micromark-extension-cjk-friendly-util 1.1.0 → 2.0.0-rc.2

package/dist/categoryUtil.d.ts

@@ -32,12 +32,19 @@ declare function isCjk(category: Category): boolean;
  */
 declare function isIvs(category: Category): boolean;
 /**
- * `true` if the code point represents a [Standard Variation Selector that can follow CJK](https://github.com/tats-u/markdown-cjk-friendly/blob/main/specification.md#svs-that-can-follow-cjk).
+ * `true` if {@link isCjk} or {@link isIvs}.
+ *
+ * @param category the return value of {@link classifyCharacter}.
+ * @returns `true` if the code point represents a CJK or IVS
+ */
+declare function isCjkOrIvs(category: Category): boolean;
+/**
+ * `true` if the code point represents a [Non-emoji General-use Variation Selector](https://github.com/tats-u/markdown-cjk-friendly/blob/main/specification.md#non-emoji-general-use-variation-selector).
  *
  * @param category the return value of `classifyCharacter`.
- * @returns `true` if the code point represents an Standard Variation Selector that can follow CJK
+ * @returns `true` if the code point represents an Non-emoji General-use Variation Selector
  */
-declare function isSvsFollowingCjk(category: Category): boolean;
+declare function isNonEmojiGeneralUseVS(category: Category): boolean;
 /**
  * `true` if the code point represents an [Unicode whitespace character](https://spec.commonmark.org/0.31.2/#unicode-whitespace-character) or an [Unicode punctuation character](https://spec.commonmark.org/0.31.2/#unicode-punctuation-character).
  *
@@ -46,4 +53,4 @@ declare function isSvsFollowingCjk(category: Category): boolean;
  */
 declare function isSpaceOrPunctuation(category: Category): boolean;
 
-export { isCjk, isIvs, isNonCjkPunctuation, isSpaceOrPunctuation, isSvsFollowingCjk, isUnicodeWhitespace };
+export { isCjk, isCjkOrIvs, isIvs, isNonCjkPunctuation, isNonEmojiGeneralUseVS, isSpaceOrPunctuation, isUnicodeWhitespace };

package/dist/categoryUtil.js

@@ -11,8 +11,9 @@ var constantsEx;
   constantsEx2.cjkPunctuation = 4098;
   constantsEx2.ivs = 8192;
   constantsEx2.cjkOrIvs = 12288;
-  constantsEx2.svsFollowingCjk = 16384;
-  constantsEx2.variationSelector = 28672;
+  constantsEx2.nonEmojiGeneralUseVS = 16384;
+  constantsEx2.variationSelector = 24576;
+  constantsEx2.ivsToCjkRightShift = 1;
 })(constantsEx || (constantsEx = {}));
 
 // src/categoryUtil.ts
@@ -28,17 +29,21 @@ function isCjk(category) {
 function isIvs(category) {
   return category === constantsEx.ivs;
 }
-function isSvsFollowingCjk(category) {
-  return category === constantsEx.svsFollowingCjk;
+function isCjkOrIvs(category) {
+  return Boolean(category & constantsEx.cjkOrIvs);
+}
+function isNonEmojiGeneralUseVS(category) {
+  return category === constantsEx.nonEmojiGeneralUseVS;
 }
 function isSpaceOrPunctuation(category) {
   return Boolean(category & constantsEx.spaceOrPunctuation);
 }
 export {
   isCjk,
+  isCjkOrIvs,
   isIvs,
   isNonCjkPunctuation,
+  isNonEmojiGeneralUseVS,
   isSpaceOrPunctuation,
-  isSvsFollowingCjk,
   isUnicodeWhitespace
 };

package/dist/characterWithNonBmp.d.ts

@@ -7,12 +7,11 @@ import { Code } from 'micromark-util-types';
  * @returns `true` if `uc` is CJK, `null` if IVS, or `false` if neither
  */
 declare function cjkOrIvs(uc: Code): boolean | null;
+declare function isCjkAmbiguousPunctuation(main: Code, vs: Code): boolean;
 /**
- * Check whether the character code represents Standard Variation Sequence that can follow an ideographic character.
- *
- * U+FE0E is used for some CJK symbols (e.g. U+3299) that can also be
+ * Check whether the character code represents Non-emoji General-use Variation Selector (U+FE00-U+FE0E).
  */
-declare const svsFollowingCjk: (code: Code) => boolean;
+declare function nonEmojiGeneralUseVS(code: Code): boolean;
 /**
  * Check whether the character code represents Unicode punctuation.
  *
@@ -55,4 +54,4 @@ declare const unicodePunctuation: (code: Code) => boolean;
  */
 declare const unicodeWhitespace: (code: Code) => boolean;
 
