0.7.0 into main (dbt-labs#372)

* Tidy up changelog * Add 0.7.0 entry to changelog * Add order_by argument to get_column_values (dbt-labs#349) * Add slugify macro to utils, use in pivot macro (dbt-labs#314) * 0.20.0 compatibility (dbt-labs#371) * Explicitly redefine Redshift -> default * Upgrade generic tests * Rm namespaces macro. New dispatch syntax * Run tests with 0.20.0rc1 * Update changelog, readme Co-authored-by: Jeremy Cohen <jeremy@fishtownanalytics.com> * Simplify concat (dbt-labs#373) * Postgres also have an alternative concat binary operation (dbt-labs#296) * Update default implementation of concat macro Co-authored-by: Christophe Duong <christophe.duong@gmail.com> Co-authored-by: Jeremy Cohen <jeremy@fishtownanalytics.com> Co-authored-by: Christophe Duong <christophe.duong@gmail.com>
yu-iskw · Jun 6, 2021 · a729044 · a729044
1 parent 0e00fb5
commit a729044
Show file tree

Hide file tree

Showing 77 changed files with 419 additions and 333 deletions.
diff --git a/.circleci/config.yml b/.circleci/config.yml
@@ -5,7 +5,7 @@ jobs:
 
   integration-postgres:
     docker:
-      - image: circleci/python:3.6.3-stretch
+      - image: circleci/python:3.6.13-stretch
       - image: circleci/postgres:9.6.5-alpine-ram
 
     steps:

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,27 +1,76 @@
+# dbt-utils v0.7.0 (unreleased)
+## Breaking changes
+
+### 🚨 New dbt version
+
+dbt v0.20.0 or greater is required for this release. If you are not ready to upgrade, consider using a previous release of this package. In accordance with the version upgrade, this package release includes breaking changes to:
+- Generic (schema) tests
+- `dispatch` functionality
+
+### 🚨 get_column_values
+The order of (optional) arguments has changed in the `get_column_values` macro.
+
+Before:
+```jinja
+{% macro get_column_values(table, column, order_by='count(*) desc', max_records=none, default=none) -%}
+...
+{% endmacro %}
+```
+
+After:
+```jinja
+{% macro get_column_values(table, column, max_records=none, default=none) -%}
+...
+{% endmacro %}
+```
+If you were relying on the position to match up your optional arguments, this may be a breaking change — in general, we recommend that you explicitly declare any optional arguments (if not all of your arguments!)
+```
+-- before: This works on previous version of dbt-utils, but on 0.7.0, the `50` would be passed through as the `order_by` argument
+{% set payment_methods = dbt_utils.get_column_values(
+        ref('stg_payments'),
+        'payment_method',
+        50
+) %}
+
+-- after
+{% set payment_methods = dbt_utils.get_column_values(
+        ref('stg_payments'),
+        'payment_method',
+        max_records=50
+) %}
+```
+
+## Features
+* Add new argument, `order_by`, to `get_column_values` (code originally in [#289](https://github.com/fishtown-analytics/dbt-utils/pull/289/) from [@clausherther](https://github.com/clausherther), merged via [#349](https://github.com/fishtown-analytics/dbt-utils/pull/349/))
+* Add `slugify` macro, and use it in the pivot macro. :rotating_light: This macro uses the `re` module, which is only available in dbt v0.19.0+. As a result, this feature introduces a breaking change. ([#314](https://github.com/fishtown-analytics/dbt-utils/pull/314))
+
+## Under the hood
+* Update the default implementation of concat macro to use `||` operator ([#373](https://github.com/fishtown-analytics/dbt-utils/pull/314) [@ChristopheDuong](https://github.com/ChristopheDuong)). Note this may be a breaking change for spark users.
+
 # dbt-utils v0.6.6
 
