SJSGI

SailsJS Gateway Interface Pass requests to sails through CGI

Description

CGI describes requests to a script with environment variables loosely according to RFC3875, depending on the implementation. This package uses those variables to make a SailsJS virtual request and output a CGI response.

Sails and the app are loaded, executed, and lowered on each request, but without running the bootstrap or HTTP server.

Installation

npm install sjsgi --save

Usage

#!/usr/bin/env node7
/**
 * cgi-bin/sails.cgi
 */

const gi = require('sjsgi')

// Default to production environment
if ( ! process.env.NODE_ENV) {
  process.env.NODE_ENV = 'production'
}

/**
 * Configure with a SailsJS config object
 * http://sailsjs.com/documentation/reference/configuration
 *
 * Note: These values take precedence over app/config/.
 */
let config = {
  // hooks: {},
  // blueprints: {},
  // ...
  // views: {}
}

/**
 * Configure with .sailsrc
 * http://sailsjs.com/documentation/concepts/configuration/using-sailsrc-files
 */
// let config = require('sails/accessible/rc')('sails')

// Load the app, make a virtual request, and lower the app.
gi.sails(config)

Errors

By default the CGIError level is set to 1, which outputs the safeMessage message. Use level 2 to output the actual message and level 3 for a stacktrace. See /lib/app/cgi/Error.js for levels and config options.

Example error levels:

#!/usr/bin/env node7

const gi = require('sjsgi')
const CGIError = require('sjsgi/lib/app/cgi/Error')

// Safe errors
// CGIError.init({
//   level: 1,
//   safeMessage: 'There was an error.'
// })

// Actual errors
// CGIError.init({
//   level: 2
// })

// Errors and stacktraces
CGIError.init({
  level: 3
})

gi.sails({})

Contribute

• Raise issues at https://github.com/jasonschmedes/sjsgi/issues
• Make pull requests at https://github.com/jasonschmedes/sjsgi/pulls
• Send suggestions to [email protected]

License

MIT License Copyright (C) 2017 Jason Schmedes