@yozora/tokenizer-autolink 2.0.4 → 2.0.5

package/lib/cjs/{index.js → index.cjs}

@@ -178,21 +178,20 @@ const uniqueName = '@yozora/tokenizer-autolink';
 
 class AutolinkTokenizer extends coreTokenizer.BaseInlineTokenizer {
     constructor(props = {}) {
-        var _a, _b;
         super({
-            name: (_a = props.name) !== null && _a !== void 0 ? _a : uniqueName,
-            priority: (_b = props.priority) !== null && _b !== void 0 ? _b : coreTokenizer.TokenizerPriority.ATOMIC,
+            name: props.name ?? uniqueName,
+            priority: props.priority ?? coreTokenizer.TokenizerPriority.ATOMIC,
         });
-        this.match = match;
-        this.parse = parse;
     }
+    match = match;
+    parse = parse;
 }
 
 exports.AutolinkTokenizer = AutolinkTokenizer;
 exports.AutolinkTokenizerName = uniqueName;
 exports.autolinkMatch = match;
 exports.autolinkParse = parse;
-exports["default"] = AutolinkTokenizer;
+exports.default = AutolinkTokenizer;
 exports.eatAbsoluteUri = eatAbsoluteUri;
 exports.eatAutolinkSchema = eatAutolinkSchema;
 exports.eatEmailAddress = eatEmailAddress;

package/lib/esm/{index.js → index.mjs}

@@ -174,14 +174,13 @@ const uniqueName = '@yozora/tokenizer-autolink';
 
 class AutolinkTokenizer extends BaseInlineTokenizer {
     constructor(props = {}) {
-        var _a, _b;
         super({
-            name: (_a = props.name) !== null && _a !== void 0 ? _a : uniqueName,
-            priority: (_b = props.priority) !== null && _b !== void 0 ? _b : TokenizerPriority.ATOMIC,
+            name: props.name ?? uniqueName,
+            priority: props.priority ?? TokenizerPriority.ATOMIC,
         });
-        this.match = match;
-        this.parse = parse;
     }
+    match = match;
+    parse = parse;
 }
 
 export { AutolinkTokenizer, uniqueName as AutolinkTokenizerName, match as autolinkMatch, parse as autolinkParse, AutolinkTokenizer as default, eatAbsoluteUri, eatAutolinkSchema, eatEmailAddress };

package/lib/types/index.d.ts

@@ -35,9 +35,9 @@ declare function eatAbsoluteUri(nodePoints: ReadonlyArray<INodePoint>, startInde
  */
 declare function eatAutolinkSchema(nodePoints: ReadonlyArray<INodePoint>, startIndex: number, endIndex: number): IResultOfRequiredEater;
 
-declare type AutolinkContentType = 'uri' | 'email';
-declare type T = LinkType;
-declare type INode = Link;
+type AutolinkContentType = 'uri' | 'email';
+type T = LinkType;
+type INode = Link;
 declare const uniqueName = "@yozora/tokenizer-autolink";
 interface IToken extends IPartialYastInlineToken<T> {
     /**
@@ -52,8 +52,8 @@ interface IDelimiter extends IYastTokenDelimiter {
      */
     contentType: AutolinkContentType;
 }
-declare type IThis = ITokenizer;
-declare type ITokenizerProps = Partial<IBaseInlineTokenizerProps>;
+type IThis = ITokenizer;
+type ITokenizerProps = Partial<IBaseInlineTokenizerProps>;
 
