Other administration tasks#

class dataikuapi.dss.admin.DSSGeneralSettings(client)#

The general settings of the DSS instance.

Important

Do not instantiate directly, use dataikuapi.DSSClient.get_general_settings() instead.

save()#

Save the changes that were made to the settings on the DSS instance

Note

This call requires an API key with admin rights

get_raw()#

Get the settings as a dictionary

Returns:

the settings

Return type:

dict

add_impersonation_rule(rule, is_user_rule=True)#

Add a rule to the impersonation settings

Parameters:
get_impersonation_rules(dss_user=None, dss_group=None, unix_user=None, hadoop_user=None, project_key=None, scope=None, rule_type=None, is_user=None, rule_from=None)#

Retrieve the user or group impersonation rules that match the parameters

Parameters:

  • dss_user (string) – a DSS user name

  • dss_group (string) – a DSS group name

  • rule_from (string) – a regex (which will be applied to user or group names)

  • unix_user (string) – a name to match the target UNIX user

  • hadoop_user (string) – a name to match the target Hadoop user

  • project_key (string) – a project key

  • scope (string) – project-scoped (‘PROJECT’) or global (‘GLOBAL’)

  • type (string) – the rule user or group matching method (‘IDENTITY’, ‘SINGLE_MAPPING’, ‘REGEXP_RULE’)

  • is_user (boolean) – True if only user-level rules should be considered, False for only group-level rules, None to consider both

remove_impersonation_rules(dss_user=None, dss_group=None, unix_user=None, hadoop_user=None, project_key=None, scope=None, rule_type=None, is_user=None, rule_from=None)#

Remove the user or group impersonation rules that matches the parameters from the settings

Parameters:

  • dss_user (string) – a DSS user name

  • dss_group (string) – a DSS group name

  • rule_from (string) – a regex (which will be applied to user or group names)

  • unix_user (string) – a name to match the target UNIX user

  • hadoop_user (string) – a name to match the target Hadoop user

  • project_key (string) – a project key

  • scope (string) – project-scoped (‘PROJECT’) or global (‘GLOBAL’)

  • type (string) – the rule user or group matching method (‘IDENTITY’, ‘SINGLE_MAPPING’, ‘REGEXP_RULE’)

  • is_user (boolean) – True if only user-level rules should be considered, False for only group-level rules, None to consider both

push_container_exec_base_images()#

Push the container exec base images to their repository

class dataikuapi.dss.admin.DSSUserImpersonationRule(raw=None)#

An user-level rule items for the impersonation settings

scope_global()#

Make the rule apply to all projects

scope_project(project_key)#

Make the rule apply to a given project

Parameters:

project_key (string) – the project this rule applies to

user_identity()#

Make the rule map each DSS user to a UNIX user of the same name

user_single(dss_user, unix_user, hadoop_user=None)#

Make the rule map a given DSS user to a given UNIX user

Parameters:

  • dss_user (string) – a DSS user

  • unix_user (string) – a UNIX user

  • hadoop_user (string) – a hadoop user (optional, defaults to unix_user)

user_regexp(regexp, unix_user, hadoop_user=None)#

Make the rule map a DSS users matching a given regular expression to a given UNIX user

Parameters:

  • regexp (string) – a regular expression to match DSS user names

  • unix_user (string) – a UNIX user

  • hadoop_user (string) – a hadoop user (optional, defaults to unix_user)

class dataikuapi.dss.admin.DSSGroupImpersonationRule(raw=None)#

A group-level rule items for the impersonation settings

group_identity()#

Make the rule map each DSS user to a UNIX user of the same name

group_single(dss_group, unix_user, hadoop_user=None)#

Make the rule map a given DSS user to a given UNIX user

Parameters:

  • dss_group (string) – a DSS group

  • unix_user (string) – a UNIX user

  • hadoop_user (string) – a hadoop user (optional, defaults to unix_user)

group_regexp(regexp, unix_user, hadoop_user=None)#

Make the rule map a DSS users matching a given regular expression to a given UNIX user

Parameters:

  • regexp (string) – a regular expression to match DSS groups

  • unix_user (string) – a UNIX user

  • hadoop_user (string) – a hadoop user (optional, defaults to unix_user)

class dataikuapi.dss.admin.DSSInstanceVariables(client, variables)#

Dict containing the instance variables.

