feat: Unit tests for dbt models

.github/workflows/coverage.yml

@@ -1,6 +1,6 @@
 # Check documentation coverage
 
-name: dbt Docs Coverage
+name: dbt Tests & Coverage
 
 on:
   push:
@@ -17,7 +17,7 @@ env:
 
 jobs:
   build:
-    name: Build docs and check coverage
+    name: Check coverage & run tests
     runs-on: ubuntu-latest
     permissions:
       contents: "read"
@@ -46,10 +46,12 @@ jobs:
           tutor local do load-xapi-test-data
       - name: Check dbt tests
         run: |
-          dbt run
-          dbt test
+          mv unit-test-seeds ci-seeds
+          dbt seed --full-refresh --selector all_tests
+          dbt run --full-refresh --selector all_tests
+          dbt test --selector all_tests
+          mv ci-seeds unit-test-seeds
       - name: Check docs coverage
         run: |
-          dbt run
           dbt docs generate
-          dbt-coverage compute doc --cov-fail-under 1.0
+          dbt-coverage compute doc --cov-fail-under 1.0 --model-path-filter models/

.gitignore

@@ -3,4 +3,4 @@ target/
 dbt_packages/
 logs/
 coverage.json
-
+.DS_Store

Makefile

@@ -2,4 +2,4 @@ format:
 	sqlfmt models macros
 
 coverage:
-	dbt-coverage compute doc --cov-fail-under 0.9
+	dbt-coverage compute doc --cov-fail-under 1.0

README.rst

@@ -33,6 +33,21 @@ Running dbt
 
 ``dbt run`` will compile and create the models defined in the "aspects" dbt project. By default, dbt will look in the ``xapi`` schema to find source tables. The ``XAPI_SCHEMA`` environment variable can be used to specify a different schema.
 
+Testing
+*******
+As of `dbt v1.8 <https://docs.getdbt.com/reference/resource-properties/unit-tests>`, models can now be tested with UNIT tests in addition to the existing DATA tests. Unit tests validate the SQL model logic by building the models using a (known to be good) dataset and comparing the results to a provided 'expected' dataset. This is especially beneficial when updating a model to ensure the output has not changed.
+
+The ``unit_tests.yaml`` file in each model directory contains any tests for the models in that directory.
+The ``unit-test-seeds`` directory contains all seed data csv files. There is one file for each base table (event_sink & xapi) and each 'expected' dataset.
+
+``dbt test`` will only run data & generic tests (NOT unit tests). This is the default mode.
+
+``dbt test --selector unit_tests`` will run all unit tests.
+These require tables to be seeded first. To do this, add 'unit-test-seeds' to ``seed-paths:`` in ``dbt_project.yml`` and run ``dbt seed --full-refresh && dbt run --full-refresh``.
+
+``dbt test --selector all_tests`` will run all data/generic/unit tests.
+
+
 More Help
 =========
 

dbt_project.yml

@@ -11,10 +11,10 @@ profile: "aspects"
 # These configurations specify where dbt should look for different types of files.
 # The `model-paths` config, for example, states that models in this project can be
 # found in the "models/" directory. You probably won't need to change these!
-model-paths: ["models"]
+model-paths: ["models",'ci-models']
 analysis-paths: ["analyses"]
 test-paths: ["tests"]
-seed-paths: ["seeds"]
+seed-paths: ["seeds","ci-seeds"]
 macro-paths: ["macros"]
 snapshot-paths: ["snapshots"]
 
@@ -34,3 +34,10 @@ models:
     # Config indicated by + and applies to all files under models/example/
     enrollment:
       +materialized: view
+
+# These are for unit test seeds. They will be used when 'unit-test-seeds' is added 
+# to seed-paths above or when CI tests run
+seeds:
+  aspects:
+    +column_types:
+      event_id: UUID

macros/get_custom_schema.sql

@@ -1,3 +1,14 @@
 {% macro generate_schema_name(custom_schema_name, node) -%}
-    {{ generate_schema_name_for_env(custom_schema_name, node) }}
+
+    {%- set default_schema = target.schema -%}
+    {%- if (
+        target.name == "prod" or node.resource_type == "seed"
+    ) and custom_schema_name is not none -%}
+
+        {{ custom_schema_name | trim }}
+
+    {%- else -%} {{ default_schema }}
+
+    {%- endif -%}
+
 {%- endmacro %}

macros/get_problem_id.sql

@@ -3,6 +3,6 @@
 -- or is followed by '/answer' or '/hint'
 {% macro get_problem_id(object_id) %}
     regexpExtract(
-        object_id, 'xblock/([\w\d-\+:@]*@problem\+block@[\w\d][^_]*)(_\d_\d)?', 1
+        object_id, 'xblock/([\w\d-\+:@]*@problem\+block@[\w\d][^_\/]*)(_\d_\d)?', 1
     )
 {% endmacro %}

