Flowfile 0.3.6__py3-none-any.whl → 0.3.8__py3-none-any.whl

flowfile_core/flowfile/flow_node/flow_node.py

@@ -1,4 +1,3 @@
-
 from typing import List, Union, Callable, Any, Optional, Generator, Literal
 from flowfile_core.configs import logger
 from flowfile_core.flowfile.flow_data_engine.flow_file_column.main import FlowfileColumn
@@ -6,6 +5,7 @@ from flowfile_core.flowfile.flow_data_engine.flow_data_engine import FlowDataEng
 from flowfile_core.utils.arrow_reader import get_read_top_n
 from flowfile_core.schemas import input_schema, schemas
 from flowfile_core.configs.flow_logger import NodeLogger
+from flowfile_core.configs.settings import SINGLE_FILE_MODE, OFFLOAD_TO_WORKER
 
 from flowfile_core.schemas.output_model import TableExample, FileColumn, NodeData
 from flowfile_core.flowfile.utils import get_hash
@@ -13,13 +13,19 @@ from flowfile_core.configs.node_store import nodes as node_interface
 from flowfile_core.flowfile.setting_generator import setting_generator, setting_updator
 from time import sleep
 from flowfile_core.flowfile.flow_data_engine.subprocess_operations import (
-    ExternalDfFetcher, ExternalSampler, results_exists, get_external_df_result, ExternalDatabaseFetcher, ExternalDatabaseWriter, ExternalCloudWriter)
+    ExternalDfFetcher, ExternalSampler, results_exists, get_external_df_result,
+    ExternalDatabaseFetcher, ExternalDatabaseWriter, ExternalCloudWriter)
 from flowfile_core.flowfile.flow_node.models import (NodeStepSettings, NodeStepInputs, NodeSchemaInformation,
                                                      NodeStepStats, NodeResults)
 from flowfile_core.flowfile.flow_node.schema_callback import SingleExecutionFuture
 
 
 class FlowNode:
+    """Represents a single node in a data flow graph.
+
+    This class manages the node's state, its data processing function,
+    and its connections to other nodes within the graph.
+    """
     parent_uuid: str
     node_type: str
     node_template: node_interface.NodeTemplate
@@ -35,12 +41,62 @@ class FlowNode:
     _setting_input: Any = None
     _hash: Optional[str] = None  # host this for caching results
     _function: Callable = None  # the function that needs to be executed when triggered
+    _name: str = None  # name of the node, used for display
     _schema_callback: Optional[SingleExecutionFuture] = None  # Function that calculates the schema without executing
     _state_needs_reset: bool = False
     _fetch_cached_df: Optional[ExternalDfFetcher | ExternalDatabaseFetcher | ExternalDatabaseWriter | ExternalCloudWriter] = None
     _cache_progress: Optional[ExternalDfFetcher | ExternalDatabaseFetcher | ExternalDatabaseWriter | ExternalCloudWriter] = None
 
+    def __init__(self, node_id: Union[str, int], function: Callable,
+                 parent_uuid: str,
+                 setting_input: Any,
+                 name: str,
+                 node_type: str,
+                 input_columns: List[str] = None,
+                 output_schema: List[FlowfileColumn] = None,
+                 drop_columns: List[str] = None,
+                 renew_schema: bool = True,
+                 pos_x: float = 0,
+                 pos_y: float = 0,
+                 schema_callback: Callable = None,
+                 ):
+        """Initializes a FlowNode instance.
+
+        Args:
+            node_id: Unique identifier for the node.
+            function: The core data processing function for the node.
+            parent_uuid: The UUID of the parent flow.
+            setting_input: The configuration/settings object for the node.
+            name: The name of the node.
+            node_type: The type identifier of the node (e.g., 'join', 'filter').
+            input_columns: List of column names expected as input.
+            output_schema: The schema of the columns to be added.
+            drop_columns: List of column names to be dropped.
+            renew_schema: Flag to indicate if the schema should be renewed.
+            pos_x: The x-coordinate on the canvas.
+            pos_y: The y-coordinate on the canvas.
+            schema_callback: A custom function to calculate the output schema.
+        """
+        self._name = None
+        self.parent_uuid = parent_uuid
+        self.post_init()
+        self.active = True
+        self.node_information.id = node_id
+        self.node_type = node_type
+        self.node_settings.renew_schema = renew_schema
+        self.update_node(function=function,
+                         input_columns=input_columns,
+                         output_schema=output_schema,
+                         drop_columns=drop_columns,
+                         setting_input=setting_input,
+                         name=name,
+                         pos_x=pos_x,
+                         pos_y=pos_y,
+                         schema_callback=schema_callback,
+                         )
+
     def post_init(self):
+        """Initializes or resets the node's attributes to their default states."""
         self.node_inputs = NodeStepInputs()
         self.node_stats = NodeStepStats()
         self.node_settings = NodeStepSettings()
@@ -54,19 +110,32 @@ class FlowNode:
         self._state_needs_reset = False
 
