Merge pull request #54 from fivetran/MagicBot/add-union-schema

Feature: Union schema compatibility

.buildkite/pipeline.yml

@@ -58,7 +58,7 @@ steps:
     commands: |
       bash .buildkite/scripts/run_models.sh redshift
 
-  - label: ":bricks: Run Tests - Databricks"
+  - label: ":databricks: Run Tests - Databricks"
     key: "run_dbt_databricks"
     plugins:
       - docker#v3.13.0:

CHANGELOG.md

@@ -1,9 +1,35 @@
-# dbt_linkedin_source v0.UPDATE.UPDATE
+# dbt_linkedin_source v0.8.0
+[PR #54](https://github.com/fivetran/dbt_linkedin_source/pull/54) includes the following updates:
 
- ## Under the Hood:
+## Breaking changes
+- Updated materializations of non-`tmp` staging models from views to tables. This is to bring the materializations into alignment with other ad reporting packages and eliminate errors in Redshift. 
+- Updated the name of the source created by this package from `linkedin` to `linkedin_ads`. This was to bring the naming used in this package in alignment with our other ad packages and for compatibility with the union schema feature.
+  - ❗ If you are using this source, you will need to update the name.
+- Updated the following identifiers for consistency with the source name and compatibility with the union schema feature:
 
+| current  | previous |
+|----------|----------|
+| linkedin_ads_account_history_identifier | linkedin_account_history_identifier
+| linkedin_ads_ad_analytics_by_creative_identifier | linkedin_ad_analytics_by_creative_identifier
+| linkedin_ads_campaign_group_history_identifier | linkedin_campaign_group_history_identifier
+| linkedin_ads_campaign_history_identifier | linkedin_campaign_history_identifier
+| linkedin_ads_creative_history_identifier | linkedin_creative_history_identifier
+| linkedin_ads_ad_analytics_by_campaign_identifier | linkedin_ad_analytics_by_campaign_identifier
+
+- If you are using the previous identifier, be sure to update to the current version!
+
+## Feature update 🎉
+- Unioning capability! This adds the ability to union source data from multiple linkedin connectors. Refer to the [Union Multiple Connectors README section](https://github.com/fivetran/dbt_linkedin_source/blob/main/README.md#union-multiple-connectors) for more details.
+
+## Under the hood 🚘
+- Updated tmp models to union source data using the `fivetran_utils.union_data` macro. 
+- To distinguish which source each field comes from, added `source_relation` column in each staging model and applied the `fivetran_utils.source_relation` macro.
+- Updated tests to account for the new `source_relation` column.
+
+[PR #51](https://github.com/fivetran/dbt_linkedin_source/pull/51) includes the following updates:
 - Incorporated the new `fivetran_utils.drop_schemas_automation` macro into the end of each Buildkite integration test job.
 - Updated the pull request [templates](/.github).
+
 # dbt_linkedin_source v0.7.0
 ## 🚨 Breaking Changes 🚨
 Due to Linkedin Ads API [change in January 2023](https://learn.microsoft.com/en-us/linkedin/marketing/integrations/recent-changes?view=li-lms-2022-12#january-2023), there have been updates in the Linkedin Ads Fivetran Connector and therefore, updates to this Linkedin package. 

README.md

@@ -47,7 +47,7 @@ If you  are **not** using the [Linkedin transformation package](https://github.c
 # packages.yml
 packages:
   - package: fivetran/linkedin_source
-    version: [">=0.7.0", "<0.8.0"]
+    version: [">=0.8.0", "<0.9.0"]
 ```
 
 ## Step 3: Define database and schema variables
@@ -61,7 +61,17 @@ vars:
 ```
 
 ## (Optional) Step 4: Additional configurations
-<details><summary>Expand for configurations</summary>
+### Union multiple connectors
+If you have multiple linkedin connectors in Fivetran and would like to use this package on all of them simultaneously, we have provided functionality to do so. The package will union all of the data together and pass the unioned table into the transformations. You will be able to see which source it came from in the `source_relation` column of each model. To use this functionality, you will need to set either the `linkedin_ads_union_schemas` OR `linkedin_ads_union_databases` variables (cannot do both) in your root `dbt_project.yml` file:
+
+```yml
+vars:
+    linkedin_ads_union_schemas: ['linkedin_usa','linkedin_canada'] # use this if the data is in different schemas/datasets of the same database/project
+    linkedin_ads_union_databases: ['linkedin_usa','linkedin_canada'] # use this if the data is in different databases/projects but uses the same schema name
+```
+Please be aware that the native `source.yml` connection set up in the package will not function when the union schema/database feature is utilized. Although the data will be correctly combined, you will not observe the sources linked to the package models in the Directed Acyclic Graph (DAG). This happens because the package includes only one defined `source.yml`.
+
+To connect your multiple schema/database sources to the package models, follow the steps outlined in the [Union Data Defined Sources Configuration](https://github.com/fivetran/dbt_fivetran_utils/tree/releases/v0.4.latest#union_data-source) section of the Fivetran Utils documentation for the union_data macro. This will ensure a proper configuration and correct visualization of connections in the DAG.
 