-export { cjkOrIvs, svsFollowingCjk, unicodePunctuation, unicodeWhitespace };
+export { cjkOrIvs, isCjkAmbiguousPunctuation, nonEmojiGeneralUseVS, unicodePunctuation, unicodeWhitespace };

package/dist/characterWithNonBmp.js

@@ -18,7 +18,7 @@ var isEmoji = function(uc) {
   fn: null
 });
 function cjkOrIvs(uc) {
-  if (!uc || uc < 0) {
+  if (!uc || uc < 4352) {
     return false;
   }
   const eaw = eastAsianWidthType(uc);
@@ -37,7 +37,13 @@ function cjkOrIvs(uc) {
       return /^\p{sc=Hangul}/u.test(String.fromCodePoint(uc));
   }
 }
-var svsFollowingCjk = regexCheck(/[\uFE00-\uFE02\uFE0E]/u);
+function isCjkAmbiguousPunctuation(main, vs) {
+  if (vs !== 65025 || !main || main < 8216) return false;
+  return main === 8216 || main === 8217 || main === 8220 || main === 8221;
+}
+function nonEmojiGeneralUseVS(code) {
+  return code !== null && code >= 65024 && code <= 65038;
+}
 var unicodePunctuation = regexCheck(/\p{P}|\p{S}/u);
 var unicodeWhitespace = regexCheck(/\s/);
 function regexCheck(regex) {
@@ -48,7 +54,8 @@ function regexCheck(regex) {
 }
 export {
   cjkOrIvs,
-  svsFollowingCjk,
+  isCjkAmbiguousPunctuation,
+  nonEmojiGeneralUseVS,
   unicodePunctuation,
   unicodeWhitespace
 };

package/dist/classifyCharacter.d.ts

@@ -7,8 +7,9 @@ declare namespace constantsEx {
     const cjkPunctuation: 4098;
     const ivs: 8192;
     const cjkOrIvs: 12288;
-    const svsFollowingCjk: 16384;
-    const variationSelector: 28672;
+    const nonEmojiGeneralUseVS: 16384;
+    const variationSelector: 24576;
+    const ivsToCjkRightShift: 1;
 }
 /**
  * Classify whether a code represents whitespace, punctuation, or something
@@ -24,6 +25,18 @@ declare namespace constantsEx {
  * @returns
  *   Group.
  */
-declare function classifyCharacter(code: Code): typeof constants.characterGroupWhitespace | typeof constants.characterGroupPunctuation | typeof constantsEx.cjk | typeof constantsEx.cjkPunctuation | typeof constantsEx.ivs | typeof constantsEx.svsFollowingCjk | 0;
+declare function classifyCharacter(code: Code): typeof constants.characterGroupWhitespace | typeof constants.characterGroupPunctuation | typeof constantsEx.cjk | typeof constantsEx.cjkPunctuation | typeof constantsEx.ivs | typeof constantsEx.nonEmojiGeneralUseVS | 0;
+/**}
+ * Classify whether a code represents whitespace, punctuation, or something else.
+ *
+ * Recognizes general-use variation selectors. Use this instead of {@linkcode classifyCharacter} for previous character.
+ *
+ * @param before result of {@linkcode classifyCharacter} of the preceding character.
+ * @param get2Previous a function that returns the code point of the character before the preceding character. Use lambda or {@linkcode Function.prototype.bind}.
+ * @param previous code point of the preceding character
+ * @returns
+ *   Group of the main code point of the preceding character. Use `isCjkOrIvs` to check whether it is CJK
+ */
+declare function classifyPrecedingCharacter(before: ReturnType<typeof classifyCharacter>, get2Previous: () => Code, previous: Code): ReturnType<typeof classifyCharacter>;
 
-export { classifyCharacter, constantsEx };
+export { classifyCharacter, classifyPrecedingCharacter, constantsEx };

package/dist/classifyCharacter.js

@@ -1,6 +1,15 @@
 // src/classifyCharacter.ts
 import { markdownLineEndingOrSpace } from "micromark-util-character";
-import { constants, codes } from "micromark-util-symbol";
+import { constants as constants2, codes } from "micromark-util-symbol";
+
+// src/categoryUtil.ts
+import { constants } from "micromark-util-symbol";
+function isUnicodeWhitespace(category) {
+  return Boolean(category & constants.characterGroupWhitespace);
+}
+function isNonEmojiGeneralUseVS(category) {
+  return category === constantsEx.nonEmojiGeneralUseVS;
+}
 