 /**
  * Autolinks are absolute URIs and email addresses inside '<' and '>'.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@yozora/tokenizer-autolink",
-  "version": "2.0.4",
+  "version": "2.0.5",
   "author": {
     "name": "guanghechen",
     "url": "https://github.com/guanghechen/"
@@ -11,33 +11,37 @@
     "directory": "tokenizers/autolink"
   },
   "homepage": "https://github.com/yozorajs/yozora/tree/release-2.x.x/tokenizers/autolink",
-  "main": "lib/cjs/index.js",
-  "module": "lib/esm/index.js",
-  "types": "lib/types/index.d.ts",
-  "source": "src/index.ts",
+  "type": "module",
+  "exports": {
+    "types": "./lib/types/index.d.ts",
+    "import": "./lib/esm/index.mjs",
+    "require": "./lib/cjs/index.cjs"
+  },
+  "source": "./src/index.ts",
+  "types": "./lib/types/index.d.ts",
+  "main": "./lib/cjs/index.cjs",
+  "module": "./lib/esm/index.mjs",
   "license": "MIT",
   "engines": {
     "node": ">= 16.0.0"
   },
   "files": [
     "lib/",
-    "!lib/**/*.js.map",
-    "!lib/**/*.d.ts.map",
+    "src/",
     "package.json",
     "CHANGELOG.md",
     "LICENSE",
     "README.md"
   ],
   "scripts": {
-    "build": "cross-env NODE_ENV=production rollup -c ../../rollup.config.js",
-    "prebuild": "rimraf lib/",
+    "build": "rimraf lib/ && cross-env NODE_ENV=production rollup -c ../../rollup.config.mjs",
     "prepublishOnly": "cross-env ROLLUP_SHOULD_SOURCEMAP=false yarn build",
-    "test": "cross-env TS_NODE_FILES=true jest --config ../../jest.config.js --rootDir ."
+    "test": "cross-env TS_NODE_FILES=true NODE_OPTIONS=--experimental-vm-modules jest --config ../../jest.config.mjs --rootDir ."
   },
   "dependencies": {
-    "@yozora/ast": "^2.0.4",
-    "@yozora/character": "^2.0.4",
-    "@yozora/core-tokenizer": "^2.0.4"
+    "@yozora/ast": "^2.0.5",
+    "@yozora/character": "^2.0.5",
+    "@yozora/core-tokenizer": "^2.0.5"
   },
-  "gitHead": "c980b95254394dcacba0cbb4bea251350b09397c"
+  "gitHead": "7ba3bab49fe65cf2f57082c0503af73da9356cf0"
 }

package/src/index.ts

@@ -0,0 +1,12 @@
+export * from './util/email'
+export * from './util/uri'
+export { match as autolinkMatch } from './match'
+export { parse as autolinkParse } from './parse'
+export { AutolinkTokenizer, AutolinkTokenizer as default } from './tokenizer'
+export { uniqueName as AutolinkTokenizerName } from './types'
+export type {
+  IThis as IAutolinkHookContext,
+  IToken as IAutolinkToken,
+  ITokenizerProps as IAutolinkTokenizerProps,
+  AutolinkContentType,
+} from './types'

package/src/match.ts

@@ -0,0 +1,79 @@
+import { LinkType } from '@yozora/ast'
+import type { INodePoint } from '@yozora/character'
+import { AsciiCodePoint } from '@yozora/character'
+import type {
+  IMatchInlineHookCreator,
+  IResultOfProcessSingleDelimiter,
+} from '@yozora/core-tokenizer'
+import { genFindDelimiter } from '@yozora/core-tokenizer'
+import type { AutolinkContentType, IContentHelper, IDelimiter, IThis, IToken, T } from './types'
+import { eatEmailAddress } from './util/email'
+import { eatAbsoluteUri } from './util/uri'
+
+const helpers: ReadonlyArray<IContentHelper> = [
+  { contentType: 'uri', eat: eatAbsoluteUri },
+  { contentType: 'email', eat: eatEmailAddress },
+]
+
+/**
+ * Autolinks are absolute URIs and email addresses inside '<' and '>'.
+ * They are parsed as links, with the URL or email address as the link label.
+ *
+ * @see https://github.github.com/gfm/#autolink
+ */
+export const match: IMatchInlineHookCreator<T, IDelimiter, IToken, IThis> = function (api) {
+  return {
+    findDelimiter: () => genFindDelimiter<IDelimiter>(_findDelimiter),
+    processSingleDelimiter,
+  }
+
+  function _findDelimiter(startIndex: number, endIndex: number): IDelimiter | null {
+    const nodePoints: ReadonlyArray<INodePoint> = api.getNodePoints()
+
+    for (let i = startIndex; i < endIndex; ++i) {
+      if (nodePoints[i].codePoint !== AsciiCodePoint.OPEN_ANGLE) continue
+
+      let nextIndex: number = endIndex
+      let contentType: AutolinkContentType | null = null
+      for (const helper of helpers) {
+        const eatResult = helper.eat(nodePoints, i + 1, endIndex)
+        nextIndex = Math.min(nextIndex, eatResult.nextIndex)
+        if (eatResult.valid) {
+          contentType = helper.contentType
+          nextIndex = eatResult.nextIndex
+          break
+        }
+      }
+
+      // Optimization: move forward to the next latest potential position.
+      if (contentType == null) {
+        i = Math.max(i, nextIndex - 1)
+        continue
+      }
+
+      if (nextIndex < endIndex && nodePoints[nextIndex].codePoint === AsciiCodePoint.CLOSE_ANGLE) {
+        return {
+          type: 'full',
+          startIndex: i,
+          endIndex: nextIndex + 1,
+          contentType,
+        }
+      }
+      i = nextIndex - 1
+    }
+    return null
+  }
+
+  function processSingleDelimiter(
+    delimiter: IDelimiter,
+  ): IResultOfProcessSingleDelimiter<T, IToken> {
+    const token: IToken = {
+      nodeType: LinkType,
+      startIndex: delimiter.startIndex,
+      endIndex: delimiter.endIndex,
+      contentType: delimiter.contentType,
+      children: api.resolveFallbackTokens([], delimiter.startIndex + 1, delimiter.endIndex - 1),
+    }
+    return [token]
+  }
+}

