npkn lets you deploy code to Napkin.io from your local command line. It is currently available as an alpha release and is subject to change.
npkn help
Show available commands, npkn install location, and current version.
npkn new <name> -r,--runtime=[:runtime] -w,--workspace=[:workspacename]
Create a new Napkin function.
Arguments
---------
<name> The name of the new function
Options
-------
-r, --runtime [:runtime]
The runtime to use. One of:
- python3.8
- node14.x
If no runtime is provided, npkn will use whatever runtime is set via the
NPKN_DEFAULT_RUNTIME environment variable. If neither is set, npkn will exit.
-w, --workspace [:workspacename]
[EXPERIMENTAL] The name of the workspace in which to create the new function. Defaults to
the default account workspace if not provided.
npkn pull <name>
Copy an existing function from napkin.io to your local machine's working directory.
Arguments
---------
<name>
The name of the function to pull. If pulling from a non-default workspace, use the following syntax:
<workspacename:functionname>. For example, if pulling a function called 'foo' from a workspace called 'bar':
npkn pull foo:bar
npkn run -n,--name=[:name]
Run a function on Napkin and get the result
Options
-------
-n, --name [:name]
Name of function to run. The function named [:name] must be in the
current working directory. If no name is provided, npkn will look for a
metadata.yaml file and function.py file in the current working directory
and run that. If neither is found, npkn will exit.
npkn deploy -n,--name=[:name]
Deploy local changes to napkin.io
Options
-------
-n, --name [:name]
Name of function to deploy. The function named [:name] must be in the
current working directory. If no name is provided, npkn will look for a
metadata.yaml file and function.py file in the current working directory
and deploy that. If neither is found, npkn will exit.
pip install git+https://github.com/NapkinHQ/npkn-py.git
- First, fill out the Google Form for access. We will send your credentials using the email you used to sign up for Napkin.
- Once you've received your credentials, set the following environment variables:
NPKN_ACCOUNT_ID=xxxxxxx
NPKN_SECRET_KEY=xxxxxxx
- Install Poetry (if not already installed)
- Clone and initialize the repo
git clone https://github.com/NapkinHQ/npkn-py.git
cd npkn-py
poetry install
- Try out via Poetry shell
poetry shell
> npkn help
poetry run pytest
These environment variables are optional and mostly intended for development purposes:
NPKN_LOG_LEVEL
NPKN_API_HOST
Create a new Napkin Python function
>>> npkn new -r python3.8 tell-time
Creating new function 'tell-time' ✅
>>> cd tell-time && ls -la
drwxrwxr-x 2 user user 42 Jun 23 22:32 .
drwxrwxr-x 4 user user 59 Jun 23 22:32 ..
-rw-rw-r-- 1 user user 90 Jun 23 22:32 function.py
-rw-rw-r-- 1 user user 97 Jun 23 22:32 meta.yaml
Run the function
>>> npkn run
Function: tell-time | Workspace: default
{'body': 'hello, world!', 'headers': {'content-type': 'text/html'}, 'logs': [], 'returnValue': 0, 'statusCode': 200, 'status_code': 200}
Update function code and re-run
function.py
from napkin import response
from datetime import datetime
response.body = f"The current time is {datetime.now()}">>> npkn run
Function: tell-time | Workspace: default
Syncing code ✅
{'body': 'The current time is 2022-06-23 22:37:23.676056', 'headers': {'content-type': 'text/html'}, 'logs': [], 'returnValue': 0, 'statusCode': 200, 'status_code': 200}
Use 3rd party modules
To use modules from PyPI, list the function module requirements in your meta.yaml file. You can check which modules are available from the Napkin UI.
meta.yaml
name: tell-time
runtime: python3.8
uid: <your user id>
workspace: <your workspace id>
modules:
pytz: "*"Update function code
function.py
from napkin import response
from datetime import datetime
from pytz import timezone
tz = timezone("America/Costa_Rica")
current_tz_time = tz.localize(datetime.now())
response.body = f"The current time in Costa Rica is {current_tz_time}"Run it again!