 ### Switching to Local Currency
 Additionally, the package allows you to select whether you want to add in costs in USD or the local currency of the ad. By default, the package uses USD. If you would like to have costs in the local currency, add the following variable to your `dbt_project.yml` file:
@@ -112,8 +122,6 @@ vars:
     linkedin_ads_<default_source_table_name>_identifier: your_table_name 
 ```
 
-</details>
-
 ## (Optional) Step 5: Orchestrate your models with Fivetran Transformations for dbt Core™
 <details><summary>Expand for more details</summary>
 

dbt_project.yml

@@ -1,18 +1,20 @@
 name: 'linkedin_source'
-version: '0.7.0'
+version: '0.8.0'
 config-version: 2
 require-dbt-version: [">=1.3.0", "<2.0.0"]
 models:
   linkedin_source:
-    +materialized: view
+    tmp:
+      +materialized: view
     +schema: linkedin_ads_source
+    +materialized: table
 vars:
   linkedin_source:
-    account_history: "{{ source('linkedin','account_history') }}"
-    ad_analytics_by_creative: "{{ source('linkedin','ad_analytics_by_creative') }}"
-    campaign_group_history: "{{ source('linkedin','campaign_group_history') }}"
-    campaign_history: "{{ source('linkedin','campaign_history') }}"
-    creative_history: "{{ source('linkedin','creative_history') }}"
-    ad_analytics_by_campaign: "{{ source('linkedin', 'ad_analytics_by_campaign') }}"
+    account_history: "{{ source('linkedin_ads','account_history') }}"
+    ad_analytics_by_creative: "{{ source('linkedin_ads','ad_analytics_by_creative') }}"
+    campaign_group_history: "{{ source('linkedin_ads','campaign_group_history') }}"
+    campaign_history: "{{ source('linkedin_ads','campaign_history') }}"
+    creative_history: "{{ source('linkedin_ads','creative_history') }}"
+    ad_analytics_by_campaign: "{{ source('linkedin_ads', 'ad_analytics_by_campaign') }}"
   linkedin_ads__campaign_passthrough_metrics: []
   linkedin_ads__creative_passthrough_metrics: []

integration_tests/ci/sample.profiles.yml

@@ -16,13 +16,13 @@ integration_tests:
       pass: "{{ env_var('CI_REDSHIFT_DBT_PASS') }}"
       dbname: "{{ env_var('CI_REDSHIFT_DBT_DBNAME') }}"
       port: 5439
-      schema: linkedin_source_integration_tests_1
+      schema: linkedin_source_integration_tests_3
       threads: 8
     bigquery:
       type: bigquery
       method: service-account-json
       project: 'dbt-package-testing'
-      schema: linkedin_source_integration_tests_1
+      schema: linkedin_source_integration_tests_3
       threads: 8
       keyfile_json: "{{ env_var('GCLOUD_SERVICE_KEY') | as_native }}"
     snowflake:
@@ -33,7 +33,7 @@ integration_tests:
       role: "{{ env_var('CI_SNOWFLAKE_DBT_ROLE') }}"
       database: "{{ env_var('CI_SNOWFLAKE_DBT_DATABASE') }}"
       warehouse: "{{ env_var('CI_SNOWFLAKE_DBT_WAREHOUSE') }}"
-      schema: linkedin_source_integration_tests_1
+      schema: linkedin_source_integration_tests_3
       threads: 8
     postgres:
       type: postgres
@@ -42,13 +42,13 @@ integration_tests:
       pass: "{{ env_var('CI_POSTGRES_DBT_PASS') }}"
       dbname: "{{ env_var('CI_POSTGRES_DBT_DBNAME') }}"
       port: 5432
-      schema: linkedin_source_integration_tests_1
+      schema: linkedin_source_integration_tests_3
       threads: 8
     databricks:
       catalog: "{{ env_var('CI_DATABRICKS_DBT_CATALOG') }}"
       host: "{{ env_var('CI_DATABRICKS_DBT_HOST') }}"
       http_path: "{{ env_var('CI_DATABRICKS_DBT_HTTP_PATH') }}"
-      schema: linkedin_source_integration_tests_1
-      threads: 2
+      schema: linkedin_source_integration_tests_3
+      threads: 8
       token: "{{ env_var('CI_DATABRICKS_DBT_TOKEN') }}"
       type: databricks

integration_tests/dbt_project.yml

@@ -1,5 +1,5 @@
 name: 'linkedin_source_integration_tests'
-version: '0.7.0'
+version: '0.8.0'
 profile: 'integration_tests'
 config-version: 2
 
@@ -12,7 +12,7 @@ vars:
     linkedin_ads_creative_history_identifier: "linkedin_creative_history_data"