package/src/parse.ts

@@ -0,0 +1,31 @@
+import type { Node } from '@yozora/ast'
+import { LinkType } from '@yozora/ast'
+import type { INodePoint } from '@yozora/character'
+import { calcStringFromNodePoints } from '@yozora/character'
+import type { IParseInlineHookCreator } from '@yozora/core-tokenizer'
+import { encodeLinkDestination } from '@yozora/core-tokenizer'
+import type { INode, IThis, IToken, T } from './types'
+
+export const parse: IParseInlineHookCreator<T, IToken, INode, IThis> = function (api) {
+  return {
+    parse: tokens =>
+      tokens.map(token => {
+        const nodePoints: ReadonlyArray<INodePoint> = api.getNodePoints()
+
+        // Backslash-escapes do not work inside autolink.
+        let url = calcStringFromNodePoints(nodePoints, token.startIndex + 1, token.endIndex - 1)
+
+        // Add 'mailto:' prefix to email address type autolink.
+        if (token.contentType === 'email') {
+          url = 'mailto:' + url
+        }
+
+        const encodedUrl = encodeLinkDestination(url)
+        const children: Node[] = api.parseInlineTokens(token.children)
+        const node: INode = api.shouldReservePosition
+          ? { type: LinkType, position: api.calcPosition(token), url: encodedUrl, children }
+          : { type: LinkType, url: encodedUrl, children }
+        return node
+      }),
+  }
+}

package/src/tokenizer.ts

@@ -0,0 +1,36 @@
+import type {
+  IInlineTokenizer,
+  IMatchInlineHookCreator,
+  IParseInlineHookCreator,
+} from '@yozora/core-tokenizer'
+import { BaseInlineTokenizer, TokenizerPriority } from '@yozora/core-tokenizer'
+import { match } from './match'
+import { parse } from './parse'
+import type { IDelimiter, INode, IThis, IToken, ITokenizerProps, T } from './types'
+import { uniqueName } from './types'
+
+/**
+ * Lexical Analyzer for Autolink.
+ * @see https://github.github.com/gfm/#autolink
+ */
+export class AutolinkTokenizer
+  extends BaseInlineTokenizer<T, IDelimiter, IToken, INode, IThis>
+  implements IInlineTokenizer<T, IDelimiter, IToken, INode, IThis>
+{
+  /* istanbul ignore next */
+  constructor(props: ITokenizerProps = {}) {
+    super({
+      name: props.name ?? uniqueName,
+      /**
+       * Autolink has the same priority as inline-code.
+       * @see https://github.github.com/gfm/#example-355
+       * @see https://github.github.com/gfm/#example-356
+       */
+      priority: props.priority ?? TokenizerPriority.ATOMIC,
+    })
+  }
+
+  public override readonly match: IMatchInlineHookCreator<T, IDelimiter, IToken, IThis> = match
+
+  public override readonly parse: IParseInlineHookCreator<T, IToken, INode, IThis> = parse
+}