     @property
-    def state_needs_reset(self):
+    def state_needs_reset(self) -> bool:
+        """Checks if the node's state needs to be reset.
+
+        Returns:
+            True if a reset is required, False otherwise.
+        """
         return self._state_needs_reset
 
     @state_needs_reset.setter
     def state_needs_reset(self, v: bool):
+        """Sets the flag indicating that the node's state needs to be reset.
+
+        Args:
+            v: The boolean value to set.
+        """
         self._state_needs_reset = v
 
     @staticmethod
     def create_schema_callback_from_function(f: Callable) -> Callable[[], List[FlowfileColumn]]:
-        """
-        Create a schema callback from a function.
-        :param f: Function that returns the schema
-        :return: Callable that returns the schema
+        """Wraps a node's function to create a schema callback that extracts the schema.
+
+        Args:
+            f: The node's core function that returns a FlowDataEngine instance.
+
+        Returns:
+            A callable that, when executed, returns the output schema.
         """
         def schema_callback() -> List[FlowfileColumn]:
             try:
@@ -79,6 +148,13 @@ class FlowNode:
 
     @property
     def schema_callback(self) -> SingleExecutionFuture:
+        """Gets the schema callback function, creating one if it doesn't exist.
+
+        The callback is used for predicting the output schema without full execution.
+
+        Returns:
+            A SingleExecutionFuture instance wrapping the schema function.
+        """
         if self._schema_callback is None:
             if self.user_provided_schema_callback is not None:
                 self.schema_callback = self.user_provided_schema_callback
@@ -88,6 +164,11 @@ class FlowNode:
 
     @schema_callback.setter
     def schema_callback(self, f: Callable):
+        """Sets the schema callback function for the node.
+
+        Args:
+            f: The function to be used for schema calculation.
+        """
         if f is None:
             return
 
@@ -101,9 +182,24 @@ class FlowNode:
 
     @property
     def is_start(self) -> bool:
+        """Determines if the node is a starting node in the flow.
+
+        A starting node requires no inputs.
+
+        Returns:
+            True if the node is a start node, False otherwise.
+        """
         return not self.has_input and self.node_template.input == 0
 
     def get_input_type(self, node_id: int) -> List:
+        """Gets the type of connection ('main', 'left', 'right') for a given input node ID.
+
+        Args:
+            node_id: The ID of the input node.
+
+        Returns:
+            A list of connection types for that node ID.
+        """
         relation_type = []
         if node_id in [n.node_id for n in self.node_inputs.main_inputs]:
             relation_type.append('main')
@@ -113,36 +209,6 @@ class FlowNode:
             relation_type.append('right')
         return list(set(relation_type))
 