     linkedin_ads_ad_analytics_by_campaign_identifier: "linkedin_ad_analytics_by_campaign_data"
 
-    linkedin_ads_schema: linkedin_source_integration_tests_1
+    linkedin_ads_schema: linkedin_source_integration_tests_3
 
 seeds:
   linkedin_source_integration_tests:

models/docs.md

@@ -0,0 +1,3 @@
+{% docs source_relation %}
+The source of the record if the unioning functionality is being used. If not this field will be empty.
+{% enddocs %}

models/src_linkedin.yml

@@ -1,7 +1,7 @@
 version: 2
 
 sources:
-  - name: linkedin
+  - name: linkedin_ads # This source will only be used if you are using a single linkedin source connector. If multiple sources are being unioned, their tables will be directly referenced via adapter.get_relation.
     schema: "{{ var('linkedin_ads_schema', 'linkedin_ads') }}"
     database: "{% if target.type != 'spark'%}{{ var('linkedin_ads_database', target.database) }}{% endif %}"
 

models/stg_linkedin.yml

@@ -7,9 +7,13 @@ models:
     tests:
       - dbt_utils.unique_combination_of_columns:
           combination_of_columns:
+            - source_relation
             - date_day
             - creative_id 
     columns:
+      - name: source_relation
+        description: "{{ doc('source_relation') }}"
+
       - name: creative_id
         description: The ID of the related creative.
         tests:
@@ -34,9 +38,13 @@ models:
     tests:
       - dbt_utils.unique_combination_of_columns:
           combination_of_columns:
+            - source_relation
             - date_day
             - campaign_id 
     columns:
+      - name: source_relation
+        description: "{{ doc('source_relation') }}"
+
       - name: campaign_id
         description: The ID of the related creative.
         tests:
@@ -59,6 +67,9 @@ models:
   - name: stg_linkedin_ads__creative_history
     description: Each record represents a 'version' of a creative.
     columns:
+      - name: source_relation
+        description: "{{ doc('source_relation') }}"
+
       - name: creative_id
         description: Unique internal ID representing the creative.
         tests:
@@ -111,16 +122,20 @@ models:
         description: The utm_term parameter of the ad, extracted from the `click_uri`.
 
       - name: is_latest_version
-        description: Boolean of whether the record is the latest version of the cretive.
+        description: Boolean of whether the record is the latest version of the creative.
 
   - name: stg_linkedin_ads__campaign_history
     description: Each record represents a 'version' of a campaign.
     tests:
       - dbt_utils.unique_combination_of_columns:
           combination_of_columns:
+            - source_relation
             - version_tag
             - campaign_id 
     columns:
+      - name: source_relation
+        description: "{{ doc('source_relation') }}"
+
       - name: campaign_id
         description: Unique internal ID representing the campaign.
         tests:
@@ -238,9 +253,13 @@ models:
     tests:
       - dbt_utils.unique_combination_of_columns:
           combination_of_columns:
+            - source_relation
             - last_modified_at
             - campaign_group_id 
     columns:
+      - name: source_relation
+        description: "{{ doc('source_relation') }}"
+
       - name: campaign_group_id
         description: Unique internal ID representing the campaign group.
         tests:
@@ -288,9 +307,13 @@ models:
     tests:
       - dbt_utils.unique_combination_of_columns:
           combination_of_columns:
+            - source_relation
             - account_id
             - version_tag 
     columns:
+      - name: source_relation
+        description: "{{ doc('source_relation') }}"
+
       - name: account_id
         description: Unique internal ID representing the account.
         tests:

models/stg_linkedin_ads__account_history.sql

@@ -14,11 +14,18 @@ with base as (
                 staging_columns=get_account_history_columns()
             )
         }}
+
+        {{ fivetran_utils.source_relation(
