Fopsy NPM | npm.io

Readme

IMPORTANT: Changes since version 1.0.0
Fopsy now an ESM module! (Breaking change!)
New methods: mkgitignore, mkscript, runscript

synchronous file operations

Lightweight wrappers around synchchronous file operations in NodeJS.

CommonJS (before version 1.0.0):

const fopsy = require("fopsy") // all exports
const {exec} = require("fopsy") // selected expors

ESM (since version 1.0.0):

import fopsy from "fopsy" // all exports
import {mkfile, rmfile} from "fopsy" // selected exports

mkfolder(path)

Create a folder recursevly at given path.

mkfile(path, content, action = "w", mode = 0o744)

Create a file at given path. Any folders are created recursevly along the way.

The action argument sets the file interaction flag like 'a' as in 'append'. (Default is 'w' as in 'write'.) See this list for all supported flags.

The mode argument sets the file permissions (chmod).

mkscript(filepath, contents, environment)

Create a shell script file at given filepath with given contents.

The file gets only created if environment argument is either missing, or, if its value equals to NODE_ENV. Meaning, you can generate scripts based on the current Node environment.

For example, if your environment is NODE_ENV=production and you create your script like this mkscript("/folder/file.sh", "echo 'test'", "development") then the script will not be created because the script is meant to be created only for "development" environments.

This is useful if you have a preflight routine that generates scripts for your and you have multiple application running the same code. Based on the environment argument of the function, you could easily run the same code on "production" and "development" and only "production" would create the required file from the example above.

mkgitignore(path, rules)

Generates a .gitignore file with given rules.

The neat thing about this function is that if there already IS a gitignore file, then it will extend it with given rules, without duplicated entries!

Since the contents of the '.gitignore' file are auto-generated, the function call will also automatically untrack itself from Git, by adding a .gitignore rule to the generated file. - This does not apply if a gitignore file already existed and had a * (ignore all) rule, because in this case everything is untracked anyways.

rmfolder(path) and rmfile(path)

Delete a folder or a file recursevly at given path.

catfolder(path, encoding)

Read file(s) at given path (path can be a string or an array of strings).

The return value is an array of objects. Each object represents one file that has been read. For example:

[
    {
        content: <Buffer 68 65 6c 6c 6f 20 77 6f 72 6c 64 0a>,
        encoding: undefined,
        mime: 'text/plain',
        size: {
            value: 12,
            byte: [Getter/Setter],
            kilobyte: [Getter/Setter],
            megabyte: [Getter/Setter],
            gigabyte: [Getter/Setter]
        }
    }
]

Without an encoding argument the content property will be a Buffer! - With an encoding argument, the content will become a string (encoded by encoding format) and the encoding property will be present. You can then use the content string as you like, or use buffers yourself to convert it into a different format. (See this documentation on encoding formats)

The size propery is an object with getters and setters and helps converting the file size into different units, like byte into megabyte and vice versa.

NOTE: If a file is missing then its content will equal null and its size will equal 0!

catfile(path, encoding)

Read file(s) at given path. - Technically, this function is just an alias to catfolder(), but instead of always returning an array of fetched files, this function returns an array only when path is passed as an array too! (If path is a string but it still points to a directory (potentially with many files in it), then the return value will also remain an array!) - Only when path is a string that points to a file, this function returns a single object. - So, this is basically a convenience function wrapper for .catfile().

sizeunit(byte)

This is a helper... but it is available too!

Calling it with an argument will set the default value in byte! For example: sizeunit(16) - From there on, you can convert the value into byte, kilobyte, megabyte or gigabyte, for example: console.log( sizeunit(16).megabyte )

You can also directly set a value in megabyte if you want, and then convert it back into byte. For example:

const mb = sizeunit()
console.log(mb) // default byte value is 0
mb.megabyte = 3 // setter will convert the value into byte internally
console.log(mb.byte)
console.log(mb.megabyte)

exec(command [, options])

This is a very light wrapper around the NodeJS execSync function and it lets you run shell commands. The function call accepts the exact same arguments as the original function, but it has a purpose and some additional value...

exec is able catch all of the errors and outputs from STDOUT, that your command produces during execution! The return value is an object with a boolean success (true, false) and a string stdout (the response received from STDOUT).

This way you can tell check if the command execution was sucessfully or not. And we can also parse the response of the command and react to any events accordingly.

Here are some use-case examples:

import {exec} from "fopsy"

const response = exec(`certbot certonly --webroot --non-interactive --agree-tos -m ${EMAIL} -w '${CHALLENGE_SAVEDIR}' -d ${DOMAIN} --work-dir '${ROOTPATH}' --deploy-hook '${RENEW_CALLBACK}'`)

if(response.success) {
    console.info(`Certbot requested a certificate renewal for '${DOMAIN}'. Server will be reloaded after the cert update.`)
    console.log(response.stdout)
} else {
    console.error(response.stdout)
    console.warn(`Certbot could not auto-renew the SSL certificate for '${DOMAIN}'! Please inspect the server logs and resolve this issues manually!`)
}

As you've noticed, the exec will not throw any errors or break your application. But sometimes throw new Error("command execution failed") is exactly what you need:

const response = exec(`certbot certonly...`)
if(!response.success) throw new Error("command failed with: ${response.stdout}")

If you want a more elegant and flexible way of asserting, feel free to use the assert function from my other package on NPM, called type-approve!

import {assert} from "type-approve"
import {exec} from "fopsy"
const run = cmd => assert(...Object.values(exec(cmd))) // just a custom shortcut to calling assert with condition and message arguments

try {
    const response = run(`certbot certonly...`)
    console.log("Success: " + response.stdout)
} catch(error) {
    console.error(error)
}

runscript(cmd)

Runs a command line call, for example: runscript("echo 'hello world'"). The great thing is that the call is running through an assert call, meaning it will throw an error if the script exists with an stderr.

Basically the implimentation is as simple as assert(...Object.values(exec(cmd))). Just a light wrapper around exec().