-    def __init__(self, node_id: Union[str, int], function: Callable,
-                 parent_uuid: str,
-                 setting_input: Any,
-                 name: str,
-                 node_type: str,
-                 input_columns: List[str] = None,
-                 output_schema: List[FlowfileColumn] = None,
-                 drop_columns: List[str] = None,
-                 renew_schema: bool = True,
-                 pos_x: float = 0,
-                 pos_y: float = 0,
-                 schema_callback: Callable = None,
-                 ):
-        self.parent_uuid = parent_uuid
-        self.post_init()
-        self.active = True
-        self.node_information.id = node_id
-        self.node_type = node_type
-        self.node_settings.renew_schema = renew_schema
-        self.update_node(function=function,
-                         input_columns=input_columns,
-                         output_schema=output_schema,
-                         drop_columns=drop_columns,
-                         setting_input=setting_input,
-                         name=name,
-                         pos_x=pos_x,
-                         pos_y=pos_y,
-                         schema_callback=schema_callback,
-                         )
-
     def update_node(self,
                     function: Callable,
                     input_columns: List[str] = None,
@@ -154,9 +220,24 @@ class FlowNode:
                     pos_y: float = 0,
                     schema_callback: Callable = None,
                     ):
+        """Updates the properties of the node.
+
+        This is called during initialization and when settings are changed.
+
+        Args:
+            function: The new core data processing function.
+            input_columns: The new list of input columns.
+            output_schema: The new schema of added columns.
+            drop_columns: The new list of dropped columns.
+            name: The new name for the node.
+            setting_input: The new settings object.
+            pos_x: The new x-coordinate.
+            pos_y: The new y-coordinate.
+            schema_callback: The new custom schema callback function.
+        """
         self.user_provided_schema_callback = schema_callback
-        self.node_information.y_position = pos_y
-        self.node_information.x_position = pos_x
+        self.node_information.y_position = int(pos_y)
+        self.node_information.x_position = int(pos_x)
         self.node_information.setting_input = setting_input
         self.name = self.node_type if name is None else name
         self._function = function
@@ -178,20 +259,40 @@ class FlowNode:
         self.setting_input = setting_input  # wait until the end so that the hash is calculated correctly
 
     @property
-    def name(self):
+    def name(self) -> str:
+        """Gets the name of the node.
+
+        Returns:
+            The node's name.
+        """
         return self._name
 
     @name.setter
     def name(self, name: str):
+        """Sets the name of the node.
+
+        Args:
+            name: The new name.
+        """
         self._name = name
         self.__name__ = name
 
     @property
-    def setting_input(self):
+    def setting_input(self) -> Any:
+        """Gets the node's specific configuration settings.
+
+        Returns:
+            The settings object.
+        """
         return self._setting_input
 
     @setting_input.setter
     def setting_input(self, setting_input: Any):
+        """Sets the node's configuration and triggers a reset if necessary.
+
+        Args:
+            setting_input: The new settings object.
+        """
         is_manual_input = (self.node_type == 'manual_input' and
                            isinstance(setting_input, input_schema.NodeManualInput) and
                            isinstance(self._setting_input, input_schema.NodeManualInput)
@@ -209,24 +310,48 @@ class FlowNode:
             self.reset()
 
     @property
-    def node_id(self):
+    def node_id(self) -> Union[str, int]:
+        """Gets the unique identifier of the node.
+
+        Returns:
+            The node's ID.
+        """
         return self.node_information.id
 
     @property
-    def left_input(self):
+    def left_input(self) -> Optional["FlowNode"]:
+        """Gets the node connected to the left input port.
+
+        Returns:
+            The left input FlowNode, or None.
+        """
         return self.node_inputs.left_input
 
     @property
-    def right_input(self):
+    def right_input(self) -> Optional["FlowNode"]:
+        """Gets the node connected to the right input port.
+
+        Returns:
+            The right input FlowNode, or None.
+        """
         return self.node_inputs.right_input
 
     @property
     def main_input(self) -> List["FlowNode"]:
+        """Gets the list of nodes connected to the main input port(s).
+
+        Returns:
+            A list of main input FlowNodes.
+        """
         return self.node_inputs.main_inputs
 
     @property
-    def is_correct(self):
-        # Check if inputs meet requirements
+    def is_correct(self) -> bool:
+        """Checks if the node's input connections satisfy its template requirements.
+
+        Returns:
+            True if connections are valid, False otherwise.
+        """
         if isinstance(self.setting_input, input_schema.NodePromise):
             return False
         return (self.node_template.input == len(self.node_inputs.get_all_inputs()) or
@@ -234,6 +359,10 @@ class FlowNode:
                 (self.node_template.multi and self.node_template.can_be_start))
 
     def set_node_information(self):
+        """Populates the `node_information` attribute with the current state.
+
+        This includes the node's connections, settings, and position.
+        """
         logger.info('setting node information')
         node_information = self.node_information
         node_information.left_input_id = self.node_inputs.left_input.node_id if self.left_input else None
@@ -248,43 +377,76 @@ class FlowNode:
         node_information.type = self.node_type
 
     def get_node_information(self) -> schemas.NodeInformation:
+        """Updates and returns the node's information object.
+
+        Returns:
+            The `NodeInformation` object for this node.
+        """
         self.set_node_information()
         return self.node_information
 