 ## Fixes
 
 - make `sequential_values` schema test use `dbt_utils.type_timestamp()` to allow for compatibility with db's without timestamp data type. [#376](https://github.com/fishtown-analytics/dbt-utils/pull/376) from [@swanderz](https://github.com/swanderz)
+
 # dbt-utils v0.6.5
 ## Features
 * Add new `accepted_range` test ([#276](https://github.com/fishtown-analytics/dbt-utils/pull/276) [@joellabes](https://github.com/joellabes))
-* Make `expression_is_true` work as a column test (code originally in [#226](https://github.com/fishtown-analytics/dbt-utils/pull/226/) from [@elliottohara](https://github.com/elliottohara), merged via [#313])
+* Make `expression_is_true` work as a column test (code originally in [#226](https://github.com/fishtown-analytics/dbt-utils/pull/226/) from [@elliottohara](https://github.com/elliottohara), merged via [#313](https://github.com/fishtown-analytics/dbt-utils/pull/313/))
 * Add new schema test, `not_accepted_values` ([#284](https://github.com/fishtown-analytics/dbt-utils/pull/284) [@JavierMonton](https://github.com/JavierMonton))
-* Support a new argument, `zero_length_range_allowed` in the `mutually_exclusive_ranges` test ([#307](https://github.com/fishtown-analytics/dbt-utils/pull/307) [@zemekeng](https://github.com/zemekeneng))
+* Support a new argument, `zero_length_range_allowed` in the `mutually_exclusive_ranges` test ([#307](https://github.com/fishtown-analytics/dbt-utils/pull/307) [@zemekeneng](https://github.com/zemekeneng))
 * Add new schema test, `sequential_values` ([#318](https://github.com/fishtown-analytics/dbt-utils/pull/318), inspired by [@hundredwatt](https://github.com/hundredwatt))
-* Support `quarter` in the `postgres__last_day` macro ([#333](https://github.com/fishtown-analytics/dbt-utils/pull/333/files), [@seunghanhong](https://github.com/seunghanhong))
-* Add new argument, `unit`, to `haversine_distance` [#340](https://github.com/fishtown-analytics/dbt-utils/pull/340) [@bastienboutonnet](https://github.com/bastienboutonnet)
-* Add new schema test, `fewer_rows_than` (code originally in [#221](https://github.com/fishtown-analytics/dbt-utils/pull/230/) from [@dmarts](https://github.com/dmarts), merged via [#343])
-
+* Support `quarter` in the `postgres__last_day` macro ([#333](https://github.com/fishtown-analytics/dbt-utils/pull/333/files) [@seunghanhong](https://github.com/seunghanhong))
+* Add new argument, `unit`, to `haversine_distance` ([#340](https://github.com/fishtown-analytics/dbt-utils/pull/340) [@bastienboutonnet](https://github.com/bastienboutonnet))
+* Add new schema test, `fewer_rows_than` (code originally in [#221](https://github.com/fishtown-analytics/dbt-utils/pull/230/) from [@dmarts](https://github.com/dmarts), merged via [#343](https://github.com/fishtown-analytics/dbt-utils/pull/343/))
 
 ## Fixes
 * Handle booleans gracefully in the unpivot macro ([#305](https://github.com/fishtown-analytics/dbt-utils/pull/305) [@avishalom](https://github.com/avishalom))
-* Fix a bug in `get_relation_by_prefix` that happens with Snowflake external tables. Now the macro will retrieve tables that match the prefix which are external tables ([#350](https://github.com/fishtown-analytics/dbt-utils/issues/350))
-* Fix `cardinality_equality` test when the two tables' column names differed ([#334](https://github.com/fishtown-analytics/dbt-utils/pull/334)) [@joellabes](https://github.com/joellabes)
+* Fix a bug in `get_relation_by_prefix` that happens with Snowflake external tables. Now the macro will retrieve tables that match the prefix which are external tables ([#351](https://github.com/fishtown-analytics/dbt-utils/pull/351))
+* Fix `cardinality_equality` test when the two tables' column names differed ([#334](https://github.com/fishtown-analytics/dbt-utils/pull/334) [@joellabes](https://github.com/joellabes))
 
 ## Under the hood
-* Fix Markdown formatting for hub rendering ([#336](https://github.com/fishtown-analytics/dbt-utils/issues/350), [@coapacetic](https://github.com/coapacetic))
+* Fix Markdown formatting for hub rendering ([#336](https://github.com/fishtown-analytics/dbt-utils/issues/350) [@coapacetic](https://github.com/coapacetic))
 * Reorder readme and improve docs
 
 # dbt-utils v0.6.4

diff --git a/README.md b/README.md
@@ -510,24 +510,50 @@ These macros run a query and return the results of the query as objects. They ar
 
 
 #### get_column_values ([source](macros/sql/get_column_values.sql))
-This macro returns the unique values for a column in a given [relation](https://docs.getdbt.com/docs/writing-code-in-dbt/class-reference/#relation).
-It takes an options `default` argument for compiling when the relation does not already exist.
+This macro returns the unique values for a column in a given [relation](https://docs.getdbt.com/docs/writing-code-in-dbt/class-reference/#relation) as an array.
+
+Arguments:
+- `table` (required): a [Relation](https://docs.getdbt.com/reference/dbt-classes#relation) (a `ref` or `source`) that contains the list of columns you wish to select from
+- `column` (required): The name of the column you wish to find the column values of
+- `order_by` (optional, default=`'count(*) desc'`): How the results should be ordered. The default is to order by `count(*) desc`, i.e. decreasing frequency. Setting this as `'my_column'` will sort alphabetically, while `'min(created_at)'` will sort by when thevalue was first observed.
+- `max_records` (optional, default=`none`): The maximum number of column values you want to return
+- `default` (optional, default=`[]`): The results this macro should return if the relation has not yet been created (and therefore has no column values).
 