models/base/schema.yml

@@ -5,29 +5,29 @@ models:
     description: "A materialized view for xAPI events"
     columns:
       - name: event_id
-        data_type: uuid
+        data_type: UUID
         description: "The unique identifier for the event"
       - name: verb_id
-        data_type: string
+        data_type: String
         description: "The xAPI verb identifier"
       - name: actor_id
-        data_type: string
+        data_type: String
         description: "The xAPI actor identifier"
       - name: object_id
-        data_type: string
+        data_type: String
         description: "The xAPI object identifier"
       - name: course_id
-        data_type: string
+        data_type: String
         description: "The fully-qualified course identifier URL"
       - name: course_key
         data_type: String
         description: "The course key for the course"
       - name: org
-        data_type: string
+        data_type: String
         description: "The organization that the course belongs to"
       - name: emission_time
-        data_type: datetime64(6)
+        data_type: DateTime64(6)
         description: "The time the event was emitted"
       - name: event
-        data_type: string
-        description: "The xAPI event as a string"
+        data_type: String
+        description: "The xAPI event as a String"

models/base/unit_tests.yaml

@@ -0,0 +1,14 @@
+unit_tests:
+  - name: test_xapi_events_all_parsed
+    model: xapi_events_all_parsed
+    config:
+      tags: 'ci'
+    given:
+      - input: source("xapi", "xapi_events_all")
+        format: sql
+        rows: |
+          select * from xapi.xapi_events_all
+    expect:
+      format: sql
+      rows: |
+        select * from xapi_events_all_parsed_expected

models/courses/course_block_names.sql

@@ -12,45 +12,14 @@
         ],
         primary_key="location",
         layout="COMPLEX_KEY_SPARSE_HASHED()",
-        lifetime="120",
+        lifetime=env_var("ASPECTS_BLOCK_NAME_CACHE_LIFETIME", "120"),
         source_type="clickhouse",
         connection_overrides={
             "host": "localhost",
         },
     )
 }}
 
-with
-    most_recent_course_blocks as (
-        select
-            location,
-            display_name,
-            toString(section)
-            || ':'
-            || toString(subsection)
-            || ':'
-            || toString(unit)
-            || ' - '
-            || display_name as display_name_with_location,
-            JSONExtractInt(xblock_data_json, 'section') as section,
-            JSONExtractInt(xblock_data_json, 'subsection') as subsection,
-            JSONExtractInt(xblock_data_json, 'unit') as unit,
-            JSONExtractBool(xblock_data_json, 'graded') as graded,
-            `order` as course_order,
-            course_key,
-            dump_id,
-            time_last_dumped,
-            row_number() over (
-                partition by location order by time_last_dumped desc
-            ) as rn
-        from {{ source("event_sink", "course_blocks") }}
-    )
 select
-    location,
-    display_name as block_name,
-    course_key,
-    graded,
-    course_order,
-    display_name_with_location
-from most_recent_course_blocks
-where rn = 1
+    location, block_name, course_key, graded, course_order, display_name_with_location
+from {{ ref("most_recent_course_blocks") }}

models/courses/course_names.sql

@@ -10,7 +10,7 @@
         ],
         primary_key="course_key",
         layout="COMPLEX_KEY_HASHED()",
-        lifetime="120",
+        lifetime=env_var("ASPECTS_COURSE_NAME_CACHE_LIFETIME", "120"),
         source_type="clickhouse",
         connection_overrides={
             "host": "localhost",
@@ -24,7 +24,11 @@ with
         from {{ source("event_sink", "course_overviews") }}
         group by org, course_key
     )
-select course_key, display_name, splitByString('+', course_key)[-1] as course_run, org
+select
+    course_key,
+    display_name as course_name,
+    splitByString('+', course_key)[-1] as course_run,
+    org
 from {{ source("event_sink", "course_overviews") }} co
 inner join
     most_recent_overviews mro

models/courses/dim_course_blocks.sql

@@ -22,6 +22,4 @@ select
         else regexpExtract(block_id, '@([^+]+)\+block@', 1)
     end as block_type
 from {{ ref("course_block_names") }} blocks
-join
-    {{ ref("course_names") }} courses on blocks.course_key = courses.course_key
-    settings join_algorithm = 'direct'
+join {{ ref("course_names") }} courses on blocks.course_key = courses.course_key

