expected texts in top values sensor updated - postgresql, redshift

Author: Patryk Jatczak

home/sensors/column/accepted_values/expected_texts_in_top_values_count/postgresql.sql.jinja2

@@ -18,9 +18,9 @@ FROM
         {% if lib.time_series is not none -%}
         top_col_values.time_period as time_period,
         top_col_values.time_period_utc as time_period_utc,
-        {%- endif -%}
+        {% endif -%}
         RANK() OVER({{- render_data_grouping('top_col_values', indentation = ' ', partition_by_enabled=true) }}
-            ORDER BY top_col_values.total_values DESC) as top_values_rank  {{- render_data_grouping('top_col_values', indentation = ' ') }}
+            ORDER BY top_col_values.total_values DESC) as top_values_rank {{- render_data_grouping('top_col_values', indentation = ' ') }}
     FROM
     (
         SELECT
@@ -31,30 +31,32 @@ FROM
         FROM
             {{ lib.render_target_table() }} AS analyzed_table
         {{- lib.render_where_clause(extra_filter = lib.render_target_column('analyzed_table') ~ ' IS NOT NULL', indentation = '        ') }}
-        GROUP BY {{ render_grouping_columns() }} top_value
-                ORDER BY {{ render_grouping_columns() }} total_values DESC
+        GROUP BY {{ render_grouping_columns() -}} top_value
+        ORDER BY {{ render_grouping_columns() -}} total_values DESC
     ) AS top_col_values
 ) AS top_values
 WHERE top_values_rank <= {{ parameters.top }}
 {%- endmacro -%}
 
 {% macro render_grouping_columns() %}
-    {%- if (lib.data_groupings is not none and (lib.data_groupings | length()) > 0) or lib.time_series is not none %}
-        {{ lib.render_grouping_column_names() }} {{ ',' }}
+    {%- if (lib.data_groupings is not none and (lib.data_groupings | length()) > 0) or lib.time_series is not none -%}
+        {{ lib.render_grouping_column_names() }} {{- ', ' -}}
     {%- endif -%}
 {% endmacro %}
 
 {%- macro render_data_grouping(table_alias_prefix = '', indentation = '', partition_by_enabled = false) -%}
 
-    {%- if lib.time_series is not none and partition_by_enabled == true -%}PARTITION BY top_col_values.time_period{%- endif -%}
+    {%- if partition_by_enabled == true -%}PARTITION BY
+        {%- if lib.time_series is not none -%}
+            {{" "}}top_col_values.time_period
+        {%- elif lib.data_groupings is none -%}
+            {{" "}}NULL
+        {%- endif -%}
+    {%- endif -%}
 
     {%- if lib.data_groupings is not none and (lib.data_groupings | length()) > 0 -%}
         {%- for attribute in lib.data_groupings -%}
-            {% if lib.time_series is none and loop.index == 1 and partition_by_enabled == true -%}
-                PARTITION BY
-            {%- else  -%}
-                {{ ',' }}
-            {%- endif -%}
+            {{- "" if loop.first and lib.time_series is none and partition_by_enabled else "," -}}
             {%- with data_grouping_level = lib.data_groupings[attribute] -%}
                 {%- if data_grouping_level.source == 'tag' -%}
                     {{ indentation }}{{ lib.make_text_constant(data_grouping_level.tag) }}
@@ -73,17 +75,17 @@ SELECT
     {{- lib.render_data_grouping_projections('analyzed_table') }}
     {{- lib.render_time_dimension_projection('analyzed_table') }}
 FROM {{ lib.render_target_table() }} AS analyzed_table
-    {%- else %}
+{%- else %}
     COUNT(DISTINCT
         CASE
             WHEN top_values.top_value IN ({{ extract_in_list(parameters.expected_values) }}) THEN top_values.top_value
             ELSE NULL
         END
     ) AS actual_value,
     MAX({{ parameters.expected_values | length }}) AS expected_value
-    {% if lib.time_series is not none -%}
-    , top_values.time_period
-    , top_values.time_period_utc
+    {%- if lib.time_series is not none -%} {{- "," }}
+    top_values.time_period,
+    top_values.time_period_utc
     {%- endif -%}
     {{- render_data_grouping('top_values', indentation = lib.eol() ~ '    ') }}
 {{ render_from_subquery() }}

home/sensors/column/accepted_values/expected_texts_in_top_values_count/redshift.sql.jinja2

