diff --git a/README.md b/README.md index 10cd87e..70c47b9 100644 --- a/README.md +++ b/README.md @@ -52,3 +52,46 @@ dbt run --var '{"UPDATE_UDFS_AND_SPS":True}' -m ... - Join the [chat](https://community.getdbt.com/) on Slack for live discussions and support - Find [dbt events](https://events.getdbt.com) near you - Check out [the blog](https://blog.getdbt.com/) for the latest news on dbt's development and best practices + + +## Applying Model Tags + +### Database / Schema level tags + +Database and schema tags are applied via the `add_database_or_schema_tags` macro. These tags are inherited by their downstream objects. To add/modify tags call the appropriate tag set function within the macro. + +``` +{{ set_database_tag_value('SOME_DATABASE_TAG_KEY','SOME_DATABASE_TAG_VALUE') }} +{{ set_schema_tag_value('SOME_SCHEMA_TAG_KEY','SOME_SCHEMA_TAG_VALUE') }} +``` + +### Model tags + +To add/update a model's snowflake tags, add/modify the `meta` model property under `config`. Only table level tags are supported at this time via DBT. + +``` +{{ config( + ..., + meta={ + 'database_tags':{ + 'table': { + 'PURPOSE': 'SOME_PURPOSE' + } + } + }, + ... +) }} +``` + +By default, model tags are pushed to Snowflake on each load. You can disable this by setting the `UPDATE_SNOWFLAKE_TAGS` project variable to `False` during a run. + +``` +dbt run --var '{"UPDATE_SNOWFLAKE_TAGS":False}' -s models/core/core__fact_blocks.sql +``` + +### Querying for existing tags on a model in snowflake + +``` +select * +from table(arbitrum.information_schema.tag_references('arbitrum.core.fact_blocks', 'table')); +``` \ No newline at end of file diff --git a/dbt_project.yml b/dbt_project.yml index b683d90..a0e6e97 100644 --- a/dbt_project.yml +++ b/dbt_project.yml @@ -30,7 +30,9 @@ tests: on-run-start: - "{{ create_sps() }}" - "{{ create_udfs() }}" - + + on-run-end: + - '{{ apply_meta_as_tags(results) }}' # Configuring models # Full documentation: https://docs.getdbt.com/docs/configuring-models @@ -42,3 +44,4 @@ on-run-start: vars: "dbt_date:time_zone": GMT UPDATE_UDFS_AND_SPS: False + UPDATE_SNOWFLAKE_TAGS: True \ No newline at end of file diff --git a/macros/tags/add_database_or_schema_tags.sql b/macros/tags/add_database_or_schema_tags.sql new file mode 100644 index 0000000..9c9f533 --- /dev/null +++ b/macros/tags/add_database_or_schema_tags.sql @@ -0,0 +1,4 @@ +{% macro add_database_or_schema_tags() %} + {{ set_database_tag_value('BLOCKCHAIN_NAME','ARBITRUM') }} + {{ set_database_tag_value('BLOCKCHAIN_TYPE','EVM, L2') }} +{% endmacro %} \ No newline at end of file diff --git a/macros/tags/snowflake_tagging.sql b/macros/tags/snowflake_tagging.sql new file mode 100644 index 0000000..bc25e69 --- /dev/null +++ b/macros/tags/snowflake_tagging.sql @@ -0,0 +1,127 @@ +{% macro apply_meta_as_tags(results) %} + {% if var("UPDATE_SNOWFLAKE_TAGS") %} + {{ log('apply_meta_as_tags', info=False) }} + {{ log(results, info=False) }} + {% if execute %} + + {%- set tags_by_schema = {} -%} + {% for res in results -%} + {% if res.node.meta.database_tags %} + + {%- set model_database = res.node.database -%} + {%- set model_schema = res.node.schema -%} + {%- set model_schema_full = model_database+'.'+model_schema -%} + {%- set model_alias = res.node.alias -%} + + {% if model_schema_full not in tags_by_schema.keys() %} + {{ log('need to fetch tags for schema '+model_schema_full, info=False) }} + {%- call statement('main', fetch_result=True) -%} + show tags in {{model_database}}.