Skip to content

neops.core.provider.base.base

NeopsBaseProvider

The neops base provider contains all methods and required data processing for a concrete provider. To create a new provider, either extend this NeopsBaseProvider or a concrete provider

the neops base provider is inherited from the BaseRunCycle

JSON Schema

JSON Schema

Class variables

deprecated: bool
Indicates if this provider will no longer be supported in the future.

description: str
Description String, to give the user an indication of the purpose of this provider

execution_updater: neops.core.provider.base.execution_updater.ExecutionUpdater
class to update logs of the current execution (is set by initialization)

json_schema: Dict
The json_schema specifies the required input values to instantiate the provider as a task.

provider_type: neops.core.provider.base.enum.ProviderTypeEnum
provider_type specifices the provider type, currently we are suppoting vor different types of providers CONFIGURE, FACTS, CHECK and EXECUTE. The provider type can be used by "frontend" applications to have different processes for the provider types. The neops.io frontend the normal handels for example CONFIGURE provider types with a two step run, first dry run then apply.

result_writer: neops.core.provider.base.base_result_writer.BaseResultWriter
class to write results (is set by initialization)

run_input_json_schema: Dict
\ The run_input_json_schema specifies the required parameter a provider uses to run, but they must be given on every run and not on task instantination.

\ On a task instantination additional parameters can be requested as input values on a task run. They are merged with the run_input_json_schema values.

run_on: neops.core.provider.base.enum.RunOnEnum
run_on specifies on which entity the provider runs, possible values are GLOBAL, GROUP, DEVICE, INTERFACE, CLIENT and GENERIC. On GENERIC the task instance must overwrite the run on value with the init_adjust_run_on method, to specify the excact run_on.

run_on_all_if_empty: bool
\ run_on_all_if_empty specifies if all elements of a given run_on type are selected when a empty list is given as execute_on parameter on the run method.

run_on_result: Union[neops.core.provider.base.enum.RunOnEnum, NoneType]
run_on_result specifies for which entity the result is written. Normally this is the same as the run_on value and if run_on_result is None then the run_on value is used.

Example for different values of run_on and run_on_result: If you build a provider which get facts for interfaces, but those facts are only collectable per device (you always get the facts for all interfaces on this device). In such a case it could make sense to set run_on to DEVICE and run_on_result to INTERFACE, then you got the result per interface and it's visible for which interfaces the facts are collected.

run_on_strict: bool
\ run_on_strict specifies if we accept loose input entities for the execute_on_type and execute_on parameters on the run method.

Example: if run_on_strict is false and run_on is INTERFACE, then we accept a list of devices on the run method and will resolve all interfaces on those devices where the task is executed. This could be a problem for some cases, e.g if you have a configure provider where you prefere to know the exact entities. Because of that run_on_strict is true per default.

short_description: str
A Short description for list views or others.

success_message: str
The success_message is used if no message is set on the results object and if a task is successfully executed on an entity.

validate_input: bool
\ validate_input specifies if the parameters on the run method should by verified against the json schema definitions of the task itself, the provider run_input_json_schema and json_schema.

Methods

add_markdown_helptext(self,md_content: neops.core.libs.helptext.markdown_content.MarkDownContent) -> NoneType
Add documentation elements to the generated provider documentations to give a better insight how this provider should work.

\ On the base provider we add only the json schema of the task instantiation to give a better overview which parameters are required for this provider.

\ param md_content: MarkDownContent of the current documentation of this element.

add_nornir_processors(self) -> List[nornir.core.processor.Processor]
Add additional Nornir Processors, for more information see https://nornir.readthedocs.io/en/v2.5.0/tutorials/intro/processors.html

init_adjust_run_on(self,execute_on: Union[List[int], NoneType] = None,execute_on_type: Union[neops.core.provider.base.enum.RunOnEnum, NoneType] = None,dry_run: Union[bool, NoneType] = None,task_input_kwargs: Union[Dict[Any, Any], NoneType] = None,search_query: str = '',task_kwargs: Union[Dict[Any, Any], NoneType] = None,**kwargs) -> NoneType
Method to overwrite run_on, run_on_strict and run_on_result values on provider initialization based on the given paremeters of the run and the task itself.

For example used by generic providers.

\ param execute_on: List of element IDs to run on \ param execute_on_type: Type of those IDs (GROUP, DEVICE, INTERFACE, CLIENT) \ param dry_run: Indicates if no configuration changes should be made \ param task_input_kwargs: Run input arguments \ param task_kwargs: Task arguments

init_before_run(self,task_input_kwargs: Dict,execute_on: List = None,execute_on_type: neops.core.provider.base.enum.RunOnEnum = device,dry_run: bool = True,search_query: str = '',**kwargs) -> NoneType
Initialization of all elements before the task runs. This methots initializes all result objects.

\ param task_input_kwargs: Run input arguments \ param execute_on: List of element IDs to run on \ param execute_on_type: Type of those IDs (GROUP, DEVICE, INTERFACE, CLIENT) \ param dry_run: Indicates if no configuration changes should be made \ param search_query: An optional search query to limit the entities to run on (not yet supported)

init_without_run(self,**kwargs) -> NoneType

print_results(self) ->
prints the current result set.

run(self,task_input_kwargs: Dict,execute_on: List = None,execute_on_type: neops.core.provider.base.enum.RunOnEnum = device,dry_run: bool = True,search_query: str = '',**kwargs) -> NoneType
Runs the provider based on the initialized task.

\ param task_input_kwargs: Run input arguments \ param execute_on: List of element IDs to run on \ param execute_on_type: Type of those IDs (GROUP, DEVICE, INTERFACE, CLIENT) \ param dry_run: Indicates if no configuration changes should be made \ param search_query: An optional search query to limit the entities to run on (not yet supported)

update_do(self,do: str,append: bool = True) -> NoneType
Wrapper method to the execution updater update_do method

\ param do: string what is to do \ param append: decides if the do string is appended or replaced

update_done(self,done: str,append: bool = True) -> NoneType

update_log(self,log: str,append: bool = True) -> NoneType
Wrapper method to the execution updater update_log method

\ param log: the log string \ param append: decides if the log string is appended or replaced

validate_schema(self,task_kwargs: Dict = None) -> NoneType
\ Validates if the parameters of the task

\ param task_kwargs: task parameters

write_results(self,subtask_finish: bool = False,task_finish: bool = False) -> NoneType
Write the result states to the database.

\ param subtask_finish: is true if i am a subtasks (pre/post run task) and the task itself is finished \ param task_finish: is true if all subtasks and the task itself is finished

Functions

inherit_json_schema(json_schema: Dict = None) -> Dict
Merges JSON Schemas: If this method is called on one of the classes children, then super().json_schema resolves, else we do a pseudo merge. \ param json_schema: \ return: Dict of merged JSON Schema

init_from_task_model(neops_task: neops.core.models.neops_task.NeopsTask,nr: nornir.core.Nornir = None,**kwargs) -> neops.core.provider.base.base.NeopsBaseProvider
Initialization of the task by a the given NeopsTask model and the Nornir inventory.

\ param neops_task: Neops Task Model \ param nr: Nornir Inventory

\ return: Instance of the provider/task itself

merge_run_input_json_schema(json_schema: Dict = None) ->
Merges the JSON Schemas of instanciated task and those of this provider

\ param json_schema: \ return: merged JSON Schema