Rapid deployment

Now that you’ve tested your step locally, it’s time to deploy to the pinexq.net server!

Deploying your step

While the template main.py is on the root level of your project directory, this isn’t necessary, as our small example suggests. However, in order to deploy your new step, we will need to edit pinexq.toml to point to it. For example:

[project]
name = "test-project"
pinexq_endpoint = "jobs.api.pinexq.net"
entrypoint = "src/small_step.py"

[deployment]
resource_preset = "Small"
max_replicas = 1

To deploy our step, we then can run the following in our project’s root directory:

foo@bar: ~/my_project$ uvx --from pinexq-cli@latest deploy --api-key="<PINEXQ_API_KEY>"

Your first job

On the Portal, under Processing Steps, you should now see your processing step, as well as any others made available to you.

Since your step takes file input, your next step is to upload your WorkData file.

Finally, navigate to Jobs, and “Create job”. You will then be able to name your test job, and assign it your ProcessingStep. Don’t start it yet! Leave “Start immediately” unchecked, and click “Create”; you will then be able to navigate to your job, and configure your inputs there.

Workflows

The Workflow feature allows you to “glue together” multiple steps. Or, in this case, multiple copies of your step. Our low-code workflow configurator lets you configure workflows with a minimum of coding knowledge. First, select your input file with a DataExplicit.

Then, drag in a ProcessingStep, configure it, and connect it.

Finally, publish your Workflow, and try running it!