 
 **Usage:**
 ```sql
 -- Returns a list of the payment_methods in the stg_payments model_
 {% set payment_methods = dbt_utils.get_column_values(table=ref('stg_payments'), column='payment_method') %}
 
-{% for state in states %}
+{% for payment_method in payment_methods %}
     ...
 {% endfor %}
 
 ...
 ```
 
-#### get_relations_by_pattern ([source](macros/sql/get_relations_by_pattern.sql))
+```sql
+-- Returns the list sorted alphabetically
+{% set payment_methods = dbt_utils.get_column_values(
+        table=ref('stg_payments'),
+        column='payment_method',
+        order_by='payment_method'
+) %}
+```
+
+```sql
+-- Returns the list sorted my most recently observed
+{% set payment_methods = dbt_utils.get_column_values(
+        table=ref('stg_payments'),
+        column='payment_method',
+        order_by='max(created_at) desc',
+        max_records=50,
+        default=['bank_transfer', 'coupon', 'credit_card']
+%}
+...
+```
 
+#### get_relations_by_pattern ([source](macros/sql/get_relations_by_pattern.sql))
 Returns a list of [Relations](https://docs.getdbt.com/docs/writing-code-in-dbt/class-reference/#relation)
 that match a given schema- or table-name pattern.
 
@@ -948,8 +974,8 @@ When an expression falls outside the range, the function returns:
 
 
 ---
-### Logger
-#### pretty_time ([source](macros/logger/pretty_time.sql))
+### Jinja Helpers
+#### pretty_time ([source](macros/jinja_helpers/pretty_time.sql))
 This macro returns a string of the current timestamp, optionally taking a datestring format.
 ```sql
 {#- This will return a string like '14:50:34' -#}
@@ -959,7 +985,7 @@ This macro returns a string of the current timestamp, optionally taking a datest
 {{ dbt_utils.pretty_time(format='%Y-%m-%d %H:%M:%S') }}
 ```
 
-#### pretty_log_format ([source](macros/logger/pretty_log_format.sql))
+#### pretty_log_format ([source](macros/jinja_helpers/pretty_log_format.sql))
 This macro formats the input in a way that will print nicely to the command line when you `log` it.
 ```sql
 {#- This will return a string like:
@@ -968,7 +994,7 @@ This macro formats the input in a way that will print nicely to the command line
 
 {{ dbt_utils.pretty_log_format("my pretty message") }}
 ```
-#### log_info ([source](macros/logger/log_info.sql))
+#### log_info ([source](macros/jinja_helpers/log_info.sql))
 This macro logs a formatted message (with a timestamp) to the command line.
 ```sql
 {{ dbt_utils.log_info("my pretty message") }}
@@ -979,6 +1005,40 @@ This macro logs a formatted message (with a timestamp) to the command line.
 11:07:31 + my pretty message
 ```
 
+#### slugify ([source](macros/jinja_helpers/slugify.sql))
+This macro is useful for transforming Jinja strings into "slugs", and can be useful when using a Jinja object as a column name, especially when that Jinja object is not hardcoded.
+
+For this example, let's pretend that we have payment methods in our payments table like `['venmo App', 'ca$h-money']`, which we can't use as a column name due to the spaces and special characters. This macro does its best to strip those out in a sensible way: `['venmo_app',
+'cah_money']`.
+
+```sql
+{%- set payment_methods = dbt_utils.get_column_values(
+    table=ref('raw_payments'),
+    column='payment_method'
+) -%}
+
+select
+order_id,
+{%- for payment_method in payment_methods %}
+sum(case when payment_method = '{{ payment_method }}' then amount end)
+  as {{ slugify(payment_method) }}_amount,
+
+{% endfor %}
+...
+```
+
+```sql
+select
+order_id,
+
+sum(case when payment_method = 'Venmo App' then amount end)
+  as venmo_app_amount,
+
+sum(case when payment_method = 'ca$h money' then amount end)
+  as cah_money_amount,
+...
+```
+
 ### Materializations
 #### insert_by_period ([source](macros/materializations/insert_by_period_materialization.sql))
 `insert_by_period` allows dbt to insert records into a table one period (i.e. day, week) at a time.
@@ -1046,29 +1106,26 @@ We welcome contributions to this repo! To contribute a new feature or a fix, ple
 **Note:** This is primarily relevant to:
 - Users and maintainers of community-supported [adapter plugins](https://docs.getdbt.com/docs/available-adapters)
 - Users who wish to override a low-lying `dbt_utils` macro with a custom implementation, and have that implementation used by other `dbt_utils` macros
+
 If you use Postgres, Redshift, Snowflake, or Bigquery, this likely does not apply to you.
 