The variables can be modified directly in the dict and persisted using its save() method.

Important

Do not instantiate directly, use dataikuapi.DSSClient.get_global_variables() instead.

save()#

Save the changes made to the instance variables.

Note

This call requires an API key with admin rights.

class dataikuapi.dss.future.DSSFuture(client, job_id, state=None, result_wrapper=<function DSSFuture.<lambda>>)#

A future represents a long-running task on a DSS instance. It allows you to track the state of the task, retrieve its result when it is ready or abort it.

Usage example:

# In this example, 'folder' is a DSSManagedFolder
future = folder.compute_metrics()

# Waits until the metrics are computed
metrics = future.wait_for_result()

Note

This class does not need to be instantiated directly. A dataikuapi.dss.future.DSSFuture is usually returned by the API calls that are initiating long running-tasks.

static from_resp(client, resp, result_wrapper=<function DSSFuture.<lambda>>)#

Creates a dataikuapi.dss.future.DSSFuture from the response of an endpoint that initiated a long-running task.

Parameters:

  • client (dataikuapi.dssclient.DSSClient) – An api client to connect to the DSS backend

  • resp (dict) – The response of the API call that initiated a long-running task.

  • result_wrapper (callable) – (optional) a function to apply to the result of the future, before it is returned by get_result or wait_for_result methods.

Returns:

dataikuapi.dss.future.DSSFuture

abort()#

Aborts the long-running task.

get_state()#

Queries the state of the future, and fetches the result if it’s ready.

Returns:

the state of the future, including the result if it is ready.

Return type:

dict

peek_state()#

Queries the state of the future, without fetching the result.

Returns:

the state of the future

Return type:

dict

get_result()#

Returns the future’s result if it’s ready.

Note

if a custom result_wrapper was provided, it will be applied on the result that will be returned.

Returns:

the result of the future

Return type:

object

Raises:

Exception – if the result is not ready (i.e. the task is still running, or it has failed)

has_result()#

Checks whether the task is completed successfully and has a result.

Returns:

True if the task has successfully returned a result, False otherwise

Return type:

bool

wait_for_result()#

Waits for the completion of the long-running task, and returns its result.

Note

if a custom result_wrapper was provided, it will be applied on the result that will be returned.

Returns:

the result of the future

Return type:

object

Raises:

Exception – if the task failed

class dataikuapi.dss.jupyternotebook.DSSJupyterNotebook(client, project_key, notebook_name)#

A handle on a Python/R/scala notebook.

Important

Do not instantiate directly, use dataikuapi.dss.project.DSSProject.get_jupyter_notebook() or dataikuapi.dss.project.DSSProject.create_jupyter_notebook().

unload(session_id=None)#

Stop this Jupyter notebook and release its resources.

get_sessions(as_objects=False)#

Get the list of running sessions of this Jupyter notebook.

Usage example:

# list the running notebooks in a project
for notebook in project.list_jupyter_notebooks(as_type="object"):
    if len(notebook.get_sessions()) > 0:
        print("Notebook %s is running" % notebook.notebook_name)
Parameters:

as_objects (boolean) – if True, each returned item will be a dataikuapi.dss.jupyternotebook.DSSNotebookSession

Returns:

a list of dataikuapi.dss.jupyternotebook.DSSNotebookSession if as_objects is True, a list of dict otherwise. The dict holds information about the kernels currently running the notebook, notably a sessionId for the Jupyter session id, and a kernelId for the Jupyter kernel id.

Return type:

list

get_content()#

Get the full contents of this Jupyter notebook.

The content comprises metadata, cells, notebook format info.

Usage example:

# collect all the source code of a notebook
lines = []
for cell in notebook.get_content().get_cells():
    if cell["cell_type"] != 'code':
        continue
    lines = lines + cell["source"]
print('\n'.join(lines))
Returns:

a list of dataikuapi.dss.jupyternotebook.DSSNotebookContent

Return type:

list

delete()#

Delete this Jupyter notebook and stop all of its active sessions.

get_object_discussions()#

Get a handle to manage discussions on the notebook.

