From b3d2e12582ae33f0587ea7b63e4cc51b8b45f7d8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A9r=C3=A9my=20Guiselin?=
 <9251353+Jrmyy@users.noreply.github.com>
Date: Mon, 20 May 2024 14:23:07 +0200
Subject: [PATCH] fix: test_not_null_proportion support on athena (#19)

---
 .../generic_tests/not_null_proportion.sql     | 30 +++++++++++++++++++
 1 file changed, 30 insertions(+)
 create mode 100644 macros/dbt_utils/generic_tests/not_null_proportion.sql

diff --git a/macros/dbt_utils/generic_tests/not_null_proportion.sql b/macros/dbt_utils/generic_tests/not_null_proportion.sql
new file mode 100644
index 0000000..3f9f459
--- /dev/null
+++ b/macros/dbt_utils/generic_tests/not_null_proportion.sql
@@ -0,0 +1,30 @@
+{% macro athena__test_not_null_proportion(model, group_by_columns) %}
+
+{% set column_name = kwargs.get('column_name', kwargs.get('arg')) %}
+{% set at_least = kwargs.get('at_least', kwargs.get('arg')) %}
+{% set at_most = kwargs.get('at_most', kwargs.get('arg', 1)) %}
+
+{% if group_by_columns|length() > 0 %}
+  {% set select_gb_cols = group_by_columns|join(' ,') + ', ' %}
+  {% set groupby_gb_cols = 'group by ' + group_by_columns|join(',') %}
+{% endif %}
+
+with validation as (
+  select
+    {{select_gb_cols}}
+    sum(case when {{ column_name }} is null then 0 else 1 end) / cast(count(*) as double) as not_null_proportion
+  from {{ model }}
+  {{groupby_gb_cols}}
+),
+validation_errors as (
+  select
+    {{select_gb_cols}}
+    not_null_proportion
+  from validation
+  where not_null_proportion < {{ at_least }} or not_null_proportion > {{ at_most }}
+)
+select
+  *
+from validation_errors
+
+{% endmacro %}