-dbt v0.18.0 introduces `adapter.dispatch()`, a reliable way to define different implementations of the same macro
-across different databases.
+dbt v0.18.0 introduced [`adapter.dispatch()`](https://docs.getdbt.com/reference/dbt-jinja-functions/adapter#dispatch), a reliable way to define different implementations of the same macro across different databases.
 
-All dispatched macros in `dbt_utils` have an override setting: a `var` named
-`dbt_utils_dispatch_list` that accepts a list of package names. If you set this
-variable in your project, when dbt searches for implementations of a dispatched
-`dbt_utils` macro, it will search through your listed packages _before_ using
-the implementations defined in `dbt_utils`.
+dbt v0.20.0 introduced a new project-level `dispatch` config that enables an "override" setting for all dispatched macros. If you set this config in your project, when dbt searches for implementations of a macro in the `dbt_utils` namespace, it will search through your list of packages instead of just looking in the `dbt_utils` package.
 
-Set a variable in your `dbt_project.yml`:
+Set the config in `dbt_project.yml`:
 ```yml
-vars:
-  dbt_utils_dispatch_list:
-    - first_package_to_search    # likely the name of your root project (only the root folder)
-    - second_package_to_search   # likely an "add-on" package, such as spark_utils
-    # dbt_utils is always the last place searched
+dispatch:
+  - macro_namespace: dbt_utils
+    search_order:
+      - first_package_to_search    # likely the name of your root project
+      - second_package_to_search   # could be a "shim" package, such as spark_utils
+      - dbt_utils                  # always include dbt_utils as the last place to search
 ```
 
 If overriding a dispatched macro with a custom implementation in your own project's `macros/` directory, you must name your custom macro with a prefix: either `default__` (note the two underscores), or the name of your adapter followed by two underscores. For example, if you're running on Postgres and wish to override the behavior of `dbt_utils.datediff` (such that `dbt_utils.date_spine` will use your version instead), you can do this by defining a macro called either `default__datediff` or `postgres__datediff`.
 
