udfs: norma exclusion petits mots, cleanup

Author: maxcorbeau

dbt/macros/udf/udf_ae_string_cleanup.sql

@@ -6,7 +6,7 @@
     to NULL for easier processing whenever we consider
     it to be empty.
 */
-DROP FUNCTION IF EXISTS {{ target.schema }}.udf_ae_string_cleanup(TEXT) CASCADE;
+DROP FUNCTION IF EXISTS {{ target.schema }}.udf_ae_string_cleanup CASCADE;
 CREATE FUNCTION {{ target.schema }}.udf_ae_string_cleanup(val TEXT)
 RETURNS TEXT AS $$
 BEGIN

dbt/macros/udf/udf_columns_concat_unique_non_empty.sql

@@ -2,7 +2,7 @@
 /*
     Concatenate strings from various columns while only retaining non-empty values
 */
-DROP FUNCTION IF EXISTS {{ target.schema }}.udf_columns_concat_unique_non_empty(VARIADIC input_columns TEXT[]) CASCADE;
+DROP FUNCTION IF EXISTS {{ target.schema }}.udf_columns_concat_unique_non_empty CASCADE;
 CREATE FUNCTION {{ target.schema }}.udf_columns_concat_unique_non_empty(VARIADIC input_columns TEXT[])
 RETURNS TEXT AS $$
 DECLARE

dbt/macros/udf/udf_columns_words_in_common_count.sql

@@ -2,7 +2,7 @@
 /*
     Count number of words in common between 2 columns
 */
-DROP FUNCTION IF EXISTS {{ target.schema }}.udf_columns_words_in_common_count(text, text) CASCADE;
+DROP FUNCTION IF EXISTS {{ target.schema }}.udf_columns_words_in_common_count CASCADE;
 CREATE FUNCTION {{ target.schema }}.udf_columns_words_in_common_count(col1 text, col2 text)
 RETURNS integer AS $$
 DECLARE

dbt/macros/udf/udf_encode_base57.sql

@@ -1,6 +1,6 @@
 {% macro create_udf_encode_base57() %}
 
-DROP FUNCTION IF EXISTS {{ target.schema }}.encode_base57(uuid) CASCADE;
+DROP FUNCTION IF EXISTS {{ target.schema }}.encode_base57 CASCADE;
 CREATE FUNCTION {{ target.schema }}.encode_base57(uuid UUID)
 RETURNS text AS $$
 DECLARE

dbt/macros/udf/udf_normalize_string_alpha_for_match.sql

@@ -9,8 +9,8 @@
     E.g. to test this function:
     SELECT udf_normalize_string_for_match(' Héllo-Wørld! Ça va? 123 ');
  */
-DROP FUNCTION IF EXISTS {{ target.schema }}.udf_normalize_string_for_match(TEXT) CASCADE;
-CREATE FUNCTION {{ target.schema }}.udf_normalize_string_for_match(input_text TEXT) RETURNS TEXT AS $$
+DROP FUNCTION IF EXISTS {{ target.schema }}.udf_normalize_string_for_match CASCADE;
+CREATE FUNCTION {{ target.schema }}.udf_normalize_string_for_match(input_text TEXT, remove_words_smaller_size INTEGER DEFAULT 2) RETURNS TEXT AS $$
 DECLARE
     normalized TEXT;
     words TEXT[];
@@ -28,7 +28,8 @@ BEGIN
     FROM (
         SELECT unnest(words) AS word
         ORDER BY word
-    ) AS words_sorted;
+    ) AS words_sorted
+    WHERE length(word) >= remove_words_smaller_size;
 
     RETURN normalized;
 END;

dbt/macros/udf/udf_safe_divmod.sql

@@ -1,5 +1,5 @@
 {% macro create_udf_safe_divmod() %}
-DROP FUNCTION IF EXISTS {{ target.schema }}.safe_divmod(numeric, numeric) CASCADE;
+DROP FUNCTION IF EXISTS {{ target.schema }}.safe_divmod CASCADE;
 CREATE FUNCTION {{ target.schema }}.safe_divmod(n numeric, d numeric)
 RETURNS TABLE(quotient numeric, remainder numeric) AS $$
 DECLARE

dbt/macros/udf/udf_uuid_to_int.sql

@@ -1,5 +1,5 @@
 {% macro create_udf_uuid_to_int() %}
-DROP FUNCTION IF EXISTS {{ target.schema }}.uuid_to_int(uuid) CASCADE;
+DROP FUNCTION IF EXISTS {{ target.schema }}.uuid_to_int CASCADE;
 CREATE FUNCTION {{ target.schema }}.uuid_to_int(uuid UUID)
 RETURNS numeric AS $$
 DECLARE