netrc --request GET \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/list \
  --header 'Accept: application/json' \
  --data '{ "path": "/Repos/me@example.com" }'

Response:

{
  "objects": [
    {
      "path": "/Repos/me@example.com/MyRepo1",
      "object_type": "REPO",
      "object_id": 234567890123456
    },
    {
      "path": "/Repos/me@example.com/MyRepo2",
      "object_type": "REPO",
      "object_id": 123456789012345
    },
    {
      "..."
    }
  ]
}

Request structure

Field Name	Type	Description
path	`STRING`	The absolute path of the notebook or directory. This field is required.

Response structure

Field Name	Type	Description
objects	An array of ObjectInfo	List of objects.

Mkdirs

Endpoint	HTTP Method
`2.0/workspace/mkdirs`	`POST`

Create the given directory and necessary parent directories if they do not exists. If there exists an object (not a directory) at any prefix of the input path, this call returns an error RESOURCE_ALREADY_EXISTS. If this operation fails it may have succeeded in creating some of the necessary parent directories.

Example

Request:

curl --netrc --request POST \
  https://dbc-a1b2345c-d6e7.cloud.databricks.com/api/2.0/workspace/mkdirs \
  --header 'Accept: application/json' \
  --data '{ "path": "/Users/me@example.com/MyFolder" }'

If successful, this endpoint returns no response.

Request structure

Field Name	Type	Description
path	`STRING`	The absolute path of the directory. If the parent directories do not exist, it will also create them. If the directory already exists, this command will do nothing and succeed. This field is required.

Data structures

In this section:

ObjectInfo
ExportFormat
Language
ObjectType

ObjectInfo

The information of the object in workspace. It is returned by list and get-status.

Field Name	Type	Description
object_type	ObjectType	The type of the object.
object_id	`INT64`	Unique identifier for the object.
path	`STRING`	The absolute path of the object.
language	Language	The language of the object. This value is set only if the object type is `NOTEBOOK`.

ExportFormat

The format for notebook import and export.

Format	Description
AUTO	(Public Preview) The item is imported/exported as either a workspace file or a notebook, depending on an analysis of the item’s extension and the header content provided in the request. See also Import a file and convert it to a notebook. If the item is imported as a notebook, then the item’s extension is automatically removed.
SOURCE	The notebook will be imported/exported as source code.
HTML	The notebook will be imported/exported as an HTML file.
JUPYTER	The notebook will be imported/exported as a Jupyter/IPython Notebook file.
DBC	The notebook will be imported/exported as Databricks archive format.

Language

The language of notebook.

Language	Description
SCALA	Scala notebook.
PYTHON	Python notebook.
SQL	SQL notebook.
R	R notebook.

ObjectType

The type of the object in workspace.

Type	Description
NOTEBOOK	Notebook
DIRECTORY	Directory
FILE	File
LIBRARY	Library
REPO	Repository