     @property
-    def function(self):
+    def function(self) -> Callable:
+        """Gets the core processing function of the node.
+
+        Returns:
+            The callable function.
+        """
         return self._function
 
-    def reset_hash(self) -> bool:
-        old_hash = self._hash
-        self._hash = None
-        if self.hash != old_hash:
-            if self.node_settings.cache_results:
-                self.remove_cache()
-            return True
-        return False
+    @function.setter
+    def function(self, function: Callable):
+        """Sets the core processing function of the node.
+
+        Args:
+            function: The new callable function.
+        """
+        self._function = function
 
     @property
     def all_inputs(self) -> List["FlowNode"]:
+        """Gets a list of all nodes connected to any input port.
+
+        Returns:
+            A list of all input FlowNodes.
+        """
         return self.node_inputs.get_all_inputs()
 
-    def calculate_hash(self, setting_input: Any):
+    def calculate_hash(self, setting_input: Any) -> str:
+        """Calculates a hash based on settings and input node hashes.
+
+        Args:
+            setting_input: The node's settings object to be included in the hash.
+
+        Returns:
+            A string hash value.
+        """
         depends_on_hashes = [_node.hash for _node in self.all_inputs]
         node_data_hash = get_hash(setting_input)
         return get_hash(depends_on_hashes + [node_data_hash, self.parent_uuid])
 
     @property
-    def hash(self):
+    def hash(self) -> str:
+        """Gets the cached hash for the node, calculating it if it doesn't exist.
+
+        Returns:
+            The string hash value.
+        """
         if not self._hash:
             self._hash = self.calculate_hash(self.setting_input)
         return self._hash
 
-    @function.setter
-    def function(self, function: Callable):
-        self._function = function
-        # self.reset()
+    def add_node_connection(self, from_node: "FlowNode",
+                            insert_type: Literal['main', 'left', 'right'] = 'main') -> None:
+        """Adds a connection from a source node to this node.
+
+        Args:
+            from_node: The node to connect from.
+            insert_type: The type of input to connect to ('main', 'left', 'right').
 
-    def add_node_connection(self, from_node: "FlowNode", insert_type: Literal['main', 'left', 'right'] = 'main'):
+        Raises:
+            Exception: If the insert_type is invalid.
+        """
         from_node.leads_to_nodes.append(self)
         if insert_type == 'main':
             if self.node_template.input <= 2 or self.node_inputs.main_inputs is None:
@@ -303,20 +465,39 @@ class FlowNode:
         self.reset()
         from_node.reset()
 
-    def evaluate_nodes(self, deep: bool = False):
+    def evaluate_nodes(self, deep: bool = False) -> None:
+        """Triggers a state reset for all directly connected downstream nodes.
+
+        Args:
+            deep: If True, the reset propagates recursively through the entire downstream graph.
+        """
         for node in self.leads_to_nodes:
             self.print(f'resetting node: {node.node_id}')
             node.reset(deep)
 
-    def get_flow_file_column_schema(self, col_name: str) -> FlowfileColumn:
+    def get_flow_file_column_schema(self, col_name: str) -> FlowfileColumn | None:
+        """Retrieves the schema for a specific column from the output schema.
+
+        Args:
+            col_name: The name of the column.
+
+        Returns:
+            The FlowfileColumn object for that column, or None if not found.
+        """
         for s in self.schema:
             if s.column_name == col_name:
                 return s
 
-    def get_predicted_schema(self, force: bool = False):
-        """
-        Method to get a predicted schema based on the columns that are dropped and added
-        :return:
+    def get_predicted_schema(self, force: bool = False) -> List[FlowfileColumn] | None:
+        """Predicts the output schema of the node without full execution.
+
+        It uses the schema_callback or infers from predicted data.
+
+        Args:
+            force: If True, forces recalculation even if a predicted schema exists.
+
+        Returns:
+            A list of FlowfileColumn objects representing the predicted schema.
         """
         if self.node_schema.predicted_schema and not force:
             return self.node_schema.predicted_schema
@@ -338,6 +519,11 @@ class FlowNode:
 
