databricks · andrewnester · Nov 28, 2024 · Dec 24, 2024 · Dec 24, 2024 · lennartkats-db
diff --git a/knowledge_base/databricks_app/.gitignore b/knowledge_base/databricks_app/.gitignore
@@ -0,0 +1,2 @@
+
+.databricks
diff --git a/knowledge_base/databricks_app/README.md b/knowledge_base/databricks_app/README.md
@@ -0,0 +1,24 @@
+# Databricks App for working with Databricks jobs
+
+This example demonstrates how to define an Databricks App in a Databricks Asset Bundle.
+
+It includes and deploys an example app and a job managed by DABs to a Databricks workspace.
+The app shows current status of the job and lists all existing runs.
+
+For more information about Databricks Apps, please refer to the [documentation](https://docs.databricks.com/en/dev-tools/databricks-apps/index.html).
+
+## Prerequisites
+
+* Databricks CLI v0.238.0 or above
+
+## Usage
+
+Modify `databricks.yml`:
+* Update the `host` field under `workspace` to the Databricks workspace to deploy to.
+
+Run `databricks bundle deploy` to deploy the app.
+
+Run `databricks bundle run job_manager` to start the app and execute app deployment. 
+When you change app code or config, you need to run `databricks bundle deploy` and `databricks bundle run job_manager` to update the app source code and configuration.
+
+Run `databricks bundle open` to navigate to the deployed app in your browser. Alternatively, run `databricks bundle summary` to display its URL.
diff --git a/knowledge_base/databricks_app/databricks.yml b/knowledge_base/databricks_app/databricks.yml
@@ -0,0 +1,14 @@
+bundle:
+  name: databricks_app
+
+include:
+  - resources/*.yml
+
+
+#workspace:
+#  host: https://myworkspace.databricks.com
+
+targets:
+  dev:
+    default: true
+    mode: development
diff --git a/knowledge_base/databricks_app/resources/hello_world.job.yml b/knowledge_base/databricks_app/resources/hello_world.job.yml
@@ -0,0 +1,16 @@
+resources:
+  jobs:
+    hello_world:
+      name: Serverless Hello World job
+      tasks:
+        - task_key: task
+          spark_python_task:
+            python_file: ../src/job/main.py
+          environment_key: default
+
+      environments:
+        - environment_key: default
+          spec:
+            client: "1"
+            dependencies:
+              - cowsay
diff --git a/knowledge_base/databricks_app/resources/job_manager.app.yml b/knowledge_base/databricks_app/resources/job_manager.app.yml
@@ -0,0 +1,29 @@
+resources:
+  apps:
+    job_manager:
+      name: "job_manager"
+      description: "App which manages job created by this bundle"
+
+      # The location of the source code for the app.
+      source_code_path: ../src/app
+
+      # The configuration for the app. This allows to pass environment variables to the app and
+      # configure how to run the app.
+      config:
+        command:
+          - flask
+          - --app
+          - app
+          - run
+          - --debug
+        env:
+          - name: JOB_ID
+            value: ${resources.jobs.hello_world.id}
+
+      # The resources which this app have an access to.
+      resources:
+        - name: "app-job"
+          description: "A job for app to be able to work with"
+          job:
+            id: ${resources.jobs.hello_world.id}
+            permission: "CAN_MANAGE_RUN"
diff --git a/knowledge_base/databricks_app/src/app/app.py b/knowledge_base/databricks_app/src/app/app.py
@@ -0,0 +1,25 @@
+import os
+
+from databricks.sdk import WorkspaceClient
+from flask import Flask, render_template, redirect, url_for
+
+app = Flask(__name__)
+
+
+@app.route("/")
+def home():
+    job_id = os.getenv("JOB_ID")
+
+    w = WorkspaceClient()
+    job = w.jobs.get(job_id)
+    runs = w.jobs.list_runs(job_id=job_id)
+    return render_template("index.html", job_name=job.settings.name, runs=runs)
+
+
+@app.route("/run")
+def run_job():
+    job_id = os.getenv("JOB_ID")
+
+    w = WorkspaceClient()
+    w.jobs.run_now(job_id=job_id)
+    return redirect(url_for("home"), code=302)
diff --git a/knowledge_base/databricks_app/src/app/templates/index.html b/knowledge_base/databricks_app/src/app/templates/index.html
@@ -0,0 +1,15 @@
+<html>
+<head>
+    <title>Databricks App managed by DABs</title>
+</head>
+<body>
+    <h1>This app manages job "{{job_name}}"</h1>
+    <h2>Job runs</h2>
+    <ul>
+        {% for run in runs %}
+            <li>{{ run.job_run_id }}: {{ run.status.state }} (<a href="{{ run.run_page_url }}">Open this job run</a>)</li>
+        {% endfor %}
+    </ul>
+    <a href="/run">Run job</a>
+</body>
+</html>
diff --git a/knowledge_base/databricks_app/src/job/main.py b/knowledge_base/databricks_app/src/job/main.py
@@ -0,0 +1,3 @@
+from cowsay import cow
+
+cow("Hello, world!")