{{model_schema}} + {%- endcall -%} + {%- set _ = tags_by_schema.update({model_schema_full: load_result('main')['table'].columns.get('name').values()|list}) -%} + {{ log('Added tags to cache', info=False) }} + {% else %} + {{ log('already have tag info for schema', info=False) }} + {% endif %} + + {%- set current_tags_in_schema = tags_by_schema[model_schema_full] -%} + {{ log('current_tags_in_schema:', info=False) }} + {{ log(current_tags_in_schema, info=False) }} + {{ log("========== Processing tags for "+model_schema_full+"."+model_alias+" ==========", info=False) }} + + {% set line -%} + node: {{ res.node.unique_id }}; status: {{ res.status }} (message: {{ res.message }}) + node full: {{ res.node}} + meta: {{ res.node.meta}} + materialized: {{ res.node.config.materialized }} + {%- endset %} + {{ log(line, info=False) }} + + {%- call statement('main', fetch_result=True) -%} + select LEVEL,UPPER(TAG_NAME) as TAG_NAME,TAG_VALUE from table(information_schema.tag_references_all_columns('{{model_schema}}.{{model_alias}}', 'table')) + {%- endcall -%} + {%- set existing_tags_for_table = load_result('main')['data'] -%} + {{ log('Existing tags for table:', info=False) }} + {{ log(existing_tags_for_table, info=False) }} + + {{ log('--', info=False) }} + {% for table_tag in res.node.meta.database_tags.table %} + + {{ create_tag_if_missing(current_tags_in_schema,table_tag|upper) }} + {% set desired_tag_value = res.node.meta.database_tags.table[table_tag] %} + + {{set_table_tag_value_if_different(model_schema,model_alias,table_tag,desired_tag_value,existing_tags_for_table)}} + {% endfor %} + {{ log("========== Finished processing tags for "+model_alias+" ==========", info=False) }} + {% endif %} + {% endfor %} + {% endif %} + {% endif %} +{% endmacro %} + + +{% macro create_tag_if_missing(all_tag_names,table_tag) %} + {% if table_tag not in all_tag_names %} + {{ log('Creating missing tag '+table_tag, info=False) }} + {%- call statement('main', fetch_result=True) -%} + create tag if not exists silver.{{table_tag}} + {%- endcall -%} + {{ log(load_result('main').data, info=False) }} + {% else %} + {{ log('Tag already exists: '+table_tag, info=False) }} + {% endif %} +{% endmacro %} + +{% macro set_table_tag_value_if_different(model_schema,table_name,tag_name,desired_tag_value,existing_tags) %} + {{ log('Ensuring tag '+tag_name+' has value '+desired_tag_value+' at table level', info=False) }} + {%- set existing_tag_for_table = existing_tags|selectattr('0','equalto','TABLE')|selectattr('1','equalto',tag_name|upper)|list -%} + {{ log('Filtered tags for table:', info=False) }} + {{ log(existing_tag_for_table[0], info=False) }} + {% if existing_tag_for_table|length > 0 and existing_tag_for_table[0][2]==desired_tag_value %} + {{ log('Correct tag value already exists', info=False) }} + {% else %} + {{ log('Setting tag value for '+tag_name+' to value '+desired_tag_value, info=False) }} + {%- call statement('main', fetch_result=True) -%} + alter table {{model_schema}}.{{table_name}} set tag {{tag_name}} = '{{desired_tag_value}}' + {%- endcall -%} + {{ log(load_result('main').data, info=False) }} + {% endif %} +{% endmacro %} + +{% macro set_column_tag_value_if_different(table_name,column_name,tag_name,desired_tag_value,existing_tags) %} + {{ log('Ensuring tag '+tag_name+' has value '+desired_tag_value+' at column level', info=False) }} + {%- set existing_tag_for_column = existing_tags|selectattr('0','equalto','COLUMN')|selectattr('1','equalto',tag_name|upper)|list -%} + {{ log('Filtered tags for column:', info=False) }} + {{ log(existing_tag_for_column[0], info=False) }} + {% if existing_tag_for_column|length > 0 and existing_tag_for_column[0][2]==desired_tag_value %} + {{ log('Correct tag value already exists', info=False) }} + {% else %} + {{ log('Setting tag value for '+tag_name+' to value '+desired_tag_value, info=False) }} + {%- call statement('main', fetch_result=True) -%} + alter table {{table_name}} modify column {{column_name}} set tag {{tag_name}} = '{{desired_tag_value}}' + {%- endcall -%} + {{ log(load_result('main').data, info=False) }} + {% endif %} +{% endmacro %} + +{% macro set_database_tag_value(tag_name,tag_value) %} + {% set query %} + create tag if not exists silver.{{tag_name}} + {% endset %} + {% do run_query(query) %} + {% set query %} + alter database {{target.database}} set tag {{target.database}}.silver.{{tag_name}} = '{{tag_value}}' + {% endset %} + {% do run_query(query) %} +{% endmacro %} + +{% macro set_schema_tag_value(target_schema,tag_name,tag_value) %} + {% set query %} + create tag if not exists silver.{{tag_name}} + {% endset %} + {% do run_query(query) %} + {% set query %} + alter schema {{target.database}}.{{target_schema}} set tag {{target.database}}.silver.{{tag_name}} = '{{tag_value}}' + {% endset %} + {% do run_query(query) %} +{% endmacro %} \ No newline at end of file diff --git a/models/sushi/sushi__dim_dex_pools.sql b/models/sushi/sushi__dim_dex_pools.sql index 17e9859..e31a5ae 100644 --- a/models/sushi/sushi__dim_dex_pools.sql +++ b/models/sushi/sushi__dim_dex_pools.sql @@ -1,5 +1,13 @@ {{ config( - materialized = 'table' + materialized = 'table', + meta={ + 'database_tags':{ + 'table': { + 'PROTOCOL': 'SUSHI', + 'PURPOSE': 'DEFI, DEX' + } + } + } ) }} SELECT diff --git a/models/sushi/sushi__dim_kashi_pairs.sql b/models/sushi/sushi__dim_kashi_pairs.sql index 97e09a1..c7a9514 100644 --- a/models/sushi/sushi__dim_kashi_pairs.sql +++ b/models/sushi/sushi__dim_kashi_pairs.sql @@ -1,5 +1,13 @@ {{ config( - materialized = 'table' + materialized = 'table', + meta={ + 'database_tags':{ + 'table': { + 'PROTOCOL': 'SUSHI', + 'PURPOSE': 'DEFI, DEX' + } + } + } ) }} SELECT diff --git a/models/sushi/sushi__ez_borrowing.sql b/models/sushi/sushi__ez_borrowing.sql index 0c215d0..2fc8ef0 100644 --- a/models/sushi/sushi__ez_borrowing.sql +++ b/models/sushi/sushi__ez_borrowing.sql @@ -4,7 +4,15 @@ persist_docs ={ "relation": true, "columns": true }, unique_key = '_log_id', - cluster_by = ['block_timestamp::DATE'] + cluster_by = ['block_timestamp::DATE'], + meta={ + 'database_tags':{ + 'table': { + 'PROTOCOL': 'SUSHI', + 'PURPOSE': 'DEFI, DEX' + } + } + } ) }} WITH borrow_txns AS ( diff --git a/models/sushi/sushi__ez_lending.sql b/models/sushi/sushi__ez_lending.sql index 0472188..2363d07 100644 --- a/models/sushi/sushi__ez_lending.sql +++ b/models/sushi/sushi__ez_lending.sql @@ -4,7 +4,15 @@ persist_docs ={ "relation": true, "columns": true }, unique_key = '_log_id', - cluster_by = ['block_timestamp::DATE'] + cluster_by = ['block_timestamp::DATE'], + meta={ + 'database_tags':{ + 'table': { + 'PROTOCOL': 'SUSHI', + 'PURPOSE': 'DEFI, DEX' + } + } + } ) }} WITH borrow_txns AS ( diff --git a/models/sushi/sushi__ez_swaps.sql b/models/sushi/sushi__ez_swaps.sql index 287b55f..6a78f47 100644 --- a/models/sushi/sushi__ez_swaps.sql +++ b/models/sushi/sushi__ez_swaps.sql @@ -3,7 +3,15 @@ persist_docs ={ "relation": true, "columns": true }, unique_key = '_log_id', - cluster_by = ['block_timestamp::DATE'] + cluster_by = ['block_timestamp::DATE'], + meta={ + 'database_tags':{ + 'table': { + 'PROTOCOL': 'SUSHI', + 'PURPOSE': 'DEFI, DEX, SWAPS' + } + } + } ) }} WITH swap_events AS (