[codex] Add scenario runner dbt step

docs/runbooks/scenario-runner.md

@@ -4,7 +4,7 @@
 
 The scenario runner executes end-to-end managed-warehouse flows against a configured dev environment. The first smoke scenario provisions a warehouse, waits for readiness, runs `SELECT 1` over PGWire with managed-hostname SNI, then deprovisions and verifies cleanup.
 
-The frozen metadata and perf scenarios provision the same fresh dev warehouse shape, create read-only views over frozen persons/events parquet supplied by `DUCKGRES_SCENARIO_FROZEN_S3_URI`, run their workload, then deprovision.
+The frozen metadata, perf, and dbt scenarios provision the same fresh dev warehouse shape, create read-only views over frozen persons/events parquet supplied by `DUCKGRES_SCENARIO_FROZEN_S3_URI`, run their workload, then deprovision.
 
 ## Required Environment
 
@@ -26,6 +26,7 @@ export DUCKGRES_SCENARIO_OUTPUT_BASE="artifacts/scenario"
 export DUCKGRES_SCENARIO_RUN_ID="scenario-smoke-manual"
 export DUCKGRES_SCENARIO_PG_PORT="5432"
 export DUCKGRES_SCENARIO_PG_CONNECT_TIMEOUT="10"
+export DUCKGRES_SCENARIO_DBT_BIN="dbt"
 export DUCKGRES_SCENARIO_MAX_RUNTIME="30m"
 export DUCKGRES_SCENARIO_GO_TEST_TIMEOUT="60m"
 ```
@@ -71,6 +72,12 @@ Run frozen perf queries:
 just scenario-frozen-perf
 ```
 
