@microbs.io/core

microbs-core

@microbs.io/core is a common library used by microbs-cli and microbs plugins. Contributors should use @microbs.io/core for things such as logging, configuration management, command execution, and HTTP requests to ensure a consistent experience in microbs.

Usage

You can import all the core modules as follows:

const { config, context, logger, rollout, state, utils } = require('@microbs.io/core')

Or you can limit your imports only to the modules you need. For example, if you only need access to config and logger, then you can use:

const { config, logger } = require('@microbs.io/core')

Reference

• config
• context
• logger
• rollout
• state
• utils

config

config provides a read-only interface to the microbs configuration file (config.yaml) and stores its values in memory as a read-only object.

config.load(filepath?: string)

Load, parse, and persist in memory the microbs config file, which is located by filepath. By default, filepath is $CWD/config.yaml, where $CWD is the current directory of the user running the microbs CLI. If the file doesn't exist, then the default filepath is $HOME/.microbs/config.yaml. Usually plugins don't need to run config.load() except during tests, because the microbs CLI will load the config file before it loads the plugins and store the filepath at context.get('path.config').

config.get(path?: string)

Get a value from the config object by its key path in the object. Because the config file can be nested, you can retrieve the values of nested keys by using dot notation. For example, given the following configuration in config.yaml...

deployment:
  plugins:
    observability: grafana-cloud

...you can use config.get('deployment.plugins.observability') to retrieve the value of the observability plugin, which the function would return as 'grafana-cloud'.

context

context provides information about the microbs execution environment. Values can be added to the context object, but they cannot be removed or changed once they are added.

The microbs CLI sets the following context variables:

• command - The command that the user ran from the microbs CLI (e.g. setup)
• args - The arguments that the user supplied with command (e.g. --kubernetes --log-level debug)
• path.cli - Absolute path to the directory of the microbs CLI
• path.user - Absolute path to the current working directory of the user running the microbs CLI
• path.config - Absolute path to the config file (config.yaml) used to store deployment inputs
• path.state - Absolute path to the state file (state.yaml) used to store post-deployment details
• path.env - Absolute path to the .env file (.env) used by microbs-secrets
• path.app - Absolute path to the installation directory of the application named in deployment.app
• path.plugin.kubernetes - Absolute path to the installation directory of the plugin in deployment.plugins.kubernetes
• path.plugin.observability - Absolute path to the installation directory of the plugin in deployment.plugins.observability
• path.plugin.alerts - Absolute path to the installation directory of the plugin in deployment.plugins.alerts

context.get(key?: string)

Get a value from the context object by its key path in the object. You can retrieve the values of nested keys by using dot notation just like in config.get().

context.set(key?: string)

Set a value in the context object by its key path in the object. You can set the values of nested keys by using dot notation just like in config.get().

logger

logger provides a standardized logger that can serialize objects, colorize outputs based on log levels, display verbose metadata, and filter outputs by log levels. Objects are serialized to JSON.

logger.debug(message: any)

Write a log message that will appear only when --log-level is set to a value that is greater than or equal to debug.

logger.info(message: any)

Write a log message that will appear only when --log-level is set to a value that is greater than or equal to info.

logger.warn(message: any)

Write a log message that will appear only when --log-level is set to a value that is greater than or equal to warn.

logger.error(message: any)

Write a log message that will appear only when --log-level is set to a value that is equal to error.

rollout

rollout rolls out a skaffold profile to the deployed Kubernetes cluster.

rollout.run(opts: object)

Roll out a skaffold profile to the deployed Kubernetes cluster.

What happens when this function is called:

1. microbs-secrets is deleted (using kubectl delete secret)
2. The state file is saved and overwritten (using state.save())
3. The .env file is saved and overwritten (using utils.objToEnv())
4. microbs-secrets is deployed using the new .env file (using kubectl create secret)
5. The skaffold profile is deployed with skaffold run

rollout.run() can be controlled with an opts object:

• opts.action - Can be "run" or "delete" to invoke either skaffold run or skaffold delete (Default: "run")
• opts.profile - Which skaffold profile to invoke from skaffold.yaml (Default: "main")
• opts.skaffoldFilepath - Path to the skaffold.yaml file to use (Default: )

state

state provides a read-write interface to the microbs state file (state.yaml) and stores its values in memory as a read-write object.

state.load(filepath?: string)

Load, parse, and persist in memory the microbs state file, which is located by filepath. By default, filepath is $CWD/state.yaml, where $CWD is the current directory of the user running the microbs CLI. If the file doesn't exist, then the default filepath is $HOME/.microbs/confstateig.yaml. Usually plugins don't need to run state.load() except during tests, because the microbs CLI will load the state file before it loads the plugins and store the filepath at context.get('path.state').

