@node-lambdas/cli
v2.0.3
Published
This CLI can be used to run or test [cloud functions](https://github.com/node-lambdas) from a terminal.
Downloads
149
Readme
@node-lambdas/cli
This CLI can be used to run or test cloud functions from a terminal.
How to install
Via NPM is the easiest:
npm i @node-lambdas/cli
If you don't have access to NPM registry, download the JS script and make it executable:
curl -o fn.mjs https://unpkg.com/@node-lambdas/cli@latest/fn.mjs
chmod +x fn.mjs
Introduction
A reusable cloud function is basically an HTTPS server:
input >> POST https://[service].jsfn.run/[function] >> output
For example: you have a JSON and you want to convert it to YAML, but you cannot install a program for that.
cat file.json | fn yaml encode | tee file.yml
Can I use it with cURL?
Absolutely!
curl -d @file.json https://yaml.jsfn.run/encode
Examples
# convert JSON to YAML
cat your.json | fn yaml
# calculate the sha256 hash for the content of file.txt
cat file.txt | fn sha 256
# convert Markdown to HTML
cat README.md | fn markdown
Run
Running any publicly available lambda or run your new project on localhost
Calling fn
with a function name will pipe stdin
to a cloud function and pipe the output back to stdout
.
echo -n 'input' | fn [name]
# pipe file.txt to a lambda running in your machine
cat file.txt | fn +local
# pipe file.txt to a lambda running in your machine that uses port 2000
cat file.txt | fn +local +port=2000
You can see the available functions and read more in the index repository.
Run without input pipe
Usually you would use the syntax a | fn b
to pipe data into a function.
You can also add input options to fn
instead, and bypass stdin.
Use +data='input data'
to send the input data instead.
You can also read from a file, using the +stdin=path/to/file
, or the shortcut, @path/to/file
.
For example:
fn data='hello' | fn base64
fn @input.txt | fn sha 512
fn +stdin=image.png | fn resize 1024x768
API Authentication
Some functions need credentials in order to run API calls.
These credentials can be stored in a file called credentials.json
.
The format is as follows:
- the top level keys are authentication groups. You use this name with
+auth
. - each key under a group is a function name
- each key under a function maps to an HTTP header that the function needs.
For example: let's say you want to call the same function with different credentials:
{
"default" {
"function-name": {
"access-key-id": "key-123",
"access-key": "cac54b35c4ab6a5c46abb"
},
"another-name": {
"authentication": "bearer cac54b35c4ab6a5c46abb"
},
},
"bob-credentials": {
"function-name": {
"access-key-id": "key-456",
"access-key": "4524bc45ba54ba45b25c2"
},
},
}
You can invoke the same function passing the authentication group to use:
# Using default credentials
echo 'let me in' | fn +auth function-name
# Using both default and 'bob-credentials' to pipe from one function call to another. The third call will not use any credentials
echo 'let me in' | fn +auth function-name | fn +auth=bob-credentials function-name | fn another-name
All options
| option | description |
| ---------------- | ------------------------------------------------------------------------ |
| +auth=[name]
| Name of an authentication group to use from './credentials.json' |
| +port=[number]
| The http port to use when calling a local server with a running function |
| +data=[data]
| Use the data passed as argument instead of stdin
for next step |
| +nodata
| Run the cloud function without any input |
| +info name
| Shows more details about a cloud function |
| +json
| Show cloud function API as JSON |