Feature/add test description (#44)

* Added macro __get_test_description and test_description to dq_log

* Remove redundant macro

* Added macro __get_test_description and test_description to dq_log

* Added integration tests for __get_test_description

* Added missing underscore

* feat: adj auto_generate_test_description

* fix: dq_tools__auto_generate_test_description: "1"

---------

Co-authored-by: Dat <dat@infinitelambda.com>

Author: il-ngocanh

dbt_project.yml

@@ -23,6 +23,7 @@ vars:
   # >> for metricflow
   # dbt_dq_tool_start_date
   # dbt_dq_tool_end_date
+  # dq_tools__auto_generate_test_description: 1 # Set 1 to automatically generate test description if not specified
 
 models:
   dq_tools:

integration_tests/dbt_project.yml

@@ -43,6 +43,7 @@ vars:
     by_database: ['dummy'] # dummy inputs
     by_schema: ['dq_tools_mart'] # exclude all dq_tools.03_mart
     by_table: ['dq_issue_log', 'dummy'] # dummy inputs
+  dq_tools__auto_generate_test_description: "1"
 
 seeds:
   dq_tools_integration_tests:

integration_tests/models/artifacts/test/utilities/_general/_general.yml

@@ -22,4 +22,9 @@ models:
   - name: test__get_where_subquery
     data_tests:
       - dq_tools.expression_is_true_db:
-          expression: actual = expected
+          expression: actual = expected
+  - name: test__get_test_description
+    data_tests:
+      - dq_tools.expression_is_true_db:
+          expression: actual = expected
+  

integration_tests/models/artifacts/test/utilities/_general/test__get_test_description.sql

@@ -0,0 +1,81 @@
+-- test not_null
+{% set result_node = {
+    "test_metadata": {
+        "name": "not_null",
+        "kwargs": {
+            "column_name": "indicator_category",
+            "model": "{{ get_where_subquery(ref('bi_column_analysis')) }}"
+        },
+    } 
+} %}
+
+select '{{ dq_tools.__get_test_description(result_node) }}' as actual,
+       'The indicator_category column in the bi_column_analysis model should not contain null values.' as expected
+
+-- test unique
+{% set result_node = {
+    "test_metadata": {
+        "name": "unique",
+        "kwargs": {
+            "column_name": "invocation_id",
+            "model": "{{ get_where_subquery(ref('test_coverage')) }}"
+        }
+    }
+} %}
+
+union all
+select '{{ dq_tools.__get_test_description(result_node) }}' as actual,
+       'The invocation_id column in the test_coverage model should be unique.' as expected
+
+-- test relationships
+{% set result_node = {
+
+    "test_metadata": {
+                "name": "relationships",
+                "kwargs": {
+                    "to": "ref('data_test_relationships_a')",
+                    "field": "a_id",
+                    "column_name": "b_id",
+                    "model": "{{ get_where_subquery(ref('data_test_relationships_b')) }}"
+                }
+    }
+} %}
+
+union all
+select '{{ dq_tools.__get_test_description((result_node)) }}' as actual,
+       'Each b_id in the data_test_relationships_b model exists as an id in the data_test_relationships_a table.' as expected
+
+-- test no test description
+{% set result_node = {
+    "test_metadata": {
+        "name": "not_null_where_db",
+        "kwargs": {
+            "column_name": "check_timestamp",
+            "model": "{{ get_where_subquery(ref('dq_tools_test')) }}"
+        }
+    }
+} %}
+
+union all
+select '{{ dq_tools.__get_test_description(result_node) }}' as actual,
+       '' as expected
+
+{% set result_node = {
+    "test_metadata": {
+                "name": "accepted_values",
+                "kwargs": {
+                    "values": [
+                        "Simple Statistics"
+                    ],
+                    "column_name": "indicator_category",
+                    "model": "{{ get_where_subquery(ref('bi_dq_metrics')) }}"
+                },
+                "namespace": null
+            },
+    "description": "The indicator_category column should only contain simple statistics value"
+} %}
+
+union all
+select 
+    '{{ dq_tools.__get_test_description(result_node) }}' as actual,
+    'The indicator_category column should only contain simple statistics value' as expected

integration_tests/models/time_spine.yml

@@ -0,0 +1,7 @@
+  models:
+    - name: metricflow_time_spine
+      time_spine:
+        standard_granularity_column: date_day    
+      columns:
+        - name: date_day
+          granularity: day

macros/artifacts/test/store_test_results.sql

@@ -46,6 +46,7 @@
       ,no_of_table_columns
       ,no_of_tables
       ,test_unique_id
+      ,test_description
     )
 
     with logs as (
@@ -75,6 +76,7 @@
               ,no_of_table_columns
               ,{{ no_of_tables }} as no_of_tables
               ,test_unique_id
+              ,test_description
 
