This sample replaces the App Engine Tasks Queue Pull Counter sample that used the taskqueue library available only in the App Engine Standard for Python 2.7 runtime.

The sample uses a Pub/Sub topic with a pull queue as recommended in Migrating pull queues to Pub/Sub to perform the same functions as the earlier, Task Queue sample.

The application has three functions:

• Viewing the home page will display a form to specify a task name, and add one more request for it. It will also show all requested tasks and their counts.

• Submitting the form on the home page will queue a task request.

• Requesting the page at /_ah/start will begin processing the queued tasks. This will run without responding to the request until the request is cancelled, or the request times out, which can be a significant amount of time.

Before you can run or deploy the sample, you will need to do the following:

1. Enable the Cloud Pub/Sub API in the Google Developers Console.

2. Check that Firestore is in Datastore mode in the Google Developers Console, and select Datastore mode if it is not.

3. Select the project and account that will be used for locally run Google Cloud SDK commands:

   $ gcloud init

4. Create a topic and subscription.

    $ gcloud pubsub topics create [YOUR_TOPIC_NAME]
    $ gcloud pubsub subscriptions create [YOUR_SUBSCRIPTION_NAME] \
        --topic=[YOUR_TOPIC_NAME]
   

5. Update the environment variables in app3.yaml for Python 3, or app.yaml for Python 2.7.

When running locally, you can use the Google Cloud SDK to provide authentication to use Google Cloud APIs. Initialize the SDK for local commands if not already done.

$ gcloud init

Install dependencies, preferably with a virtualenv:

$ virtualenv env
$ source env/bin/activate
$ pip install -r requirements.txt

Set environment variables before starting your application:

$ export GOOGLE_CLOUD_PROJECT=[YOUR_PROJECT_NAME]
$ export TOPIC=[YOUR_TOPIC_NAME]
$ export SUBSCRIPTION=[YOUR_SUBSCRIPTION_NAME]

Run the application locally:

$ python main.py

In the current directory, edit the environment variables in app.yaml or app3.yaml, depending on whether you are going to use Python 2.7 or Python 3, and then deploy using gcloud.

For Python 2.7 you must first install the required libraries in the lib folder:

$ pip -t lib -r requirements.txt
$ gcloud app deploy app.yaml

For Python 3, you only need to run the deploy command:

$ gcloud app deploy app3.yaml

You can now access the application using the gcloud app browse command.