+            union_schema_variable='linkedin_ads_union_schemas', 
+            union_database_variable='linkedin_ads_union_databases') 
+        }}
+
     from base
 
 ), fields as (
 
     select 
+        source_relation,
         id as account_id,
         name as account_name,
         currency,
@@ -27,7 +34,7 @@ with base as (
         type,
         cast(last_modified_time as {{ dbt.type_timestamp() }}) as last_modified_at,
         cast(created_time as {{ dbt.type_timestamp() }}) as created_at,
-        row_number() over (partition by id order by last_modified_time desc) = 1 as is_latest_version
+        row_number() over (partition by source_relation, id order by last_modified_time desc) = 1 as is_latest_version
 
     from macro
 

models/stg_linkedin_ads__ad_analytics_by_campaign.sql

@@ -15,12 +15,19 @@ macro as (
                 staging_columns=get_ad_analytics_by_campaign_columns()
             )
         }}
+
+        {{ fivetran_utils.source_relation(
+            union_schema_variable='linkedin_ads_union_schemas', 
+            union_database_variable='linkedin_ads_union_databases') 
+        }}
+
     from base
 ),
 
 fields as (
 
     select 
+        source_relation,
         {{ dbt.date_trunc('day', 'day') }} as date_day,
         campaign_id,
         clicks,

models/stg_linkedin_ads__ad_analytics_by_creative.sql

@@ -14,11 +14,18 @@ with base as (
                 staging_columns=get_ad_analytics_by_creative_columns()
             )
         }}
+
+        {{ fivetran_utils.source_relation(
+            union_schema_variable='linkedin_ads_union_schemas', 
+            union_database_variable='linkedin_ads_union_databases') 
+        }}
+
     from base
 
 ), fields as (
 
     select
+        source_relation,
         {{ dbt.date_trunc('day', 'day') }} as date_day,
         creative_id,
         clicks, 

models/stg_linkedin_ads__campaign_group_history.sql

@@ -14,11 +14,18 @@ with base as (
                 staging_columns=get_campaign_group_history_columns()
             )
         }}
+
+        {{ fivetran_utils.source_relation(
+            union_schema_variable='linkedin_ads_union_schemas', 
+            union_database_variable='linkedin_ads_union_databases') 
+        }}
+
     from base
 
 ), fields as (
 
     select 
+        source_relation,
         id as campaign_group_id,
         name as campaign_group_name,
         account_id,
@@ -28,7 +35,7 @@ with base as (
         cast(run_schedule_end as {{ dbt.type_timestamp() }}) as run_schedule_end_at,
         cast(last_modified_time as {{ dbt.type_timestamp() }}) as last_modified_at,
         cast(created_time as {{ dbt.type_timestamp() }}) as created_at,
-        row_number() over (partition by id order by last_modified_time desc) = 1 as is_latest_version
+        row_number() over (partition by source_relation, id order by last_modified_time desc) = 1 as is_latest_version
 
     from macro
 

models/stg_linkedin_ads__campaign_history.sql

@@ -14,11 +14,18 @@ with base as (
                 staging_columns=get_campaign_history_columns()
             )
         }}
+
+        {{ fivetran_utils.source_relation(
+            union_schema_variable='linkedin_ads_union_schemas', 
+            union_database_variable='linkedin_ads_union_databases') 
+        }}
+
     from base
 
 ), fields as (
 
     select 
+        source_relation,
         id as campaign_id,
         name as campaign_name,
         cast(version_tag as numeric) as version_tag,
@@ -43,7 +50,7 @@ with base as (
         cast(run_schedule_end as {{ dbt.type_timestamp() }}) as run_schedule_end_at,
         cast(last_modified_time as {{ dbt.type_timestamp() }}) as last_modified_at,
         cast(created_time as {{ dbt.type_timestamp() }}) as created_at,
-        row_number() over (partition by id order by last_modified_time desc) = 1 as is_latest_version
+        row_number() over (partition by source_relation, id order by last_modified_time desc) = 1 as is_latest_version
 
     from macro