Use ESM & update micromark

.gitignore

@@ -1,9 +1,7 @@
-.nyc_output/
 coverage/
-dist/
-micromark/
 node_modules/
+/lib/
+/index.js
 *.log
 .DS_Store
 yarn.lock
-mdast-util-from-markdown.min.js

.prettierignore

@@ -1,4 +1,2 @@
 coverage/
-micromark/
-*.json
 *.md

dev/index.js

@@ -0,0 +1 @@
+export {fromMarkdown} from './lib/index.js'

lib/index.js → dev/lib/index.js

@@ -1,32 +1,26 @@
-'use strict'
-
-module.exports = fromMarkdown
-
-// These three are compiled away in the `dist/`
-var codes = require('micromark/dist/character/codes')
-var constants = require('micromark/dist/constant/constants')
-var types = require('micromark/dist/constant/types')
-
-var toString = require('mdast-util-to-string')
-var assign = require('micromark/dist/constant/assign')
-var own = require('micromark/dist/constant/has-own-property')
-var normalizeIdentifier = require('micromark/dist/util/normalize-identifier')
-var safeFromInt = require('micromark/dist/util/safe-from-int')
-var parser = require('micromark/dist/parse')
-var preprocessor = require('micromark/dist/preprocess')
-var postprocess = require('micromark/dist/postprocess')
-var decode = require('parse-entities/decode-entity')
-var stringifyPosition = require('unist-util-stringify-position')
-
-function fromMarkdown(value, encoding, options) {
+import {toString} from 'mdast-util-to-string'
+import {parse} from 'micromark/lib/parse.js'
+import {preprocess} from 'micromark/lib/preprocess.js'
+import {postprocess} from 'micromark/lib/postprocess.js'
+import {normalizeIdentifier} from 'micromark-util-normalize-identifier'
+import {codes} from 'micromark-util-symbol/codes.js'
+import {values} from 'micromark-util-symbol/values.js'
+import {constants} from 'micromark-util-symbol/constants.js'
+import {types} from 'micromark-util-symbol/types.js'
+import {decodeEntity} from 'parse-entities/decode-entity.js'
+import {stringifyPosition} from 'unist-util-stringify-position'
+
+const own = {}.hasOwnProperty
+
+export function fromMarkdown(value, encoding, options) {
   if (typeof encoding !== 'string') {
     options = encoding
     encoding = undefined
   }
 
   return compiler(options)(
     postprocess(
-      parser(options).document().write(preprocessor()(value, encoding, true))
+      parse(options).document().write(preprocess()(value, encoding, true))
     )
   )
 }
@@ -155,15 +149,15 @@ function compiler(options) {
     var listStart
 
     var context = {
-      stack: stack,
-      tokenStack: tokenStack,
-      config: config,
-      enter: enter,
-      exit: exit,
-      buffer: buffer,
-      resume: resume,
-      setData: setData,
-      getData: getData
+      stack,
+      tokenStack,
+      config,
+      enter,
+      exit,
+      buffer,
+      resume,
+      setData,
+      getData
     }
 
     while (++index < events.length) {
@@ -189,7 +183,10 @@ function compiler(options) {
 
       if (own.call(handler, events[index][1].type)) {
         handler[events[index][1].type].call(
-          assign({sliceSerialize: events[index][2].sliceSerialize}, context),
+          Object.assign(
+            {sliceSerialize: events[index][2].sliceSerialize},
+            context
+          ),
           events[index][1]
         )
       }
@@ -472,16 +469,18 @@ function compiler(options) {
 
   function onexitcodefenced() {
     var data = this.resume()
+
     this.stack[this.stack.length - 1].value = data.replace(
       /^(\r?\n|\r)|(\r?\n|\r)$/g,
       ''
     )
+
     setData('flowCodeInside')
   }
 
   function onexitcodeindented() {
     var data = this.resume()
-    this.stack[this.stack.length - 1].value = data
+    this.stack[this.stack.length - 1].value = data.replace(/(\r?\n|\r)$/g, '')
   }
 
   function onexitdefinitionlabelstring(token) {
@@ -679,15 +678,15 @@ function compiler(options) {
     var tail
 
     if (type) {
-      value = safeFromInt(
+      value = parseNumericCharacterReference(
         data,
         type === types.characterReferenceMarkerNumeric
           ? constants.numericBaseDecimal
           : constants.numericBaseHexadecimal
       )
       setData('characterReferenceType')
     } else {
-      value = decode(data)
+      value = decodeEntity(data)
     }
 
     tail = this.stack.pop()
@@ -816,3 +815,39 @@ function extension(config, extension) {
     }
   }
 }
+
+// To do: externalize this from `micromark/lib/compile`
+/**
+ * Turn the number (in string form as either hexa- or plain decimal) coming from
+ * a numeric character reference into a character.
+ *
+ * @param {string} value
+ * @param {number} base
+ * @returns {string}
+ */
+function parseNumericCharacterReference(value, base) {
+  const code = Number.parseInt(value, base)
+
+  if (
+    // C0 except for HT, LF, FF, CR, space
+    code < codes.ht ||
+    code === codes.vt ||
+    (code > codes.cr && code < codes.space) ||
+    // Control character (DEL) of the basic block and C1 controls.
+    (code > codes.tilde && code < 160) ||
+    // Lone high surrogates and low surrogates.
+    /* c8 ignore next */
+    (code > 55295 && code < 57344) ||
+    // Noncharacters.
+    /* c8 ignore next */
+    (code > 64975 && code < 65008) ||
+    (code & 65535) === 65535 ||
+    (code & 65535) === 65534 ||
+    // Out of range
+    code > 1114111
+  ) {
+    return values.replacementCharacter
+  }
+
+  return String.fromCharCode(code)
+}