     @property
     def is_setup(self) -> bool:
+        """Checks if the node has been properly configured and is ready for execution.
+
+        Returns:
+            True if the node is set up, False otherwise.
+        """
         if not self.node_information.is_setup:
             if self.function.__name__ != 'placeholder':
                 self.node_information.is_setup = True
@@ -345,9 +531,24 @@ class FlowNode:
         return self.node_information.is_setup
 
     def print(self, v: Any):
+        """Helper method to log messages with node context.
+
+        Args:
+            v: The message or value to log.
+        """
         logger.info(f'{self.node_type}, node_id: {self.node_id}: {v}')
 
     def get_resulting_data(self) -> FlowDataEngine | None:
+        """Executes the node's function to produce the actual output data.
+
+        Handles both regular functions and external data sources.
+
+        Returns:
+            A FlowDataEngine instance containing the result, or None on error.
+
+        Raises:
+            Exception: Propagates exceptions from the node's function execution.
+        """
         if self.is_setup:
             if self.results.resulting_data is None and self.results.errors is None:
                 self.print('getting resulting data')
@@ -375,6 +576,13 @@ class FlowNode:
             return self.results.resulting_data
 
     def _predicted_data_getter(self) -> FlowDataEngine | None:
+        """Internal helper to get a predicted data result.
+
+        This calls the function with predicted data from input nodes.
+
+        Returns:
+            A FlowDataEngine instance with predicted data, or an empty one on error.
+        """
         try:
             fl = self._function(*[v.get_predicted_resulting_data() for v in self.all_inputs])
             return fl
@@ -391,6 +599,13 @@ class FlowNode:
             logger.warning(e)
 
     def get_predicted_resulting_data(self) -> FlowDataEngine:
+        """Creates a `FlowDataEngine` instance based on the predicted schema.
+
+        This avoids executing the node's full logic.
+
+        Returns:
+            A FlowDataEngine instance with a schema but no data.
+        """
         if self.needs_run(False) and self.schema_callback is not None or self.node_schema.result_schema is not None:
             self.print('Getting data based on the schema')
 
@@ -404,17 +619,28 @@ class FlowNode:
             return fl
 
     def add_lead_to_in_depend_source(self):
+        """Ensures this node is registered in the `leads_to_nodes` list of its inputs."""
         for input_node in self.all_inputs:
             if self.node_id not in [n.node_id for n in input_node.leads_to_nodes]:
                 input_node.leads_to_nodes.append(self)
 
     def get_all_dependent_nodes(self) -> Generator["FlowNode", None, None]:
+        """Yields all downstream nodes recursively.
+
+        Returns:
+            A generator of all dependent FlowNode objects.
+        """
         for node in self.leads_to_nodes:
             yield node
             for n in node.get_all_dependent_nodes():
                 yield n
 
     def get_all_dependent_node_ids(self) -> Generator[int, None, None]:
+        """Yields the IDs of all downstream nodes recursively.
+
+        Returns:
+            A generator of all dependent node IDs.
+        """
         for node in self.leads_to_nodes:
             yield node.node_id
             for n in node.get_all_dependent_node_ids():
@@ -422,6 +648,13 @@ class FlowNode:
 
     @property
     def schema(self) -> List[FlowfileColumn]:
+        """Gets the definitive output schema of the node.
+
+        If not already run, it falls back to the predicted schema.
+
+        Returns:
+            A list of FlowfileColumn objects.
+        """
         try:
             if self.is_setup and self.results.errors is None:
                 if self.node_schema.result_schema is not None and len(self.node_schema.result_schema) > 0:
@@ -434,31 +667,42 @@ class FlowNode:
                 return self.node_schema.result_schema
             else:
                 return []
-        except:
+        except Exception as e:
+            logger.error(e)
             return []
 