state.get(path?: string)

Get a value from the state object by its key path in the object. You can retrieve the values of nested keys by using dot notation just like in config.get().

state.set(path: string, value: string)

Set a value from the state object by its key path in the object. You can set the values of nested keys by using dot notation just like in config.get(). Note that this will not persist to the state file until you call state.save()

state.save(filepath: string)

Save the entire contents of the state object to the state file, optionally located by its filepath. Usually plugins don't need to set filepath except during tests, because the microbs CLI will locate the state file and store the filepath at context.get('path.state').

utils

utils provides commonly used functions to simplify the development of plugins and help ensure consistent behavior across plugins. The most commonly used functions for plugins include:

• utils.http() - Submit an HTTP request
• utils.exec() - Execute a synchronous command.
• utils.sanitize() - Escape an input for safe usage in utils.exec()
• utils.loadFile() - Load the contents of file
• utils.loadJson() - Load and parse the contents of a JSON file

utils.exec(command: string, hideStdout: boolean)

Execute a synchronous command. The command will be logged if --log-level is set to debug. Setting hideStdout to true lets you capture the contents of stdout and stderr, which is recommended in most cases.

Returns a result object with the following properties:

• result.code - The exit code of the process
• result.stdout - The contents of stdout
• result.stderr - The contents of stderr if present
• result.err - The error object if an exception was thrown

Note: You should use utils.sanitize() to escape potentially unsafe characters from any inputs that you pass to the command. If your command concatenates or interpolates other variables, those variables should be escaped with utils.sanitize().

Example usage:

const { utils } = require('@microbs.io/core')
const result = utils.exec('echo "hello world"', true)
//result = { code: 0, stdout: 'hello world\n' }

utils.expandvars(str: string, values: object)

Given a string (str) substitute the environment variables with an object of values (values).

Example usage:

const { utils } = require('@microbs.io/core')
const result = utils.expandvars('FOO=${BAR}', { BAR: 'BAZ' })
// result = 'FOO=BAZ'

utils.flatten(obj: object)

Convert a nested object to a flattened object. This function is not intended to be used directly by plugins, but it can be used to test and understand the behavior of the function. The microbs CLI uses this function when creating the .env file for microbs-secrets, normalizing the nested contents of the config.yaml and state.yaml files into a flat structure used for .env.

Example usage:

const { utils } = require('@microbs.io/core')
const output = utils.flatten({
  'a': {
    'b': {
      'c.d': 'foo'
    }
  }
})
// output = { 'a.b.c.d': 'foo' }

utils.http(opts?: object) (async)

Submit an asynchronous HTTP request. utils.http() is a lightweight wrapper around axios.request() that includes debugging logging for the request and response when --log-level is set to debug.

Example usage:

const { utils } = require('@microbs.io/core')
const result = await utils.http({
  method: 'get',
  url: 'https://microbs.io/'
})

utils.loadFile(filepath: string)

Read the contents of a file located by its filepath.

utils.loadJson(filepath: string)

Read and parse the contents of a JSON-formatted file located by its filepath.

utils.loadTemplate(filepath: string, values: object)

Read the contents of a file located by its filepath and substitute any environment variables. utils.expandvars() is what performs the environment variable substitution.

utils.loadTemplateJson(filepath: string, values: object)

Read and parse the contents of a JSON-formatted file located by its filepath and substitute any environment variables. utils.expandvars() is what performs the environment variable substitution.

utils.objToEnv(obj: object)

Serialize the contents of object to environment variable syntax suitable to be saved as an .env file. Field names are flattened and uppercased, and dots are replaced with underscores. This function is not intended to be used directly by plugins, but it can be used to test and understand the behavior of the function.

Example usage:

const { utils } = require('@microbs.io/core')
const output = utils.objToEnv({
  foo: {
    bar.baz: 'hello world'
  }
})
// output = 'FOO_BAR_BAZ=hello world'

utils.sanitize(value: string)

Escape potentially unsafe characters in a string (value) so that the value can be used safely as an input to utils.exec().

Example usage:

const { utils } = require('@microbs.io/core')
const output = utils.sanitize('foo;bar')
// output = 'foo\\;bar'

utils.sleep(ms: number)

Pause the execution of the current process by ms number of milliseconds.

Example usage:

const { utils } = require('@microbs.io/core')
utils.sleep(1000) // Wait for 1 second