@@ -18,9 +18,9 @@ FROM
         {% if lib.time_series is not none -%}
         top_col_values.time_period as time_period,
         top_col_values.time_period_utc as time_period_utc,
-        {%- endif -%}
+        {% endif -%}
         RANK() OVER({{- render_data_grouping('top_col_values', indentation = ' ', partition_by_enabled=true) }}
-            ORDER BY top_col_values.total_values DESC) as top_values_rank  {{- render_data_grouping('top_col_values', indentation = ' ') }}
+            ORDER BY top_col_values.total_values DESC) as top_values_rank {{- render_data_grouping('top_col_values', indentation = ' ') }}
     FROM
     (
         SELECT
@@ -31,30 +31,32 @@ FROM
         FROM
             {{ lib.render_target_table() }} AS analyzed_table
         {{- lib.render_where_clause(extra_filter = lib.render_target_column('analyzed_table') ~ ' IS NOT NULL', indentation = '        ') }}
-        GROUP BY {{ render_grouping_columns() }} top_value
-        ORDER BY {{ render_grouping_columns() }} total_values DESC
+        GROUP BY {{ render_grouping_columns() -}} top_value
+        ORDER BY {{ render_grouping_columns() -}} total_values DESC
     ) AS top_col_values
 ) AS top_values
 WHERE top_values_rank <= {{ parameters.top }}
 {%- endmacro -%}
 
 {% macro render_grouping_columns() %}
-    {%- if (lib.data_groupings is not none and (lib.data_groupings | length()) > 0) or lib.time_series is not none %}
-        {{ lib.render_grouping_column_names() }} {{ ',' }}
+    {%- if (lib.data_groupings is not none and (lib.data_groupings | length()) > 0) or lib.time_series is not none -%}
+        {{ lib.render_grouping_column_names() }} {{- ', ' -}}
     {%- endif -%}
 {% endmacro %}
 
 {%- macro render_data_grouping(table_alias_prefix = '', indentation = '', partition_by_enabled = false) -%}
 
-    {%- if lib.time_series is not none and partition_by_enabled == true -%}PARTITION BY top_col_values.time_period{%- endif -%}
+    {%- if partition_by_enabled == true -%}PARTITION BY
+        {%- if lib.time_series is not none -%}
+            {{" "}}top_col_values.time_period
+        {%- elif lib.data_groupings is none -%}
+            {{" "}}NULL
+        {%- endif -%}
+    {%- endif -%}
 
     {%- if lib.data_groupings is not none and (lib.data_groupings | length()) > 0 -%}
         {%- for attribute in lib.data_groupings -%}
-            {% if lib.time_series is none and loop.index == 1 and partition_by_enabled == true -%}
-                PARTITION BY
-            {%- else  -%}
-                {{ ',' }}
-            {%- endif -%}
+            {{- "" if loop.first and lib.time_series is none and partition_by_enabled else "," -}}
             {%- with data_grouping_level = lib.data_groupings[attribute] -%}
                 {%- if data_grouping_level.source == 'tag' -%}
                     {{ indentation }}{{ lib.make_text_constant(data_grouping_level.tag) }}
@@ -73,17 +75,17 @@ SELECT
     {{- lib.render_data_grouping_projections('analyzed_table') }}
     {{- lib.render_time_dimension_projection('analyzed_table') }}
 FROM {{ lib.render_target_table() }} AS analyzed_table
-    {%- else %}
+{%- else %}
     COUNT(DISTINCT
         CASE
             WHEN top_values.top_value IN ({{ extract_in_list(parameters.expected_values) }}) THEN top_values.top_value
             ELSE NULL
         END
     ) AS actual_value,
     MAX({{ parameters.expected_values | length }}) AS expected_value
-    {% if lib.time_series is not none -%}
-    , top_values.time_period
-    , top_values.time_period_utc
+    {%- if lib.time_series is not none -%} {{- "," }}
+    top_values.time_period,
+    top_values.time_period_utc
     {%- endif -%}
     {{- render_data_grouping('top_values', indentation = lib.eol() ~ '    ') }}
 {{ render_from_subquery() }}