-When running on Spark, if dbt needs to dispatch `dbt_utils.datediff`, it will search for the following in order:
+Let's say we have the config defined above, and we're running on Spark. When dbt goes to dispatch `dbt_utils.datediff`, it will search for macros the following in order:
 ```
 first_package_to_search.spark__datediff
 first_package_to_search.default__datediff

diff --git a/dbt_project.yml b/dbt_project.yml
@@ -1,7 +1,8 @@
 name: 'dbt_utils'
 version: '0.1.0'
 
-require-dbt-version: [">=0.18.0", "<0.20.0"]
+require-dbt-version: [">=0.20.0", "<0.21.0"]
+
 config-version: 2
 
 target-path: "target"

diff --git a/integration_tests/dbt_project.yml b/integration_tests/dbt_project.yml
@@ -19,8 +19,9 @@ clean-targets:         # directories to be removed by `dbt clean`
     - "target"
     - "dbt_modules"
 
-vars:
-  dbt_utils_dispatch_list: ['dbt_utils_integration_tests']
+dispatch:
+  - macro_namespace: 'dbt_utils'
+    search_order: ['dbt_utils_integration_tests', 'dbt_utils']
 
 seeds:
 

diff --git a/integration_tests/macros/limit_zero.sql b/integration_tests/macros/limit_zero.sql
@@ -1,5 +1,9 @@
+{% macro my_custom_macro() %}
+    whatever
+{% endmacro %}
+
 {% macro limit_zero() %}
-    {{ return(adapter.dispatch('limit_zero', dbt_utils._get_utils_namespaces())()) }}
+    {{ return(adapter.dispatch('limit_zero', 'dbt_utils')()) }}
 {% endmacro %}
 
 {% macro default__limit_zero() %}

diff --git a/integration_tests/macros/tests.sql b/integration_tests/macros/tests.sql
@@ -1,14 +1,12 @@
 
-{% macro test_assert_equal(model, actual, expected) %}
-select count(*) from {{ model }} where {{ actual }} != {{ expected }}
+{% test assert_equal(model, actual, expected) %}
+select * from {{ model }} where {{ actual }} != {{ expected }}
 
-{% endmacro %}
+{% endtest %}
 
 
-{% macro test_not_empty_string(model, arg) %}
+{% test not_empty_string(model, column_name) %}
 
-{% set column_name = kwargs.get('column_name', kwargs.get('arg')) %}
+select * from {{ model }} where {{ column_name }} = ''
 
-select count(*) from {{ model }} where {{ column_name }} = ''
-
-{% endmacro %}
+{% endtest %}
diff --git a/integration_tests/models/schema_tests/schema.yml b/integration_tests/models/schema_tests/schema.yml
@@ -86,7 +86,6 @@ models:
       - name: id
         tests:
           - dbt_utils.relationships_where:
-              from: id
               to: ref('data_test_relationships_where_table_1')
               field: id
               from_condition: id <> 4

diff --git a/integration_tests/models/sql/schema.yml b/integration_tests/models/sql/schema.yml
@@ -4,7 +4,7 @@ models:
   - name: test_generate_series
     tests:
       - dbt_utils.equality:
-          arg: ref('data_generate_series')
+          compare_model: ref('data_generate_series')
 
   - name: test_get_column_values
     columns:

diff --git a/integration_tests/models/sql/test_get_column_values.sql b/integration_tests/models/sql/test_get_column_values.sql
@@ -1,5 +1,5 @@
 
-{% set columns = dbt_utils.get_column_values(ref('data_get_column_values'), 'field', default = []) %}
+{% set columns = dbt_utils.get_column_values(ref('data_get_column_values'), 'field', default=[], order_by="field") %}
 
 
 {% if target.type == 'snowflake' %}

diff --git a/...er/assert_pretty_output_msg_is_string.sql → ...rs/assert_pretty_output_msg_is_string.sql b/...er/assert_pretty_output_msg_is_string.sql → ...rs/assert_pretty_output_msg_is_string.sql
diff --git a/...s/logger/assert_pretty_time_is_string.sql → ..._helpers/assert_pretty_time_is_string.sql b/...s/logger/assert_pretty_time_is_string.sql → ..._helpers/assert_pretty_time_is_string.sql
diff --git a/integration_tests/tests/jinja_helpers/test_slugify.sql b/integration_tests/tests/jinja_helpers/test_slugify.sql
@@ -0,0 +1,7 @@
+{% if dbt_utils.slugify('!Hell0 world-hi') == 'hell0_world_hi' %}
+    {# Return 0 rows for the test to pass #}
+    select 1 limit 0
+{% else %}
+    {# Return >0 rows for the test to fail #}
+    select 1
+{% endif %}
diff --git a/macros/cross_db_utils/_get_utils_namespaces.sql b/macros/cross_db_utils/_get_utils_namespaces.sql
diff --git a/macros/cross_db_utils/cast_bool_to_text.sql b/macros/cross_db_utils/cast_bool_to_text.sql
@@ -1,5 +1,5 @@
 {% macro cast_bool_to_text(field) %}
-  {{ adapter.dispatch('cast_bool_to_text', packages = dbt_utils._get_utils_namespaces()) (field) }}
+  {{ adapter.dispatch('cast_bool_to_text', 'dbt_utils') (field) }}
 {% endmacro %}
 
 

diff --git a/macros/cross_db_utils/concat.sql b/macros/cross_db_utils/concat.sql
@@ -1,23 +1,9 @@
 
 
 {% macro concat(fields) -%}
-  {{ return(adapter.dispatch('concat', packages = dbt_utils._get_utils_namespaces())(fields)) }}
+  {{ return(adapter.dispatch('concat', 'dbt_utils')(fields)) }}
 {%- endmacro %}
 
 {% macro default__concat(fields) -%}
-    concat({{ fields|join(', ') }})
-{%- endmacro %}
-
-{% macro alternative_concat(fields) %}
     {{ fields|join(' || ') }}
-{% endmacro %}
-
-
-{% macro redshift__concat(fields) %}
-    {{ dbt_utils.alternative_concat(fields) }}
-{% endmacro %}
-
-
-{% macro snowflake__concat(fields) %}
-    {{ dbt_utils.alternative_concat(fields) }}
-{% endmacro %}
+{%- endmacro %}