Merge pull request #72 from get-select/fix_query_text_no_comments

Fix query_text_no_comments

.changes/1.5.0.md

@@ -0,0 +1,7 @@
+## dbt-snowflake-monitoring 1.5.0 - January 25, 2023
+
+### Fixes
+
+- Fix query_text_no_comments regex ([#72](https://github.com/get-select/dbt-snowflake-monitoring/pull/72))
+
+

CHANGELOG.md

@@ -5,6 +5,14 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html),
 and is generated by [Changie](https://github.com/miniscruff/changie).
 
+## dbt-snowflake-monitoring 1.5.0 - January 25, 2023
+
+### Fixes
+
+- Fix query_text_no_comments regex ([#72](https://github.com/get-select/dbt-snowflake-monitoring/pull/72))
+
+
+
 ## dbt-snowflake-monitoring 1.4.3 - January 20, 2023
 
 ### Fixes

README.md

@@ -15,7 +15,7 @@ Add the following to your `packages.yml` file:
 ```yaml
 packages:
   - package: get-select/dbt_snowflake_monitoring
-    version: 1.4.2
+    version: 1.5.0
 ```
 
 To attribute costs to individual models via the `dbt_metadata` column in the `query_history_enriched` model, add the following to `dbt_project.yml`:
@@ -29,7 +29,7 @@ query-comment:
 The generate URLs to dbt Cloud jobs and runs in the `dbt_queries` model, add the following variable to `dbt_project.yml`:
 ```yaml
 vars:
-  dbt_cloud_account_id: 12345 # https://cloud.getdbt.com/deploy/<this_number>/projects/<not_this_number>/jobs 
+  dbt_cloud_account_id: 12345 # https://cloud.getdbt.com/deploy/<this_number>/projects/<not_this_number>/jobs
 ```
 
 ### Only want to use the get_query_comment macro?

dbt_project.yml

@@ -1,5 +1,5 @@
 name: 'dbt_snowflake_monitoring'
-version: '1.4.3'
+version: '1.5.0'
 config-version: 2
 
 profile: dbt_snowflake_monitoring

macros/create_regex_udf.sql

@@ -0,0 +1,14 @@
+{% macro create_regexp_replace_udf() %}
+
+create or replace function {{ target.database }}.{{ target.schema }}.dbt_snowflake_monitoring_regexp_replace(subject text, pattern text, replacement text)
+returns string
+language javascript
+comment = 'Created by dbt-snowflake-monitoring dbt package.'
+as
+$$
+    const p = SUBJECT;
+    let regex = new RegExp(PATTERN, 'g')
+    return p.replace(regex, REPLACEMENT);
+$$
+
+{% endmacro %}

models/query_history_enriched.sql

@@ -1,14 +1,16 @@
-{{ config(materialized='incremental', unique_key=['query_id', 'start_time']) }}
+{{ config(
+    materialized='incremental',
+    unique_key=['query_id', 'start_time'],
+    pre_hook="{{ create_regexp_replace_udf() }}"
+) }}
 
 with
 query_history as (
     select
         *,
 
-        -- this removes comments enclosed by /* <comment text> */
-        regexp_replace(query_text, '(/\*.*\*/)') as _query_text_no_comments,
-        -- this removes single line comments starting with -- and either ending with a new line or end of string
-        regexp_replace(_query_text_no_comments, '(--.*$)|(--.*\n)') as query_text_no_comments,
+        -- this removes comments enclosed by /* <comment text> */ and single line comments starting with -- and either ending with a new line or end of string
+        {{ target.database }}.{{ target.schema }}.dbt_snowflake_monitoring_regexp_replace(query_text, $$(/\*(.|\n)+?\*/)|(--.*$)|(--.*\n)$$, '') as query_text_no_comments,
 
         regexp_substr(query_text, '/\\*\\s({"app":\\s"dbt".*})\\s\\*/', 1, 1, 'ie') as _dbt_json_meta,
         try_parse_json(_dbt_json_meta) as dbt_metadata