package/src/types.ts

@@ -0,0 +1,46 @@
+import type { Link, LinkType } from '@yozora/ast'
+import type { INodePoint } from '@yozora/character'
+import type {
+  IBaseInlineTokenizerProps,
+  IPartialYastInlineToken,
+  IResultOfRequiredEater,
+  ITokenizer,
+  IYastTokenDelimiter,
+} from '@yozora/core-tokenizer'
+
+// Content type of autolink
+export type AutolinkContentType = 'uri' | 'email'
+
+export type T = LinkType
+export type INode = Link
+export const uniqueName = '@yozora/tokenizer-autolink'
+
+export interface IToken extends IPartialYastInlineToken<T> {
+  /**
+   * Autolink content type: absolute uri or email.
+   */
+  contentType: AutolinkContentType
+}
+
+export interface IDelimiter extends IYastTokenDelimiter {
+  type: 'full'
+  /**
+   * Autolink content type: absolute uri or email.
+   */
+  contentType: AutolinkContentType
+}
+
+export type IThis = ITokenizer
+
+export type ITokenizerProps = Partial<IBaseInlineTokenizerProps>
+
+export type ContentEater = (
+  nodePoints: ReadonlyArray<INodePoint>,
+  startIndex: number,
+  endIndex: number,
+) => IResultOfRequiredEater
+
+export interface IContentHelper {
+  contentType: AutolinkContentType
+  eat: ContentEater
+}

package/src/util/email.ts

@@ -0,0 +1,99 @@
+import type { INodePoint } from '@yozora/character'
+import {
+  AsciiCodePoint,
+  isAlphanumeric,
+  isAsciiDigitCharacter,
+  isAsciiLetter,
+} from '@yozora/character'
+import type { IResultOfOptionalEater, IResultOfRequiredEater } from '@yozora/core-tokenizer'
+
+/**
+ * An email address, for these purposes, is anything that matches the
+ * non-normative regex from the HTML5 spec:
+ *
+ *  /^[a-zA-Z0-9.!#$%&'*+/=?^_`{|}~-]+@[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}
+ *   [a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*$/
+ *
+ * @see https://github.github.com/gfm/#email-address
+ */
+export function eatEmailAddress(
+  nodePoints: ReadonlyArray<INodePoint>,
+  startIndex: number,
+  endIndex: number,
+): IResultOfRequiredEater {
+  let i = startIndex
+
+  // Match /[a-zA-Z0-9.!#$%&'*+/=?^_`{|}~-]+/
+  for (; i < endIndex; i += 1) {
+    const c = nodePoints[i].codePoint
+    if (isAsciiLetter(c) || isAsciiDigitCharacter(c)) continue
+    if (
+      c !== AsciiCodePoint.DOT &&
+      c !== AsciiCodePoint.EXCLAMATION_MARK &&
+      c !== AsciiCodePoint.NUMBER_SIGN &&
+      c !== AsciiCodePoint.DOLLAR_SIGN &&
+      c !== AsciiCodePoint.PERCENT_SIGN &&
+      c !== AsciiCodePoint.AMPERSAND &&
+      c !== AsciiCodePoint.SINGLE_QUOTE &&
+      c !== AsciiCodePoint.ASTERISK &&
+      c !== AsciiCodePoint.PLUS_SIGN &&
+      c !== AsciiCodePoint.SLASH &&
+      c !== AsciiCodePoint.EQUALS_SIGN &&
+      c !== AsciiCodePoint.QUESTION_MARK &&
+      c !== AsciiCodePoint.CARET &&
+      c !== AsciiCodePoint.UNDERSCORE &&
+      c !== AsciiCodePoint.BACKTICK &&
+      c !== AsciiCodePoint.OPEN_BRACE &&
+      c !== AsciiCodePoint.VERTICAL_SLASH &&
+      c !== AsciiCodePoint.CLOSE_BRACE &&
+      c !== AsciiCodePoint.TILDE &&
+      c !== AsciiCodePoint.MINUS_SIGN
+    )
+      break
+  }
+
+  if (
+    i === startIndex ||
+    i + 1 >= endIndex ||
+    nodePoints[i].codePoint !== AsciiCodePoint.AT_SIGN ||
+    !isAlphanumeric(nodePoints[i + 1].codePoint)
+  )
+    return { valid: false, nextIndex: i + 1 }
+
+  i = eatAddressPart0(nodePoints, i + 2, endIndex)
+
+  // Match /(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*/
+  for (; i + 1 < endIndex; ) {
+    let c = nodePoints[i].codePoint
+    if (c !== AsciiCodePoint.DOT) break
+
+    c = nodePoints[i + 1].codePoint
+    if (!isAsciiLetter(c) && !isAsciiDigitCharacter(c)) break
+    i = eatAddressPart0(nodePoints, i + 2, endIndex)
+  }
+
+  return { valid: true, nextIndex: i }
+}
+
+/**
+ * Match regex /(?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?/
+ *
+ */
+function eatAddressPart0(
+  nodePoints: ReadonlyArray<INodePoint>,
+  startIndex: number,
+  endIndex: number,
+): IResultOfOptionalEater {
+  let i = startIndex,
+    result = -1
+
+  for (let _endIndex = Math.min(endIndex, i + 62); i < _endIndex; ++i) {
+    const c = nodePoints[i].codePoint
+    if (isAsciiLetter(c) || isAsciiDigitCharacter(c)) {
+      result = i
+      continue
+    }
+    if (c !== AsciiCodePoint.MINUS_SIGN) break
+  }
+  return result >= startIndex ? result + 1 : startIndex
+}