     from      logs;
 

macros/artifacts/test/utilities/general/__get_test_description.sql

@@ -0,0 +1,39 @@
+{% macro __get_test_description(test_node) %}
+
+    {{ return(adapter.dispatch('__get_test_description', 'dq_tools')(test_node)) }}
+
+{% endmacro %}
+
+{% macro default____get_test_description(test_node) %}
+
+    {% if test_node.description is defined and test_node.description %}
+
+        {{ return(test_node.description) }}
+    
+    {% elif dq_tools.__get_test_type(test_node) != 'singular' and var("dq_tools__auto_generate_test_description", "0") == "1" %}
+
+        {%- set test_name = dq_tools.__get_test_name(test_node) -%}
+        {%- set column_name = dq_tools.__get_column_name(test_node) -%}
+        {%- set testing_model = dq_tools.__get_test_model(test_node) -%}
+        {%- set model_name = dq_tools.__get_relation(testing_model).name | lower -%}
+
+        {%- set generated_description = '' -%}
+
+        {%- if test_name == 'unique' -%}
+            {%- set generated_description = 'The ' ~ column_name ~ ' column in the ' ~ model_name ~ ' model should be unique.' -%}
+        {%- elif test_name == 'not_null' -%}
+            {%- set generated_description = 'The ' ~ column_name ~ ' column in the ' ~ model_name ~ ' model should not contain null values.' -%}
+        {%- elif test_name == 'accepted_values' -%}
+            {%- set accepted_values = test_node.test_metadata.kwargs['values'] | join(', ') -%}
+            {%- set generated_description = 'The ' ~ column_name ~ ' column in the ' ~ model_name ~ ' should be one of ' ~ accepted_values ~ ' values.' -%}
+        {%- elif test_name == 'relationships' -%}
+            {%- set to_model = test_node.test_metadata.kwargs.to -%}
+            {%- set related_model = to_model.split('\'')[1].strip() -%}
+            {%- set generated_description = 'Each ' ~ column_name ~ ' in the ' ~ model_name ~ ' model exists as an id in the ' ~ related_model ~ ' table.' -%}
+        {%- endif -%}
+
+        {{ return(generated_description) }}
+
+    {% endif %}
+
+{% endmacro %}

macros/artifacts/test/utilities/general/__select_test_result.sql

@@ -9,6 +9,8 @@
   {%- set testing_model = dq_tools.__get_test_model(result.node) -%}
   {%- set testing_model_relation = dq_tools.__get_relation(testing_model) -%}
   {%- set materialization = dq_tools.__get_test_model_materialization(testing_model.name) -%}
+  {%- set test_description = dq_tools.__get_test_description(result.node) -%}
+
   /* {{ testing_model }} */
 
   select   '{{ result.node.unique_id }}' as test_unique_id
@@ -42,6 +44,7 @@
             ){% else %}null{% endif %} as no_of_records_scanned
           ,coalesce({{ result.failures or 'null' }}, 0) as no_of_records_failed
           ,'{{ test_type }}' as test_type
+          ,'{{ test_description }}' as test_description
           ,'{{ result.execution_time }}' as execution_time_seconds
           ,'{{ result.node.original_file_path }}' as file_test_defined
           ,'{{ target.name }}' as dbt_target_name

macros/artifacts/test/utilities/general/_general.yml

@@ -50,6 +50,9 @@ macros:
         General type of the test which are singular or generic.
         See {{'{{'}} __get_test_type {{'}}'}}
 
+      - `test_description`:
+        Description of a test.
+
       - `execution_time_seconds`:
         Value generated by dbt which is `result.execution_time`
 
@@ -114,3 +117,16 @@ macros:
   - name: __get_test_model_materialization
     description: |
       Return the materialization type of the model being tested (eg: table, view, ephemeral etc).
+  
+
+  - name: __get_test_description
+    description: |
+      This macro retrieves the description of a test from a test node object.
+      If the description is not defined, it generates one based on the test type (e.g., unique, not_null, accepted_values, or relationships). 
+      Available for tests starting in dbt v1.9. 
+      Returns an empty string if no description is found or generated.
+    arguments:
+      - name: test_node
+        type: string
+        description: The unique identifier of the test node in the graph
+    return_value: (string) The test description or an empty string if none is found.

models/01_lake/dq_issue_log.sql

@@ -30,6 +30,7 @@ select   cast(null as {{ dbt.type_timestamp() }}) as check_timestamp
         ,cast(null as {{ dbt.type_int() }}) as no_of_table_columns
         ,cast(null as {{ dbt.type_int() }}) as no_of_tables
         ,cast(null as {{ dbt.type_string() }}) as test_unique_id
+        ,cast(null as {{ dbt.type_string() }}) as test_description
 
 from    dummy