Skip to content

Commit

Permalink
still needs 1 - 2 more passes
Browse files Browse the repository at this point in the history
  • Loading branch information
@MylesBorins
MylesBorins committed Feb 24, 2020
1 parent c1a7cb8 commit 0fde0ab
Showing 1 changed file with 62 additions and 220 deletions.
282 changes: 62 additions & 220 deletions src/index.jade
View file
Expand Up @@ -158,7 +158,7 @@ html
h3 With your host
h4
a(href='https://twitter.com/MylesBorins') @MylesBorins
p Node+JS Interactive 2019
p Fullstack Academy

section(data-bespoke-backdrop='hello')
h1 Bonjour
Expand All @@ -169,9 +169,9 @@ html

section
h3 I am gainfully employed by Google as a Developer Advocate
h4 Focusing on the Node.js ecosystem and GCP
h4 Doing lots of JavaScript (and a few other things too)

img(src='images/gcp.png' alt='Google Cloud Platform' style='background-color:white;')
//- img(src='images/gcp.png' alt='Google Cloud Platform' style='background-color:white;')
h5.headroom The opinions expressed in this talk are solely my own

Expand Down Expand Up @@ -372,7 +372,7 @@ html
h2 To determine Resolution

section
h2 Node.js's CJS resolution algorithm supports:
h2 Node.js's CJS loader supports:
ul
li.hide
li Bare imports
Expand Down Expand Up @@ -403,11 +403,13 @@ html
"lodash": "/node_modules/lodash-es/lodash.js"
}
}
---
import 'lodash';

section
h2 import-maps
h4.headroom You can use in Chrome today
h2 chrome://flags
h2 chrome://flags/#enable-built-in-module-infra
h4.headroom Built-in module infra and import maps
//- section(data-bespoke-backdrop='danger3')
//- h3 ⚠️ DANGER ⚠️
Expand Down Expand Up @@ -440,25 +442,25 @@ html
h3 Denoted with Pragma

section
h2 Module goal has extra grammar changes
h2 ESModule goal has extra grammar changes
ul
li.hide
li No HTML Comments
li Await is a reserved keyword
li Divergence may increase over time

section
h3 ESModules do not have in source way
h2 of determining goal

section
h2 Potential Future Goals
ul
li.hide
li WASM
li HTML Modules
li WebPackage
li BinaryAST
//- section
//- h3 ESModules do not have in source way
//- h2 of determining goal
//-
//- section
//- h2 Potential Future Goals
//- ul
//- li.hide
//- li Wasm
//- li HTML Modules
//- li WebPackage
//- li BinaryAST
//- section
//- h1 So what happened next?
Expand All @@ -469,7 +471,7 @@ html
//- h3 ⚠️ DANGER ⚠️
//-
section
h3 ESM does not have a way
h3 ESModules does not have a way
h2 to inject lexically scope variables

section
Expand All @@ -486,6 +488,7 @@ html

section
h2 import.meta to the rescue
h3 🎉 `import.meta.url` 🎉

section
h1 You can use ESM in Node.js Today!
Expand All @@ -496,187 +499,21 @@ html
section
h2 in Node.js 13

section
h3 Recently backported (still with a flag)
h2 in Node.js 12

section
h2 About our current implementation
ul
li.hide
li Getting close to stable, but could still change
li ECMAScript compliant implementation
li ESM files need to use .mjs extension
li Unless you set type to module in package.json
li Support for Bare Imports
li Support for dynamic import in both ESM & CJS
li import.meta.url

section
h2 What are the limitations?
ul
li.hide
li Full file path is mandatory
li No support for folder resolution
li No named imports for `import 'cjs'`
li Only experimental support for `import 'json'`
li No support for `require('esm')`