-    def load_from_cache(self) -> FlowDataEngine:
-        if results_exists(self.hash):
-            try:
-                return FlowDataEngine(self._fetch_cached_df.get_result())
-            except Exception as e:
-                logger.error(e)
-
     def remove_cache(self):
+        """Removes cached results for this node.
+
+        Note: Currently not fully implemented.
+        """
+
         if results_exists(self.hash):
             logger.warning('Not implemented')
 
     def needs_run(self, performance_mode: bool, node_logger: NodeLogger = None,
                   execution_location: schemas.ExecutionLocationsLiteral = "auto") -> bool:
-        if execution_location == "local":
+        """Determines if the node needs to be executed.
+
+        The decision is based on its run state, caching settings, and execution mode.
+
+        Args:
+            performance_mode: True if the flow is in performance mode.
+            node_logger: The logger instance for this node.
+            execution_location: The target execution location.
+
+        Returns:
+            True if the node should be run, False otherwise.
+        """
+        if execution_location == "local" or SINGLE_FILE_MODE:
             return False
+
         flow_logger = logger if node_logger is None else node_logger
         cache_result_exists = results_exists(self.hash)
         if not self.node_stats.has_run_with_current_setup:
             flow_logger.info('Node has not run, needs to run')
             return True
         if self.node_settings.cache_results and cache_result_exists:
-
             return False
         elif self.node_settings.cache_results and not cache_result_exists:
             return True
@@ -468,9 +712,44 @@ class FlowNode:
             return True
 
     def __call__(self, *args, **kwargs):
+        """Makes the node instance callable, acting as an alias for execute_node."""
         self.execute_node(*args, **kwargs)
 
+    def execute_full_local(self, performance_mode: bool = False) -> None:
+        """Executes the node's logic locally, including example data generation.
+
+        Args:
+            performance_mode: If True, skips generating example data.
+
+        Raises:
+            Exception: Propagates exceptions from the execution.
+        """
+        def example_data_generator():
+            example_data = None
+
+            def get_example_data():
+                nonlocal example_data
+                if example_data is None:
+                    example_data = resulting_data.get_sample(100).to_arrow()
+                return example_data
+            return get_example_data
+        resulting_data = self.get_resulting_data()
+
+        if not performance_mode:
+            self.results.example_data_generator = example_data_generator()
+            self.node_schema.result_schema = self.results.resulting_data.schema
+            self.node_stats.has_completed_last_run = True
+
     def execute_local(self, flow_id: int, performance_mode: bool = False):
+        """Executes the node's logic locally.
+
+        Args:
+            flow_id: The ID of the parent flow.
+            performance_mode: If True, skips generating example data.
+
+        Raises:
+            Exception: Propagates exceptions from the execution.
+        """
         try:
             resulting_data = self.get_resulting_data()
             if not performance_mode:
@@ -495,7 +774,15 @@ class FlowNode:
                     step.node_settings.streamable = self.node_settings.streamable
 
     def execute_remote(self, performance_mode: bool = False, node_logger: NodeLogger = None):
-        # flow_logger = logger if flow_logger is None else flow_logger
+        """Executes the node's logic remotely or handles cached results.
+
+        Args:
+            performance_mode: If True, skips generating example data.
+            node_logger: The logger for this node execution.
+
+        Raises:
+            Exception: If the node_logger is not provided or if execution fails.
+        """
         if node_logger is None:
             raise Exception('Node logger is not defined')
         if self.node_settings.cache_results and results_exists(self.hash):
@@ -552,11 +839,15 @@ class FlowNode:
                 self._fetch_cached_df = None
 
     def prepare_before_run(self):
+        """Resets results and errors before a new execution."""
+
         self.results.errors = None
         self.results.resulting_data = None
         self.results.example_data = None
 
     def cancel(self):
+        """Cancels an ongoing external process if one is running."""
+
         if self._fetch_cached_df is not None:
             self._fetch_cached_df.cancel()
             self.node_stats.is_canceled = True
@@ -566,6 +857,18 @@ class FlowNode:
 