package/src/util/uri.ts

@@ -0,0 +1,84 @@
+import type { INodePoint } from '@yozora/character'
+import {
+  AsciiCodePoint,
+  isAlphanumeric,
+  isAsciiCharacter,
+  isAsciiControlCharacter,
+  isAsciiLetter,
+  isWhitespaceCharacter,
+} from '@yozora/character'
+import type { IResultOfRequiredEater } from '@yozora/core-tokenizer'
+
+/**
+ * Try to find to autolink absolute uri strictly start from the give `startIndex`.
+ *
+ * An absolute URI, for these purposes, consists of a scheme followed by a
+ * colon (:) followed by zero or more characters other than ASCII whitespace
+ * and control characters, `<`, and `>`. If the URI includes these characters,
+ * they must be percent-encoded (e.g. %20 for a space).
+ *
+ * @see https://github.github.com/gfm/#absolute-uri
+ */
+export function eatAbsoluteUri(
+  nodePoints: ReadonlyArray<INodePoint>,
+  startIndex: number,
+  endIndex: number,
+): IResultOfRequiredEater {
+  const schema = eatAutolinkSchema(nodePoints, startIndex, endIndex)
+  let { nextIndex } = schema
+
+  if (
+    !schema.valid ||
+    nextIndex >= endIndex ||
+    nodePoints[nextIndex].codePoint !== AsciiCodePoint.COLON
+  )
+    return { valid: false, nextIndex }
+
+  for (nextIndex += 1; nextIndex < endIndex; ++nextIndex) {
+    const c = nodePoints[nextIndex].codePoint
+    if (
+      !isAsciiCharacter(c) ||
+      isWhitespaceCharacter(c) ||
+      isAsciiControlCharacter(c) ||
+      c === AsciiCodePoint.OPEN_ANGLE ||
+      c === AsciiCodePoint.CLOSE_ANGLE
+    )
+      break
+  }
+  return { valid: true, nextIndex }
+}
+
+/**
+ * Try to find to autolink schema strictly start from the give `startIndex`.
+ *
+ * A scheme is any sequence of 2–32 characters beginning with an ASCII letter
+ * and followed by any combination of ASCII letters, digits, or the symbols
+ * plus (`+`), period (`.`), or hyphen (`-`).
+ *
+ * @see https://github.github.com/gfm/#scheme
+ */
+export function eatAutolinkSchema(
+  nodePoints: ReadonlyArray<INodePoint>,
+  startIndex: number,
+  endIndex: number,
+): IResultOfRequiredEater {
+  let i = startIndex
+  const c = nodePoints[i].codePoint
+  if (!isAsciiLetter(c)) return { valid: false, nextIndex: i + 1 }
+
+  for (i += 1; i < endIndex; ++i) {
+    const d = nodePoints[i].codePoint
+    if (
+      isAlphanumeric(d) ||
+      d === AsciiCodePoint.PLUS_SIGN ||
+      d === AsciiCodePoint.DOT ||
+      d === AsciiCodePoint.MINUS_SIGN
+    )
+      continue
+    break
+  }
+
+  const count = i - startIndex
+  if (count < 2 || count > 32) return { valid: false, nextIndex: i + 1 }
+  return { valid: true, nextIndex: i }
+}