+Run frozen dbt lifecycle:
+
+```bash
+just scenario-frozen-dbt
+```
+
 Run a specific scenario file:
 
 ```bash
@@ -92,6 +99,13 @@ The frozen perf scenario uses:
 
 Perf artifacts are written under `artifacts/scenario/<run_id>/perf/` using the existing `tests/perf/core` artifact schema, including `query_results.csv`, `summary.json`, and `server_metrics.prom`.
 
+The frozen dbt scenario uses:
+
+- `tests/scenario/scenarios/posthog_frozen_dbt.yaml`
+- `tests/scenario/dbt/posthog_frozen_project/`
+
+dbt artifacts are written under `artifacts/scenario/<run_id>/dbt/`, including per-command stdout/stderr logs, `target/` artifacts, and dbt logs. Install `dbt-postgres` locally or set `DUCKGRES_SCENARIO_DBT_BIN` to the dbt executable to use.
+
 `DUCKGRES_SCENARIO_FROZEN_S3_URI` must point at a dev-owned frozen dataset prefix with `persons/` and `events/` parquet children.
 The provisioned Duckgres worker role also needs read/list access to that prefix; the runner process only supplies the URI, while the worker performs the S3 reads during `read_parquet`.
 

justfile

@@ -384,6 +384,11 @@ scenario-frozen-metadata:
 scenario-frozen-perf:
     ./scripts/scenario_run.sh tests/scenario/scenarios/posthog_frozen_perf.yaml
 
+# Run the dev frozen dataset dbt scenario
+[group('test')]
+scenario-frozen-dbt:
+    ./scripts/scenario_run.sh tests/scenario/scenarios/posthog_frozen_dbt.yaml
+
 # Lint (matches CI — uses golangci-lint, not go vet)
 [group('test')]
 lint:

scripts/scenario_run.sh

@@ -20,6 +20,7 @@ Optional environment:
   DUCKGRES_SCENARIO_PG_CONNECT_TIMEOUT
   DUCKGRES_SCENARIO_FLIGHT_ADDR (required by frozen perf scenarios)
   DUCKGRES_SCENARIO_FLIGHT_INSECURE_SKIP_VERIFY
+  DUCKGRES_SCENARIO_DBT_BIN
   DUCKGRES_SCENARIO_MAX_RUNTIME
   DUCKGRES_SCENARIO_GO_TEST_TIMEOUT
   DUCKGRES_SCENARIO_FROZEN_S3_URI (required by frozen dataset scenarios)

tests/scenario/dbt/errors.go

@@ -0,0 +1,28 @@
+package dbt
+
+const (
+	ErrorClassConfig          = "dbt_config"
+	ErrorClassDBT             = "dbt_execution"
+	ErrorClassUnsupportedStep = "unsupported_step"
+)
+
+type classifiedError struct {
+	class string
+	err   error
+}
+
+func (e classifiedError) Error() string {
+	return e.err.Error()
+}
+
+func (e classifiedError) Unwrap() error {
+	return e.err
+}
+
+func (e classifiedError) ErrorClass() string {
+	return e.class
+}
+
+func classified(class string, err error) error {
+	return classifiedError{class: class, err: err}
+}

tests/scenario/dbt/posthog_frozen_project/dbt_project.yml

@@ -0,0 +1,13 @@
+name: posthog_frozen_scenario
+version: "1.0.0"
+profile: duckgres_scenario
+
+model-paths: ["models"]
+test-paths: ["tests"]
+
+models:
+  posthog_frozen_scenario:
+    staging:
+      +materialized: view
+    marts:
+      +materialized: table

tests/scenario/dbt/posthog_frozen_project/models/marts/daily_event_counts.sql

@@ -0,0 +1,6 @@
+SELECT
+    date_trunc('day', event_timestamp) AS event_day,
+    event,
+    count(*) AS events
+FROM {{ ref('stg_events') }}
+GROUP BY 1, 2

tests/scenario/dbt/posthog_frozen_project/models/marts/person_event_rollup.sql

@@ -0,0 +1,8 @@
+SELECT
+    person_id,
+    count(*) AS events,
+    min(event_timestamp) AS first_event_timestamp,
+    max(event_timestamp) AS last_event_timestamp
+FROM {{ ref('stg_events') }}
+WHERE person_id IS NOT NULL
+GROUP BY 1

tests/scenario/dbt/posthog_frozen_project/models/schema.yml

@@ -0,0 +1,38 @@
+version: 2
+
+models:
+  - name: stg_persons_daily
+    columns:
+      - name: person_day
+        tests:
+          - not_null
+      - name: persons
+        tests:
+          - not_null
+
+  - name: stg_events
+    columns:
+      - name: event
+        tests:
+          - not_null
+      - name: event_timestamp
+        tests:
+          - not_null
+
+  - name: daily_event_counts
+    columns:
+      - name: event_day
+        tests:
+          - not_null
+      - name: events
+        tests:
+          - not_null
+
+  - name: person_event_rollup
+    columns:
+      - name: person_id
+        tests:
+          - not_null
+      - name: events
+        tests:
+          - not_null

tests/scenario/dbt/posthog_frozen_project/models/sources.yml

@@ -0,0 +1,8 @@
+version: 2
+
+sources:
+  - name: frozen_v1
+    schema: frozen_v1
+    tables:
+      - name: persons_file_view
+      - name: events_file_view

tests/scenario/dbt/posthog_frozen_project/models/staging/stg_events.sql

@@ -0,0 +1,6 @@
+SELECT
+    event,
+    person_id,
+    "timestamp" AS event_timestamp
+FROM {{ source('frozen_v1', 'events_file_view') }}
+WHERE "timestamp" IS NOT NULL

tests/scenario/dbt/posthog_frozen_project/models/staging/stg_persons_daily.sql

@@ -0,0 +1,6 @@
+SELECT
+    date_trunc('day', _timestamp) AS person_day,
+    count(*) AS persons
+FROM {{ source('frozen_v1', 'persons_file_view') }}
+WHERE _timestamp IS NOT NULL
+GROUP BY 1

tests/scenario/dbt/posthog_frozen_project/profiles.yml

@@ -0,0 +1,14 @@
+duckgres_scenario:
+  target: dev
+  outputs:
+    dev:
+      type: postgres
+      host: "{{ env_var('DUCKGRES_DBT_HOST') }}"
+      port: "{{ env_var('DUCKGRES_DBT_PORT') | int }}"
+      user: "{{ env_var('DUCKGRES_DBT_USER') }}"
+      pass: "{{ env_var('DBT_ENV_SECRET_DUCKGRES_PASSWORD') }}"
+      dbname: "{{ env_var('DUCKGRES_DBT_DBNAME') }}"
+      schema: "{{ env_var('DUCKGRES_DBT_SCHEMA') }}"
+      sslmode: "{{ env_var('DUCKGRES_DBT_SSLMODE', 'require') }}"
+      connect_timeout: "{{ env_var('DUCKGRES_DBT_CONNECT_TIMEOUT', '10') | int }}"
+      threads: 1