     def execute_node(self, run_location: schemas.ExecutionLocationsLiteral, reset_cache: bool = False,
                      performance_mode: bool = False, retry: bool = True, node_logger: NodeLogger = None):
+        """Orchestrates the execution, handling location, caching, and retries.
+
+        Args:
+            run_location: The location for execution ('local', 'remote').
+            reset_cache: If True, forces removal of any existing cache.
+            performance_mode: If True, optimizes for speed over diagnostics.
+            retry: If True, allows retrying execution on recoverable errors.
+            node_logger: The logger for this node execution.
+
+        Raises:
+            Exception: If the node_logger is not defined.
+        """
         if node_logger is None:
             raise Exception('Flow logger is not defined')
         # node_logger = flow_logger.get_node_logger(self.node_id)
@@ -575,7 +878,8 @@ class FlowNode:
             self.node_stats.has_completed_last_run = False
         if self.is_setup:
             node_logger.info(f'Starting to run {self.__name__}')
-            if self.needs_run(performance_mode, node_logger, run_location):
+            if (self.needs_run(performance_mode, node_logger, run_location) or self.node_template.node_group == "output"
+                    and not (run_location == 'local' or SINGLE_FILE_MODE)):
                 self.prepare_before_run()
                 try:
                     if ((run_location == 'remote' or (self.node_default.transform_type == 'wide')
@@ -605,13 +909,28 @@ class FlowNode:
                     else:
                         self.results.errors = str(e)
                         node_logger.error(f'Error with running the node: {e}')
-
+            elif ((run_location == 'local' or SINGLE_FILE_MODE) and
+                  (not self.node_stats.has_run_with_current_setup or self.node_template.node_group == "output")):
+                try:
+                    node_logger.info('Executing fully locally')
+                    self.execute_full_local(performance_mode)
+                except Exception as e:
+                    self.results.errors = str(e)
+                    node_logger.error(f'Error with running the node: {e}')
+                    self.node_stats.error = str(e)
+                    self.node_stats.has_completed_last_run = False
+                self.node_stats.has_run_with_current_setup = True
             else:
                 node_logger.info('Node has already run, not running the node')
         else:
             node_logger.warning(f'Node {self.__name__} is not setup, cannot run the node')
 
     def store_example_data_generator(self, external_df_fetcher: ExternalDfFetcher | ExternalSampler):
+        """Stores a generator function for fetching a sample of the result data.
+
+        Args:
+            external_df_fetcher: The process that generated the sample data.
+        """
         if external_df_fetcher.status is not None:
             file_ref = external_df_fetcher.status.file_ref
             self.results.example_data_path = file_ref
@@ -620,9 +939,21 @@ class FlowNode:
             logger.error('Could not get the sample data, the external process is not ready')
 
     def needs_reset(self) -> bool:
+        """Checks if the node's hash has changed, indicating an outdated state.
+
+        Returns:
+            True if the calculated hash differs from the stored hash.
+        """
         return self._hash != self.calculate_hash(self.setting_input)
 
     def reset(self, deep: bool = False):
+        """Resets the node's execution state and schema information.
+
+        This also triggers a reset on all downstream nodes.
+
+        Args:
+            deep: If True, forces a reset even if the hash hasn't changed.
+        """
         needs_reset = self.needs_reset() or deep
         if needs_reset:
             logger.info(f'{self.node_id}: Node needs reset')
@@ -637,10 +968,19 @@ class FlowNode:
             self.node_schema.predicted_schema = None
             self._hash = None
             self.node_information.is_setup = None
+            self.results.errors = None
             self.evaluate_nodes()
             _ = self.hash  # Recalculate the hash after reset
 
     def delete_lead_to_node(self, node_id: int) -> bool:
+        """Removes a connection to a specific downstream node.
+
+        Args:
+            node_id: The ID of the downstream node to disconnect.
+
+        Returns:
+            True if the connection was found and removed, False otherwise.
+        """
         logger.info(f'Deleting lead to node: {node_id}')
         for i, lead_to_node in enumerate(self.leads_to_nodes):
             logger.info(f'Checking lead to node: {lead_to_node.node_id}')
@@ -652,7 +992,16 @@ class FlowNode:
 
     def delete_input_node(self, node_id: int, connection_type: input_schema.InputConnectionClass = 'input-0',
                           complete: bool = False) -> bool:
-        #  connection type must be in right, left or main
+        """Removes a connection from a specific input node.
+
+        Args:
+            node_id: The ID of the input node to disconnect.
+            connection_type: The specific input handle (e.g., 'input-0', 'input-1').
+            complete: If True, tries to delete from all input types.
+
+        Returns:
+            True if a connection was found and removed, False otherwise.
+        """
         deleted: bool = False
         if connection_type == 'input-0':
             for i, node in enumerate(self.node_inputs.main_inputs):
@@ -675,17 +1024,32 @@ class FlowNode:
             self.reset()
         return deleted
 
-    def __repr__(self):
+    def __repr__(self) -> str:
+        """Provides a string representation of the FlowNode instance.
+
+        Returns:
+            A string showing the node's ID and type.
+        """
         return f"Node id: {self.node_id} ({self.node_type})"
 