 // src/characterWithNonBmp.ts
 import { eastAsianWidthType } from "get-east-asian-width";
@@ -22,7 +31,7 @@ var isEmoji = function(uc) {
   fn: null
 });
 function cjkOrIvs(uc) {
-  if (!uc || uc < 0) {
+  if (!uc || uc < 4352) {
     return false;
   }
   const eaw = eastAsianWidthType(uc);
@@ -41,7 +50,13 @@ function cjkOrIvs(uc) {
       return /^\p{sc=Hangul}/u.test(String.fromCodePoint(uc));
   }
 }
-var svsFollowingCjk = regexCheck(/[\uFE00-\uFE02\uFE0E]/u);
+function isCjkAmbiguousPunctuation(main, vs) {
+  if (vs !== 65025 || !main || main < 8216) return false;
+  return main === 8216 || main === 8217 || main === 8220 || main === 8221;
+}
+function nonEmojiGeneralUseVS(code) {
+  return code !== null && code >= 65024 && code <= 65038;
+}
 var unicodePunctuation = regexCheck(/\p{P}|\p{S}/u);
 var unicodeWhitespace = regexCheck(/\s/);
 function regexCheck(regex) {
@@ -59,17 +74,18 @@ var constantsEx;
   constantsEx2.cjkPunctuation = 4098;
   constantsEx2.ivs = 8192;
   constantsEx2.cjkOrIvs = 12288;
-  constantsEx2.svsFollowingCjk = 16384;
-  constantsEx2.variationSelector = 28672;
+  constantsEx2.nonEmojiGeneralUseVS = 16384;
+  constantsEx2.variationSelector = 24576;
+  constantsEx2.ivsToCjkRightShift = 1;
 })(constantsEx || (constantsEx = {}));
 function classifyCharacter(code) {
   if (code === codes.eof || markdownLineEndingOrSpace(code) || unicodeWhitespace(code)) {
-    return constants.characterGroupWhitespace;
+    return constants2.characterGroupWhitespace;
   }
   let value = 0;
   if (code >= 4352) {
-    if (svsFollowingCjk(code)) {
-      return constantsEx.svsFollowingCjk;
+    if (nonEmojiGeneralUseVS(code)) {
+      return constantsEx.nonEmojiGeneralUseVS;
     }
     switch (cjkOrIvs(code)) {
       case null:
@@ -80,11 +96,23 @@ function classifyCharacter(code) {
     }
   }
   if (unicodePunctuation(code)) {
-    value |= constants.characterGroupPunctuation;
+    value |= constants2.characterGroupPunctuation;
   }
   return value;
 }
+function classifyPrecedingCharacter(before, get2Previous, previous) {
+  if (!isNonEmojiGeneralUseVS(before)) {
+    return before;
+  }
+  const twoPrevious = get2Previous();
+  const twoBefore = classifyCharacter(twoPrevious);
+  return !twoPrevious || isUnicodeWhitespace(twoBefore) ? before : isCjkAmbiguousPunctuation(twoPrevious, previous) ? constantsEx.cjkPunctuation : stripIvs(twoBefore);
+}
+function stripIvs(twoBefore) {
+  return twoBefore & ~constantsEx.ivs;
+}
 export {
   classifyCharacter,
+  classifyPrecedingCharacter,
   constantsEx
 };

package/dist/index.d.ts

@@ -1,5 +1,5 @@
-export { isCjk, isIvs, isNonCjkPunctuation, isSpaceOrPunctuation, isSvsFollowingCjk, isUnicodeWhitespace } from './categoryUtil.js';
-export { classifyCharacter, constantsEx } from './classifyCharacter.js';
-export { isCodeHighSurrogate, isCodeLowSurrogate, tryGetCodeTwoBefore, tryGetGenuineNextCode, tryGetGenuinePreviousCode } from './codeUtil.js';
+export { isCjk, isCjkOrIvs, isIvs, isNonCjkPunctuation, isNonEmojiGeneralUseVS, isSpaceOrPunctuation, isUnicodeWhitespace } from './categoryUtil.js';
+export { classifyCharacter, classifyPrecedingCharacter, constantsEx } from './classifyCharacter.js';
+export { TwoPreviousCode, isCodeHighSurrogate, isCodeLowSurrogate, tryGetCodeTwoBefore, tryGetGenuineNextCode, tryGetGenuinePreviousCode } from './codeUtil.js';
 import 'micromark-util-symbol';
 import 'micromark-util-types';

package/dist/index.js

@@ -1,3 +1,7 @@
+var __defProp = Object.defineProperty;
+var __defNormalProp = (obj, key, value) => key in obj ? __defProp(obj, key, { enumerable: true, configurable: true, writable: true, value }) : obj[key] = value;
+var __publicField = (obj, key, value) => __defNormalProp(obj, typeof key !== "symbol" ? key + "" : key, value);
+
 // src/categoryUtil.ts
 import { constants as constants2 } from "micromark-util-symbol";
 
