✨ ✅ Validate output data and add tests (#20) (#66)

* validate also output data and add tests

* remove convert parameters in test

* add check if dataset is already validated

---------

Co-authored-by: Manuel Spierenburg <manuel.spierenburg@sbb.ch>

kedro_pandera/framework/hooks/pandera_hook.py

@@ -1,7 +1,10 @@
 import logging
+from typing import Any, Dict, Set
 
 from kedro.framework.context import KedroContext
 from kedro.framework.hooks import hook_impl
+from kedro.io import DataCatalog
+from kedro.pipeline.node import Node
 from pandera.errors import SchemaError
 
 from kedro_pandera.framework.config.resolvers import (
@@ -22,6 +25,9 @@
 
 
 class PanderaHook:
+    def __init__(self) -> None:
+        self._validated_datasets: Set[str] = set()
+
     @property
     def _logger(self) -> logging.Logger:
         return logging.getLogger(__name__)
@@ -45,17 +51,18 @@ def after_context_created(
             }
         )
 
-    @hook_impl
-    def before_node_run(  # noqa : PLR0913
-        self, node, catalog, inputs, is_async, session_id
+    def _validate_datasets(
+        self, node: Node, catalog: DataCatalog, datasets: Dict[str, Any]
     ):
-        for name, data in inputs.items():
+        for name, data in datasets.items():
             if (
                 catalog._datasets[name].metadata is not None
                 and "pandera" in catalog._datasets[name].metadata
+                and name not in self._validated_datasets
             ):
                 try:
                     catalog._datasets[name].metadata["pandera"]["schema"].validate(data)
+                    self._validated_datasets.add(name)
                 except SchemaError as err:
                     self._logger.error(
                         f"Dataset '{name}' pandera validation failed before running '{node.name}', see details in the error message. "
@@ -71,5 +78,20 @@ def before_node_run(  # noqa : PLR0913
                     f"(kedro-pandera) Dataset '{name}' was successfully validated with pandera"
                 )
 
+    @hook_impl
+    def before_node_run(  # noqa : PLR0913
+        self,
+        node: Node,
+        catalog: DataCatalog,
+        inputs: Dict[str, Any],
+        is_async,
+        session_id,
+    ):
+        self._validate_datasets(node, catalog, inputs)
+
+    @hook_impl
+    def after_node_run(self, node: Node, catalog: DataCatalog, outputs: Dict[str, Any]):
+        self._validate_datasets(node, catalog, outputs)
+
 
 pandera_hook = PanderaHook()

tests/data/iris_schema.yml

@@ -1,85 +1,84 @@
-_example_iris_data_schema:
-  schema_type: dataframe
-  version: 0.18.3
-  columns:
-    sepal_length:
-      title: null
-      description: null
-      dtype: float64
-      nullable: false
-      checks:
-        greater_than_or_equal_to: 4.3
-        less_than_or_equal_to: 7.9
-      unique: false
-      coerce: false
-      required: true
-      regex: false
-    sepal_width:
-      title: null
-      description: null
-      dtype: float64
-      nullable: false
-      checks:
-        greater_than_or_equal_to: 2.0
-        less_than_or_equal_to: 4.4
-      unique: false
-      coerce: false
-      required: true
-      regex: false
-    petal_length:
-      title: null
-      description: null
-      dtype: float64
-      nullable: false
-      checks:
-        greater_than_or_equal_to: 1.0
-        less_than_or_equal_to: 6.9
-      unique: false
-      coerce: false
-      required: true
-      regex: false
-    petal_width:
-      title: null
-      description: null
-      dtype: float64
-      nullable: false
-      checks:
-        greater_than_or_equal_to: 0.1
-        less_than_or_equal_to: 2.5
-      unique: false
-      coerce: false
-      required: true
-      regex: false
-    species:
-      title: null
-      description: null
-      dtype: object
-      nullable: false
-      checks: null
-      unique: false
-      coerce: false
-      required: true
-      regex: false
-  checks: null
-  index:
-  - title: null
+schema_type: dataframe
+version: 0.18.3
+columns:
+  sepal_length:
+    title: null
     description: null
-    dtype: int64
+    dtype: float32
     nullable: false
     checks:
-      greater_than_or_equal_to: 0.0
-      less_than_or_equal_to: 149.0
-    name: null
+      greater_than_or_equal_to: 4.3
+      less_than_or_equal_to: 7.9
     unique: false
     coerce: false
-  dtype: null
-  coerce: true
-  strict: false
-  name: null
-  ordered: false
-  unique: null
-  report_duplicates: all
-  unique_column_names: false
-  add_missing_columns: false
-  title: null
+    required: true
+    regex: false
+  sepal_width:
+    title: null
+    description: null
+    dtype: float64
+    nullable: false
+    checks:
+      greater_than_or_equal_to: 2.0
+      less_than_or_equal_to: 4.4
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+  petal_length:
+    title: null
+    description: null
+    dtype: float64
+    nullable: false
+    checks:
+      greater_than_or_equal_to: 1.0
+      less_than_or_equal_to: 6.9
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+  petal_width:
+    title: null
+    description: null
+    dtype: float64
+    nullable: false
+    checks:
+      greater_than_or_equal_to: 0.1
+      less_than_or_equal_to: 2.5
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+  species:
+    title: null
+    description: null
+    dtype: string
+    nullable: false
+    checks: null
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+checks: null
+index:
+- title: null
   description: null
+  dtype: int64
+  nullable: false
+  checks:
+    greater_than_or_equal_to: 0.0
+    less_than_or_equal_to: 149.0
+  name: null
+  unique: false
+  coerce: false
+dtype: null
+coerce: true
+strict: false
+name: null
+ordered: false
+unique: null
+report_duplicates: all
+unique_column_names: false
+add_missing_columns: false
+title: null
+description: null

tests/data/iris_schema_fail.yml

@@ -0,0 +1,85 @@
+schema_type: dataframe
+version: 0.18.3
+columns:
+  sepal_length:
+    title: null
+    description: null
+    dtype: float32
+    nullable: false
+    checks:
+      greater_than_or_equal_to: 4.3
+      less_than_or_equal_to: 7.9
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+  sepal_width:
+    title: null
+    description: null
+    dtype: float64
+    nullable: false
+    checks:
+      greater_than_or_equal_to: 2.0
+      less_than_or_equal_to: 4.4
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+  petal_length:
+    title: null
+    description: null
+    dtype: float64
+    nullable: false
+    checks:
+      greater_than_or_equal_to: 1.0
+      less_than_or_equal_to: 6.9
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+  petal_width:
+    title: null
+    description: null
+    dtype: float64
+    nullable: false
+    checks:
+      greater_than_or_equal_to: 0.1
+      less_than_or_equal_to: 2.5
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+  species:
+    title: null
+    description: null
+    dtype: string
+    nullable: false
+    checks:
+      isin: ["satosa", "versicolor", "virginica"]
+    unique: false
+    coerce: false
+    required: true
+    regex: false
+checks: null
+index:
+- title: null
+  description: null
+  dtype: int64
+  nullable: false
+  checks:
+    greater_than_or_equal_to: 0.0
+    less_than_or_equal_to: 149.0
+  name: null
+  unique: false
+  coerce: false
+dtype: null
+coerce: true
+strict: false
+name: null
+ordered: false
+unique: null
+report_duplicates: all
+unique_column_names: false
+add_missing_columns: false
+title: null
+description: null