-    def _get_readable_schema(self):
+    def _get_readable_schema(self) -> List[dict] | None:
+        """Helper to get a simplified, dictionary representation of the output schema.
+
+        Returns:
+            A list of dictionaries, each with 'column_name' and 'data_type'.
+        """
         if self.is_setup:
             output = []
             for s in self.schema:
                 output.append(dict(column_name=s.column_name, data_type=s.data_type))
             return output
 
-    def get_repr(self):
+    def get_repr(self) -> dict:
+        """Gets a detailed dictionary representation of the node's state.
+
+        Returns:
+            A dictionary containing key information about the node.
+        """
         return dict(FlowNode=
                     dict(node_id=self.node_id,
                          step_name=self.__name__,
@@ -693,30 +1057,66 @@ class FlowNode:
                          output_schema=self._get_readable_schema()))
 
     @property
-    def number_of_leads_to_nodes(self) -> int:
+    def number_of_leads_to_nodes(self) -> int | None:
+        """Counts the number of downstream node connections.
+
+        Returns:
+            The number of nodes this node leads to.
+        """
         if self.is_setup:
             return len(self.leads_to_nodes)
 
     @property
     def has_next_step(self) -> bool:
+        """Checks if this node has any downstream connections.
+
+        Returns:
+            True if it has at least one downstream node.
+        """
         return len(self.leads_to_nodes) > 0
 
     @property
     def has_input(self) -> bool:
+        """Checks if this node has any input connections.
+
+        Returns:
+            True if it has at least one input node.
+        """
         return len(self.all_inputs) > 0
 
     @property
     def singular_input(self) -> bool:
+        """Checks if the node template specifies exactly one input.
+
+        Returns:
+            True if the node is a single-input type.
+        """
         return self.node_template.input == 1
 
     @property
     def singular_main_input(self) -> "FlowNode":
+        """Gets the input node, assuming it is a single-input type.
+
+        Returns:
+            The single input FlowNode, or None.
+        """
         if self.singular_input:
             return self.all_inputs[0]
 
     def get_table_example(self, include_data: bool = False) -> TableExample | None:
+        """Generates a `TableExample` model summarizing the node's output.
+
+        This can optionally include a sample of the data.
+
+        Args:
+            include_data: If True, includes a data sample in the result.
+
+        Returns:
+            A `TableExample` object, or None if the node is not set up.
+        """
         self.print('Getting a table example')
         if self.is_setup and include_data and self.node_stats.has_completed_last_run:
+
             if self.node_template.node_group == 'output':
                 self.print('getting the table example')
                 return self.main_input[0].get_table_example(include_data)
@@ -749,10 +1149,16 @@ class FlowNode:
                                 table_schema=schema, columns=columns,
                                 data=[])
 
-    def calculate_settings_out_select(self):
-        pass
-
     def get_node_data(self, flow_id: int, include_example: bool = False) -> NodeData:
+        """Gathers all necessary data for representing the node in the UI.
+
+        Args:
+            flow_id: The ID of the parent flow.
+            include_example: If True, includes data samples.
+
+        Returns:
+            A `NodeData` object.
+        """
         node = NodeData(flow_id=flow_id,
                         node_id=self.node_id,
                         has_run=self.node_stats.has_run_with_current_setup,
@@ -772,15 +1178,30 @@ class FlowNode:
         return node
 
     def get_output_data(self) -> TableExample:
+        """Gets the full output data sample for this node.
+
+        Returns:
+            A `TableExample` object with data.
+        """
         return self.get_table_example(True)
 
     def get_node_input(self) -> schemas.NodeInput:
+        """Creates a `NodeInput` schema object for representing this node in the UI.
+
+        Returns:
+            A `NodeInput` object.
+        """
         return schemas.NodeInput(pos_y=self.setting_input.pos_y,
                                  pos_x=self.setting_input.pos_x,
                                  id=self.node_id,
                                  **self.node_template.__dict__)
 
     def get_edge_input(self) -> List[schemas.NodeEdge]:
+        """Generates `NodeEdge` objects for all input connections to this node.
+
+        Returns:
+            A list of `NodeEdge` objects.
+        """
         edges = []
         if self.node_inputs.main_inputs is not None:
             for i, main_input in enumerate(self.node_inputs.main_inputs):