Returns:

the handle to manage discussions

Return type:

dataikuapi.dss.discussion.DSSObjectDiscussions

class dataikuapi.dss.jupyternotebook.DSSJupyterNotebookListItem(client, data)#

An item in a list of Jupyter notebooks.

Important

Do not instantiate directly, use dataikuapi.dss.project.DSSProject.list_jupyter_notebooks()

to_notebook()#

Get a handle on the corresponding notebook

Return type:

DSSJupyterNotebook

property name#

Get the name of the notebook.

Used as identifier.

property language#

Get the language of the notebook.

Possible values are ‘python’, ‘R’, ‘toree’ (scala)

Return type:

string

property kernel_spec#

Get the raw kernel spec of the notebook.

The kernel spec is internal data for Jupyter, that identifies the kernel.

Returns:

the Jupyter spec of a Jupyter kernel. Top-level fields are:

  • name : identifier of the kernel in Jupyter, for example ‘python2’ or ‘python3’

  • display_name : name of the kernel as shown in the Jupyter UI

  • language : language of the notebook (informative field)

Return type:

dict

class dataikuapi.dss.jupyternotebook.DSSNotebookSession(client, session)#

Metadata associated to the session of a Jupyter Notebook.

Important

Do not instantiate directly, use dataikuapi.dss.jupyternotebook.DSSJupyterNotebook.get_sessions()

unload()#

Stop this Jupyter notebook and release its resources.

class dataikuapi.dss.jupyternotebook.DSSNotebookContent(client, project_key, notebook_name, content)#

The content of a Jupyter Notebook.

This is the actual notebook data, see the nbformat doc .

Important

Do not instantiate directly, use dataikuapi.dss.jupyternotebook.DSSJupyterNotebook.get_content()

get_raw()#

Get the raw content of this Jupyter notebook.

The content comprises metadata, cells, notebook format info.

Returns:

a dict containing the full content of a notebook. Notable fields are:

  • metadata : the metadata of the notebook, as a dict (see get_metadata())

  • nbformat and nbformat_minor : the version of the notebook format

  • cells : list of cells, each one a dict (see get_cells())

Return type:

dict

get_metadata()#

Get the metadata associated to this Jupyter notebook.

Returns:

the Jupyter metadata of the notebook, with fields:

  • kernelspec : identification of the kernel used to run the notebook

  • creator : name of the user that created the notebook

  • createdOn : timestamp of creation, in milliseconds

  • modifiedBy : name of last modifier

  • language_info : information on the language of the notebook

Return type:

dict

get_cells()#

Get the cells of this Jupyter notebook.

Returns:

a list of cells, as defined by Jupyter. Each cell is a dict, with notable fields:

  • cell_type : type of cell, for example ‘code’

  • metadata : notebook metadata in the cell

  • executionCount : index of the last run of this cell

  • source : content of the cell, as a list of string

  • output : list of outputs of the last run of the cell, as a list of dict with fields output_type, name and text. Exact contents and meaning depend on the cell type.

Return type:

list[dict]

save()#

Save the content of this Jupyter notebook.

class dataikuapi.dss.notebook.DSSNotebook(client, project_key, notebook_name, state=None)#

A Python/R/Scala notebook on the DSS instance.

Attention

Deprecated. Use dataikuapi.dss.jupyternotebook.DSSJupyterNotebook

unload(session_id=None)#

Stop this Jupyter notebook and release its resources.

Attention

Deprecated. Use dataikuapi.dss.jupyternotebook.DSSJupyterNotebook

get_state()#

Get the metadata associated to this Jupyter notebook.

Attention

Deprecated. Use dataikuapi.dss.jupyternotebook.DSSJupyterNotebook

get_sessions()#

Get the list of running sessions of this Jupyter notebook.

Attention

Deprecated. Use dataikuapi.dss.jupyternotebook.DSSJupyterNotebook

get_object_discussions()#

Get a handle to manage discussions on the notebook.