models/courses/most_recent_course_blocks.sql

@@ -0,0 +1,37 @@
+{{
+    config(
+        materialized="materialized_view",
+        schema=env_var("ASPECTS_EVENT_SINK_DATABASE", "event_sink"),
+        engine=get_engine("ReplacingMergeTree()"),
+        order_by="(location)",
+        post_hook="OPTIMIZE TABLE {{ this }} {{ on_cluster() }} FINAL",
+    )
+}}
+
+select
+    location,
+    display_name as block_name,
+    toString(section)
+    || ':'
+    || toString(subsection)
+    || ':'
+    || toString(unit)
+    || ' - '
+    || display_name as display_name_with_location,
+    JSONExtractInt(xblock_data_json, 'section') as section,
+    JSONExtractInt(xblock_data_json, 'subsection') as subsection,
+    JSONExtractInt(xblock_data_json, 'unit') as unit,
+    JSONExtractBool(xblock_data_json, 'graded') as graded,
+    order as course_order,
+    course_key,
+    dump_id,
+    time_last_dumped
+from {{ source("event_sink", "course_blocks") }} course_blocks
+join
+    (
+        select location, max(time_last_dumped) as max_time_last_dumped
+        from {{ source("event_sink", "course_blocks") }}
+        group by location
+    ) latest_course_blocks
+    on course_blocks.location = latest_course_blocks.location
+    and course_blocks.time_last_dumped = latest_course_blocks.max_time_last_dumped

models/courses/schema.yml

@@ -23,29 +23,51 @@ models:
         data_type: String
         description: "The block's name"
       - name: section_number
-        data_type: string
+        data_type: String
         description: "The section this block belongs to, formatted as <section location>:0:0"
       - name: subsection_number
-        data_type: string
+        data_type: String
         description: "The subsection this block belongs to, formatted as <section location>:<subsection location>:0"
       - name: hierarchy_location
-        data_type: string
+        data_type: String
         description: "The full section:subsection:unit hierarchy in which this block belongs"
       - name: display_name_with_location
         data_type: String
         description: "The block's display name with section, subsection, and unit prepended to the name. This provides additional context when looking at block names and can help data consumers understand which block they are analyzing"
+      - name: course_order
+        data_type: Int32
+        description: "The sort order of this block in the course across all course blocks"
       - name: graded
         data_type: Boolean
         description: "Whether the block is graded"
       - name: block_type
         data_type: String
         description: "The type of block. This can be a section, subsection, unit, or the block type"
+
+  - name: course_block_names
+    description: "A table of course blocks with their names"
+    columns:
+      - name: location
+        data_type: String
+        description: "The location of the block"
+      - name: block_name
+        data_type: String
+        description: "The name of the block"
+      - name: course_key
+        data_type: String
+        description: "The course which the block belongs to"
+      - name: graded
+        data_type: Boolean
+        description: "Whether the block is graded"
+      - name: display_name_with_location
+        data_type: String
+        description: "The block's display name with section, subsection, and unit prepended to the name. This provides additional context when looking at block names and can help data consumers understand which block they are analyzing"
       - name: course_order
         data_type: Int32
         description: "The sort order of this block in the course across all course blocks"
 
-  - name: course_block_names
-    description: "A table of course blocks with their names"
+  - name: most_recent_course_blocks
+    description: "A materialized view of course blocks with their display names and additional metadata. Only stores the most recent row per block location."
     columns:
       - name: location
         data_type: String
@@ -65,6 +87,21 @@ models:
       - name: course_order
         data_type: Int32
         description: "The sort order of this block in the course across all course blocks"
+      - name: section
+        data_type: Int32
+        description: "The section number that this block falls under in the course. Starts at 1."
+      - name: subsection
+        data_type: Int32
+        description: "The subsection number that this block falls under in the section. Starts at 1."
+      - name: unit
+        data_type: Int32
+        description: "The unit number that this block falls under in the subsection. Starts at 1."
+      - name: dump_id
+        data_type: UUID
+        description: "The UUID of the event sink run that published this block to ClickHouse. When a course is published all blocks inside it are sent with the same dump_id."
+      - name: time_last_dumped
+        data_type: String
+        description: "The Datetime of the event sink run that published this block to ClickHouse. When a course is published all blocks inside it are sent with the same time_last_dumped."
 
   - name: course_names
     description: "A table of courses with their names"
@@ -104,13 +141,13 @@ models:
         data_type: String
         description: "The block's name"
       - name: section_number
-        data_type: string
+        data_type: String
         description: "The section this block belongs to, formatted as <section location>:0:0"
       - name: subsection_number
-        data_type: string
+        data_type: String
         description: "The subsection this block belongs to, formatted as <section location>:<subsection location>:0"
       - name: hierarchy_location
-        data_type: string
+        data_type: String
         description: "The full section:subsection:unit hierarchy in which this block belongs"
       - name: display_name_with_location
         data_type: String
@@ -122,10 +159,10 @@ models:
         data_type: String
         description: "The type of block. This can be a section, subsection, unit, or the block type"
       - name: section_with_name
-        data_type: string
+        data_type: String
         description: "The name of the section this block belongs to, with section_number prepended"
       - name: subsection_with_name
-        data_type: string
+        data_type: String
         description: "The name of the section this subsection belongs to, with subsection_number prepended"
       - name: course_order
         data_type: Int32