package.json

@@ -26,56 +26,49 @@
   "contributors": [
     "Titus Wormer <tituswormer@gmail.com> (https://wooorm.com)"
   ],
+  "sideEffects": false,
+  "type": "module",
+  "main": "index.js",
   "files": [
-    "dist/",
+    "dev/",
     "lib/",
-    "index.js",
-    "types/index.d.ts"
+    "index.js"
   ],
-  "types": "types",
+  "exports": {
+    "development": "./dev/index.js",
+    "default": "./index.js"
+  },
   "dependencies": {
     "@types/mdast": "^3.0.0",
-    "mdast-util-to-string": "^2.0.0",
-    "micromark": "~2.11.0",
-    "parse-entities": "^2.0.0",
-    "unist-util-stringify-position": "^2.0.0"
+    "mdast-util-to-string": "^3.0.0",
+    "micromark": "^3.0.0-alpha.3",
+    "micromark-util-normalize-identifier": "^1.0.0-alpha.3",
+    "micromark-util-symbol": "^1.0.0-alpha.3",
+    "parse-entities": "^3.0.0",
+    "unist-util-stringify-position": "^3.0.0"
   },
   "devDependencies": {
-    "@babel/cli": "^7.0.0",
-    "@babel/core": "^7.0.0",
-    "babel-plugin-inline-constants": "^1.0.0",
-    "browserify": "^17.0.0",
-    "commonmark.json": "^0.29.0",
-    "dtslint": "^4.0.0",
+    "c8": "^7.0.0",
+    "commonmark.json": "^0.30.0",
     "gzip-size-cli": "^5.0.0",
-    "hast-util-to-html": "^7.0.0",
-    "mdast-util-to-hast": "^10.0.0",
-    "nyc": "^15.0.0",
+    "hast-util-to-html": "^8.0.0",
+    "mdast-util-to-hast": "^11.0.0",
+    "micromark-build": "^1.0.0-alpha.1",
     "prettier": "^2.0.0",
     "rehype-parse": "^7.0.0",
     "rehype-stringify": "^8.0.0",
     "remark-cli": "^9.0.0",
     "remark-preset-wooorm": "^8.0.0",
     "tape": "^5.0.0",
-    "tinyify": "^3.0.0",
     "unified": "^9.0.0",
-    "xo": "^0.38.0"
+    "xo": "^0.39.0"
   },
   "scripts": {
+    "build": "micromark-build",
     "format": "remark . -qfo && prettier . -w --loglevel warn && xo --fix",
-    "generate-dist": "babel lib/ --out-dir dist/ --quiet --retain-lines; prettier dist/index.js --loglevel error --write",
-    "generate-size": "browserify . -p tinyify -s mdast-util-from-markdown -o mdast-util-from-markdown.min.js; gzip-size mdast-util-from-markdown.min.js --raw",
-    "generate": "npm run generate-dist && npm run generate-size",
-    "test-api": "node test",
-    "test-coverage": "nyc --reporter lcov tape test/index.js",
-    "test-types": "dtslint types",
-    "test": "npm run format && npm run generate && npm run test-coverage && npm run test-types"
-  },
-  "nyc": {
-    "check-coverage": true,
-    "lines": 100,
-    "functions": 100,
-    "branches": 100
+    "test-api": "node --conditions development test/index.js",
+    "test-coverage": "c8 --check-coverage --branches 100 --functions 100 --lines 100 --statements 100 --reporter lcov node --conditions development test/index.js",
+    "test": "npm run build && npm run format && npm run test-coverage"
   },
   "prettier": {
     "tabWidth": 2,
@@ -87,10 +80,12 @@
   },
   "xo": {
     "prettier": true,
-    "esnext": false,
     "rules": {
+      "no-var": "off",
+      "prefer-arrow-callback": "off",
       "complexity": "off",
       "guard-for-in": "off",
+      "unicorn/prefer-switch": "off",
       "unicorn/explicit-length-check": "off",
       "unicorn/no-array-callback-reference": "off",
       "unicorn/prefer-includes": "off",

readme.md

@@ -12,6 +12,9 @@
 
 ## Install
 
+This package is [ESM only](https://gist.github.com/sindresorhus/a39789f98801d908bbc7ff3ecc99d99c):
+Node 12+ is needed to use it and it must be `import`ed instead of `require`d.
+
 [npm][]:
 
 ```sh
@@ -29,12 +32,12 @@ Say we have the following markdown file, `example.md`:
 And our script, `example.js`, looks as follows:
 
 ```js
-var fs = require('fs')
-var fromMarkdown = require('mdast-util-from-markdown')
+import fs from 'node:fs'
+import {fromMarkdown} from 'mdast-util-from-markdown'
 
-var doc = fs.readFileSync('example.md')
+const doc = fs.readFileSync('example.md')
 
-var tree = fromMarkdown(doc)
+const tree = fromMarkdown(doc)
 
 console.log(tree)
 ```
@@ -63,6 +66,9 @@ Now, running `node example` yields (positional info removed for brevity):
 
 ## API
 
+This package exports the following identifier: `fromMarkdown`.
+There is no default export.
+
 ### `fromMarkdown(doc[, encoding][, options])`
 
 Parse markdown to a **[mdast][]** tree.

test/fixtures/blockquote.json

@@ -233,9 +233,9 @@
               "offset": 111
             },
             "end": {
-              "line": 13,
-              "column": 1,
-              "offset": 144
+              "line": 12,
+              "column": 29,
+              "offset": 143
             }
           }
         }
@@ -247,9 +247,9 @@
           "offset": 95
         },
         "end": {
-          "line": 13,
-          "column": 1,
-          "offset": 144
+          "line": 12,
+          "column": 29,
+          "offset": 143
         }
       }
     },