>>> npkn run
Function: tell-time | Workspace: default
Syncing code ✅
Syncing modules ✅
{'body': '"The current time in Costa Rica is 2022-06-23 23:00:26.558193-06:00"', 'headers': {'content-type': 'text/html'}, 'logs': [], 'returnValue': 0, 'statusCode': 200, 'status_code': 200}
Use Environment Variables
Add your environment variables to the env field of your meta.yaml file.
meta.yaml
name: tell-time
runtime: python3.8
uid: <your user id>
workspace: <your workspace id>
modules:
pytz: "*"
env:
TZ: America/Costa_RicaThen use the environment variable as you normally would in your function.
function.py
from napkin import response
from datetime import datetime
from pytz import timezone
import os
tz = timezone(os.getenv("TZ"))
current_tz_time = tz.localize(datetime.now())
response.body = f"The current time in Costa Rica is {current_tz_time}"Running it again...
>>> npkn run
Function: tell-time | Workspace: default
Syncing code ✅
Syncing environment ✅
{'body': '"The current time in Costa Rica is 2022-06-23 23:02:26.5589233-06:00"', 'headers': {'content-type': 'text/html'}, 'logs': [], 'returnValue': 0, 'statusCode': 200, 'status_code': 200}
Deploying Code
When you're ready to deploy your code to your function's endpoint, just run npkn deploy.
>>> npkn deploy
Function: tell-time | Workspace: default
Deploying ✅
>>> curl https://my-account.npkn.net/tell-time
The current time in Costa Rica is 2022-06-23 23:08:27.104832-06:00
Under the hood, npkn-py uses Napkin's public REST API (currently in alpha and subject to changes) to interact with the Napkin platform. Official documentation for the public API remains under development; however, below is included a brief API reference to help users that wish to fork or contribute to this repo.
The following headers should be set in the API request.
napkin-api-key: xxxxxx
napkin-user-uid: xxxxx
The values for the headers are the same as the account id and secret key that you use for the Napkin CLI.
Get user account object
GET /api/1/account
Response
{
"workspaces": [
{
"members": [
// ...
],
"name": "xxx",
"objects": {
"xxxxxxxxx": {
"description": null,
"name": "testing2",
"path": "/",
"requireAuth": "0",
"runtime": "python3.8",
"type": "napkin",
"uid": "xxxxxxxxxx",
"version": "v1",
"visibility": "private"
},
// ...
},
"runtimes": {
"nodejs14.x": "PROVISIONED",
"python3.8": "PROVISIONED"
},
"type": "PRO_TEAM",
"uid": "xxxxxxxxx",
"version": "v1"
}
]
}Sync function modules
POST /api/1/sync/modules
Request Body
{
"modules": {
"requests": "*"
},
"function_uid": "xxxxx"
}Response
{
"success": true,
"modules": {
"requests": "2.9.2"
}
}Sync function environment variables
POST /api/1/sync/env
Request Body
{
"env": {
"MY_VAR": "hello"
},
"function_uid": "xxxxx"
}Response
{
"env": {
"MY_VAR": "hello"
}
}Sync function code (pre-deploy)
POST /api/1/sync/code
Request Body
{
"code_str": "print(\"hello world\")\n",
"function_uid": "xxxxx"
}Response
{
"function": {
// ...
}
}Retrieve a function object
GET /api/1/function
Request Args
uid=xxxxx
Response
{
"function": {
// function properties
},
"env": {
"MY_VAR": "hello"
}
}Create a new Napkin function
POST /api/1/new
Request Body
{
"runtime": "python3.8",
"workspace_id": "xxxxxx",
"function_name": "my-new-function"
}Response
{
"napkin": {
// function properties
},
"workspace": {
// function workspace properties
}
}Run a Napkin function using current, un-deployed code
POST /api/1/run
Request Body
{
"uid": "functionUidXXX",
"workspaceUid": "workspaceUid"
}Response
{
"logs": [],
"statusCode": 200,
"status_code": 200, // duplicated for legacy reasons
"headers": {},
"body": {}
}Deploy Napkin function code
POST /api/1/deploy
Request Body
{
"uid": "functionUidXXX"
}Response
{
"success": true
}