Make clean_stale_db_objects configurable

Also update local development instructions in README

README.md

@@ -12,38 +12,25 @@ An Airflow-based dashboard for the LA Metro ETL pipeline!
 
 Perform the following steps from your terminal.
 
-1. Clone this repository and its submodule, then `cd` into the superproject.
+1. Clone [the LA Metro Councilmatic repository](ttps://github.com/Metro-Records/la-metro-councilmatic) and follow the instructions in its
+README to build and run the application.
 
-    ```bash
-    git clone --recursive https://github.com/Metro-Records/la-metro-dashboard.git
-    cd la-metro-dashboard
-    ```
-2. Build `la-metro-dashboard` application, and create a local `.env` file. Fill
-in the absolute location of your GPG keyring, usually the absolute path for ` ~/.gnupg`.
+2. Clone this repository and create a local `.env` file. 
 
-    ```bash
-    docker-compose build
-    cp .env.example .env
-    # Fill in the correct value for GPG_KEYRING_PATH
-    ```
-
-3. Once the command exits, follow the instructions to build the [LA Metro Councilmatic application](https://github.com/Metro-Records/la-metro-councilmatic#setup)
+```bash
+cp .env.example .env
+```
 
-4. In order to run the `la-metro-dashboard` application, the `la-metro-councilmatic`
-app must already be running. Open a new shell, move into the `la-metro-councilmatic`
-application, and run it.
+Fill in the absolute location of your GPG keyring, usually the absolute path for ` ~/.gnupg`.
 
-	```bash
-    cd la-metro-councilmatic && docker-compose up app
-    ```
 
-	Once la-metro-councilmatic is running, in your first shell, run the la-metro-dashboard application.
+3. Build and run the dashboard:
 
-	```bash
-	docker-compose up
-	```
+```bash
+docker-compose up
+```
 
-5. Finally, to visit the dashboard app, go to http://localhost:8080/admin/. The
+4. Finally, to visit the dashboard app, go to http://localhost:8080/admin/. The
 Councilmatic app runs on http://localhost:8001/.
 
 See the Airflow documentation for more on [navigating the UI](https://airflow.apache.org/docs/stable/ui.html)

dags/clean_stale_db_objects.py

@@ -1,14 +1,14 @@
 from datetime import timedelta
 
-from airflow import DAG
+from airflow.decorators import dag, task
 
 from constants import (
     LA_METRO_DATABASE_URL,
     LA_METRO_SEARCH_URL,
     LA_METRO_DOCKER_IMAGE_TAG,
     LA_METRO_STAGING_DATABASE_URL,
     START_DATE,
-    LA_SCRAPERS_IMAGE_URL
+    LA_SCRAPERS_IMAGE_URL,
 )
 from operators.blackbox_docker_operator import BlackboxDockerOperator
 
@@ -41,16 +41,31 @@
     },
 }
 
-with DAG(
-    "clean_stale_db_objects",
-    default_args=default_args,
+
+@dag(
     schedule_interval="0 0 * * 0",
     description="Deletes objects from the database that have not"
     "been seen in a recent scrape",
-) as dag:
+    default_args=default_args,
+    params={"window": 7, "max": 25, "report": False},
+)
+def clean_stale_db_objects(window=7, max=25, report=False):
+    @task
+    def get_flags(**kwargs):
+        if kwargs["params"]["report"]:
+            return "--report"
+        else:
+            return f"--window={kwargs['params']['window']} --max={kwargs['params']['max']} --yes"
 
-    BlackboxDockerOperator(
+    flags = get_flags()
+
+    pupa_clean = BlackboxDockerOperator(
         task_id="clean_stale_db_objects",
         environment=docker_base_environment,
-        command="pupa clean --noinput",
+        command=f"pupa clean {flags}",
     )
+
+    flags >> pupa_clean
+
+
+clean_stale_db_objects()

docker-compose.yml

@@ -82,3 +82,4 @@ volumes:
 networks:
   app_net:
     name: la-metro-councilmatic_default
+    external: true