//- section
//- h1 Interoperability
//-
//- section
//- h2 Loading ESM from CJS
//- ul
//- li.hide
//- li dynamic `import(esm)` in cjs
//- li Cannot require(esm)
//- section(data-bespoke-backdrop='grumpy')
//- h2 Some of our decisions may have made people grumpy
//- section
//- h3 Introducing
//- h2 The Node.js Modules Team
//-
//- section(data-bespoke-backdrop='modules-team-init')
//- h2 Formed 10 months ago
//-
//- section
//- h1 49 members
//-
//- section
//- h3 Representation from at least 10 countries:
//- ul
//- li.hide
//- li Canada
//- li China
//- li Germany
//- li Israel
//- li Italy
//- li Japan
//- li Nigeria
//- li United Kingdom
//- li South Africa
//- li USA
//-
//- section
//- h3 Representation from at least 10 corporate organizations:
//- ul
//- li.hide
//- li airbnb
//- li applitools
//- li Bloomberg
//- li Godaddy
//- li Google
//- li Groupon
//- li IBM
//- li Microsoft
//- li Mozilla
//- li NearForm
//- li Node Source
//-
//- section
//- h3 Key stakeholders from at least 14 OSS projects
//- ul
//- li.hide
//- li angular
//- li babel
//- li bluebird
//- li fastify
//- li libuv
//- li jQuery
//- li jspm
//-
//- section
//- h3 Key stakeholders from at least 14 OSS projects
//- ul
//- li.hide
//- li lodash
//- li Meteor
//- li npm
//- li polymer
//- li tapjs
//- li V8
//- li webpack
//-
//- section
//- h3 Members of the team participate
//- h3 in at least 3 standards bodies including:
//- ul
//- li.hide
//- li tc39
//- li whatwg
//- li w3c
//- section(data-bespoke-backdrop='kernel')
//- h1 Minimal Kernal
//-
//- section
//- h2 Phase 1
//- ul
//- li.hide
//- li Can only import ESM
//- li Must have .mjs file extension
//- li No support for Node.js specifier resolution algorithm
//-
//- section
//- h1 That's it
//- section
//- h1 CJS Interop
//-
//- section.small-code
//- pre
//- code.language-javascript.
//- // cjs.js
//- module.exports = 'cjs';
//-
//- // esm.mjs
//- import { createRequire } from 'module';
//-
//- const require = createRequire(import.meta.url);
//-
//- const cjs = require('./cjs');
//- cjs === 'cjs'; // true
//- section
//- h2 Phase 2
//- h4 Uncontentious features to enhance UX
//-
//- section
//- h2 Phase 3
//- h4 Extensible "loaders" and deliver an environment that allows user-land experimentation.
//-
//- section
//- h2 Phase 4
//- h4 Addressing the user feedback gathered from the experimentation enabled by Phase 3.
//- section
h1 Maybe import.meta.require?
//- section
//- h1 The Future
//-
//- section
//- h2 ESM is one of many goals
//-
//- section
//- h3 .mjs is in IETF Standard track
//- ul
//- li.hide
//- li Will land in official mime registry
//- li mime types determined by file extension
//-
//- section
//- h1 .js will always be ambiguous
//-
section
h2 What we are working on?
ul
li.hide
li Package Exports
li Conditional Exports
li Self-Reference
li JSON Modules
li WASM Modules
li Package Exports & Conditional Exports

section.small-code
h2 Package Exports
Expand All @@ -692,42 +529,20 @@ html
}

section.small-code
h3 Conditional Exports (Experimental)
h4 https://github.com/nodejs/node/pull/29978/
pre
code.language-javascript.
{
"main": "index-legacy.cjs",
"type": "module",
"exports": {
".": {
"module": "./index.js",
"node": "./index.cjs",
"browser": "./index-browser.js"
}
}
}

section.small-code
h3 Conditional Exports (Experimental)
h4 https://github.com/nodejs/node/pull/29978/
h3 Conditional Exports
pre
code.language-javascript.
{
"main": "index-legacy.cjs",
"type": "module",
"exports": {
".": {
"import": "./index.js",
"require": "./index.cjs",
"browser": "./index-browser.js"
}
"browser": "./index-browser.js"
"import": "./index.mjs",
"default": "./index.js",
}
}

section.small-code
h3 Self Reference (Experimental)
h4 --experimental-resolve-self
h3 Self Reference
pre
code.language-javascript.
{
Expand All @@ -740,13 +555,40 @@ html
// index.mjs
import { wow } from 'cool-module/great-feature'

section
h2 What are the limitations?
ul
li.hide
li Full file path is mandatory
li No support for folder resolution
li No named imports for `import 'cjs'`
li Only experimental support for `import 'json'`
li No support for `require('esm')`

//-
//- section
//- h3 .mjs is in IETF Standard track
//- ul
//- li.hide
//- li Will land in official mime registry
//- li mime types determined by file extension
//-
//- section
//- h1 .js will always be ambiguous
//-
section
h2 What we are working on?
ul
li.hide
li JSON Modules
li WASM Modules

section
h2 Roadmap:
ul
li.hide
li Unflag Conditional Exports in Q1 20'
li Backport all changes to 12 LTS in Q1 20'
li Unflag in 12 in Q2 20'
li Aiming to unflag in 12.x around April

section
h2 We are all hard at work
Expand Down

0 comments on commit 0fde0ab

Please sign in to comment.