Skip to content

Latest commit

 

History

History
95 lines (62 loc) · 2.43 KB

API.md

File metadata and controls

95 lines (62 loc) · 2.43 KB
Raw

API

The library exposes a function to write directly to Stackdriver from your own application. The example below shows how this can be done using pino-multi-stream.

Example:

const stackdriver = require('pino-stackdriver')
const pinoms = require('pino-multi-stream')
// create the stackdriver destination stream
const credentials = '/credentials.json'
const projectId = 'my-project'
const writeStream = stackdriver.createWriteStream({ credentials, projectId })
// create pino loggger
const logger = pinoms({ streams: [{ stream: writeStream }] })
// log some events
logger.info('Informational message')
logger.error(new Error('things got bad'), 'error message')

Functions

createWriteStream

The createWriteStream function creates a writestream that pino-multi-stream can use to send logs to.

Example:

const writeStream = stackdriver.createWriteStream({
  credentials: '/credentials.json',
  projectId: 'my-project'
})

credentials

Type: String or { client_email: String, private_key: String } (optional)

Full path to the JSON file containing the Google Service Credentials, you can also use an object parameter with the client_email and the private_key instead of the path. Defaults to the GOOGLE_APPLICATION_CREDENTIALS environment variable. At least one has to be available.

projectId

Type: String (required)

The name of the project.

logName

Type: String (optional)

The name of the log. Defaults to "pino_log".

resource

Type: { type: String, labels: Object } (optional)

The resource to send logs to. Defaults to { type: "global" }.

keys

Type: { [key]: key } (optional)

Customize additional fields to pull from log messages and include in meta. Currently supports httpRequest, trace. Defaults to { httpRequest: "httpRequest" }.

fallback

Type: Boolean (optional)

Set the gRPC fallback option for the Google Stackdriver API.

Batching

Log entries will automatically be batched by the Google cloud logging library that's being used internally using a fire and forget meganism.

Prefixing messages

Prefixing message is supported via a prefix property from the log data:

logger.info({ prefix: 'foo' }, 'Info message')
logger.child({ prefix: 'foo' }).info('Info message')

Will send the following JSON payload to Stackdriver:

{
  "prefix": "foo",
  "message": "[foo] Info message"
  // ...
}