Attention

Deprecated. Use dataikuapi.dss.jupyternotebook.DSSJupyterNotebook

Returns:

the handle to manage discussions

Return type:

dataikuapi.dss.discussion.DSSObjectDiscussions

class dataikuapi.dss.admin.DSSGlobalApiKey(client, key)#

A global API key on the DSS instance

delete()#

Delete the api key

Note

This call requires an API key with admin rights

get_definition()#

Get the API key’s definition

Note

This call requires an API key with admin rights

Returns:

the API key definition, as a dict. The API key should be in a key field, distinct of the id field which contains an identifier of the key. The dict additionally contains the definition of the permissions attached to the key.

Return type:

dict

set_definition(definition)#

Set the API key’s definition.

Note

This call requires an API key with admin rights

Important

You should only set_definition() using an object that you obtained through get_definition(), not create a new dict.

Usage example

# make an API key able to create projects
key = client.get_global_api_key('my_api_key_secret')
definition = key.get_definition()
definition["globalPermissions"]["mayCreateProjects"] = True
key.set_definition(definition)
Parameters:

definition (dict) – the definition for the API key

class dataikuapi.dss.admin.DSSGlobalApiKeyListItem(client, data)#

An item in a list of personal API key.

Important

Do not instantiate directly, use dataikuapi.DSSClient.list_global_api_keys() instead.

to_global_api_key()#

Gets a handle corresponding to this item

Return type:

DSSGlobalApiKey

property id#

Get the identifier of the API key

Return type:

string

property user_for_impersonation#

Get the user associated to the API key

Return type:

string

property key#

Get the API key

Return type:

string

property label#

Get the label of the API key

Return type:

string

property description#

Get the description of the API key

Return type:

string

property created_on#

Get the timestamp of when the API key was created

Return type:

datetime.datetime

property created_by#

Get the login of the user who created the API key

Return type:

string

class dataikuapi.dss.admin.DSSGlobalUsageSummary(data)#

The summary of the usage of the DSS instance.

Important

Do not instantiate directly, use dataikuapi.DSSClient.get_global_usage_summary() instead.

property raw#

Get the usage summary structure

The summary report has top-level fields per object type, like projectSummaries or datasets, each containing counts, usually a all global count, then several XXXXByType dict with counts by object sub-type (for example, for datasets the sub-type would be the type of the connection they’re using)

Return type:

dict

property projects_count#

Get the number of projects on the instance

Return type:

int

property total_datasets_count#

Get the number of datasets on the instance

Return type:

int

property total_recipes_count#

Get the number of recipes on the instance

Return type:

int

property total_jupyter_notebooks_count#

Get the number of code nobteooks on the instance

Return type:

int

property total_sql_notebooks_count#

Get the number of sql notebooks on the instance

Return type:

int

property total_scenarios_count#

Get the number of scenarios on the instance

Return type:

int

property total_active_with_trigger_scenarios_count#

Get the number of active scenarios on the instance

Return type:

int

class dataikuapi.dss.admin.DSSPersonalApiKey(client, id_)#

A personal API key on the DSS instance.

Important

Do not instantiate directly, use dataikuapi.DSSClient.get_personal_api_key() instead.

get_definition()#

Get the API key’s definition

Returns:

the personal API key definition, as a dict. The key itself is in a key field, and the login of the user of this personal key in a user field.

Return type:

dict

delete()#

Delete the API key

class dataikuapi.dss.admin.DSSPersonalApiKeyListItem(client, data)#

An item in a list of personal API key.

Important

Do not instantiate directly, use dataikuapi.DSSClient.list_personal_api_keys() or dataikuapi.DSSClient.list_all_personal_api_keys() instead.

to_personal_api_key()#

Gets a handle corresponding to this item

Return type:

DSSPersonalApiKey

property id#

Get the identifier of the API key

Return type:

string

property user#

Get the user associated to the API key

Return type:

string

property key#

Get the API key

Return type:

string

property label#

Get the label of the API key

Return type:

string

property description#

Get the description of the API key

Return type:

string

property created_on#

Get the timestamp of when the API key was created

Return type:

datetime.datetime

property created_by#

Get the login of the user who created the API key

Return type:

string