Deploying a CloudFlare Worker

Tools to install

First, install nvm

curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.37.2/install.sh | bash

Then, install npm with nvm. If you don't do this, will be really hard to install the other tools

nvm install node

Afterwards, install wrangler

sudo -H npm i @cloudflare/wrangler -g

Next, login to Cloudflare to authenticate with wrangler

wrangler login

Creating a worker

Initiating the worker

First, initiate a worker. The command below creates a new directory and initiates a dummy project

wrangler generate $NAME [$TEMPLATE] [--type=$TYPE] [--site]

$NAME is the name of the workers project. Both the name of the directory and the name property in the wrangler.toml config file

Adding data to your KV namespace

First, create a namespace through the web view.

When you have the id of the namespace, add it to the wrangler.toml file in the format below

kv-namespaces = [
  {binding = "BINDING_1", id = "ID_1"},
  {binding = "BINDING_2", id = "ID_2"},
  ...
]

Also create a preview namespace, so you can test it out on local dev

wrangler kv:namespace create "BINDING_1" --preview

And then add the generated PREVIEW_ID to wrangler.toml too

kv-namespaces = [
  {binding = "BINDING_1", preview_id = "PREVIEW_ID_1", id = "ID_1"},
  {binding = "BINDING_2", preview_id = "PREVIEW_ID", id = "ID_2"},
  ...
]

To do a bulk upload of data, use the following

wrangler kv:bulk put --binding=BINDING_NAME FILENAME

Where FILENAME is a JSON file that contains the data in the [{key: value}, {key: value}, ... ] format

And to put the same file in the preview namespace

wrangler kv:bulk put --binding=BINDING_NAME --preview FILENAME

Testing out the code

wrangler dev #starts a local server

To test out whether it works, run the following codeblock in python

import requests
url = 'http://127.0.0.1:8787'
r = requests.get(url, json=JSON_PAYLOAD)
r.content #shows the value returned. Can also run r.json()

Publishing the code in production

wrangler publish

Looking at production logs

You can use wrangler tail to look at logs in real-time. But before that, cloudflared must be installed. To install it, just run

curl https://bin.equinox.io/c/VdrWdbjqyF/cloudflared-stable-darwin-amd64.tgz | tar xzC /usr/local/bin