@@ -25,7 +29,7 @@ var isEmoji = function(uc) {
   fn: null
 });
 function cjkOrIvs(uc) {
-  if (!uc || uc < 0) {
+  if (!uc || uc < 4352) {
     return false;
   }
   const eaw = eastAsianWidthType(uc);
@@ -44,7 +48,13 @@ function cjkOrIvs(uc) {
       return /^\p{sc=Hangul}/u.test(String.fromCodePoint(uc));
   }
 }
-var svsFollowingCjk = regexCheck(/[\uFE00-\uFE02\uFE0E]/u);
+function isCjkAmbiguousPunctuation(main, vs) {
+  if (vs !== 65025 || !main || main < 8216) return false;
+  return main === 8216 || main === 8217 || main === 8220 || main === 8221;
+}
+function nonEmojiGeneralUseVS(code) {
+  return code !== null && code >= 65024 && code <= 65038;
+}
 var unicodePunctuation = regexCheck(/\p{P}|\p{S}/u);
 var unicodeWhitespace = regexCheck(/\s/);
 function regexCheck(regex) {
@@ -62,8 +72,9 @@ var constantsEx;
   constantsEx2.cjkPunctuation = 4098;
   constantsEx2.ivs = 8192;
   constantsEx2.cjkOrIvs = 12288;
-  constantsEx2.svsFollowingCjk = 16384;
-  constantsEx2.variationSelector = 28672;
+  constantsEx2.nonEmojiGeneralUseVS = 16384;
+  constantsEx2.variationSelector = 24576;
+  constantsEx2.ivsToCjkRightShift = 1;
 })(constantsEx || (constantsEx = {}));
 function classifyCharacter(code) {
   if (code === codes.eof || markdownLineEndingOrSpace(code) || unicodeWhitespace(code)) {
@@ -71,8 +82,8 @@ function classifyCharacter(code) {
   }
   let value = 0;
   if (code >= 4352) {
-    if (svsFollowingCjk(code)) {
-      return constantsEx.svsFollowingCjk;
+    if (nonEmojiGeneralUseVS(code)) {
+      return constantsEx.nonEmojiGeneralUseVS;
     }
     switch (cjkOrIvs(code)) {
       case null:
@@ -87,6 +98,17 @@ function classifyCharacter(code) {
   }
   return value;
 }
+function classifyPrecedingCharacter(before, get2Previous, previous) {
+  if (!isNonEmojiGeneralUseVS(before)) {
+    return before;
+  }
+  const twoPrevious = get2Previous();
+  const twoBefore = classifyCharacter(twoPrevious);
+  return !twoPrevious || isUnicodeWhitespace(twoBefore) ? before : isCjkAmbiguousPunctuation(twoPrevious, previous) ? constantsEx.cjkPunctuation : stripIvs(twoBefore);
+}
+function stripIvs(twoBefore) {
+  return twoBefore & ~constantsEx.ivs;
+}
 
 // src/categoryUtil.ts
 function isUnicodeWhitespace(category) {
@@ -101,8 +123,11 @@ function isCjk(category) {
 function isIvs(category) {
   return category === constantsEx.ivs;
 }
-function isSvsFollowingCjk(category) {
-  return category === constantsEx.svsFollowingCjk;
+function isCjkOrIvs(category) {
+  return Boolean(category & constantsEx.cjkOrIvs);
+}
+function isNonEmojiGeneralUseVS(category) {
+  return category === constantsEx.nonEmojiGeneralUseVS;
 }
 function isSpaceOrPunctuation(category) {
   return Boolean(category & constantsEx.spaceOrPunctuation);
@@ -159,6 +184,40 @@ function tryGetCodeTwoBefore(previousCode, nowPoint, sliceSerialize) {
   }
   return twoPreviousLast;
 }
+var TwoPreviousCode = class {
+  /**
+   * @see {@link tryGetCodeTwoBefore}
+   *
+   * @param previousCode a previous code point. Should be greater than 65,535 if it represents a [Supplementary Character](https://www.unicode.org/glossary/#supplementary_character).
+   * @param nowPoint `this.now()` (`this` = `TokenizeContext`)
+   * @param sliceSerialize `this.sliceSerialize` (`this` = `TokenizeContext`)
+   */
+  constructor(previousCode, nowPoint, sliceSerialize) {
+    this.previousCode = previousCode;
+    this.nowPoint = nowPoint;
+    this.sliceSerialize = sliceSerialize;
+    __publicField(this, "cachedValue");
+  }
+  /**
+   * Returns the return value of {@link tryGetCodeTwoBefore}.
+   *
+   * If the value has not been computed yet, it will be computed and cached.
+   *
+   * @see {@link tryGetCodeTwoBefore}
+   *
+   * @returns a value greater than 65,535 if the code point two positions before represents a [Supplementary Character](https://www.unicode.org/glossary/#supplementary_character), a value less than 65,536 for a [BMP Character](https://www.unicode.org/glossary/#bmp_character), or `null` if not found
+   */
+  value() {
+    if (this.cachedValue === void 0) {
+      this.cachedValue = tryGetCodeTwoBefore(
+        this.previousCode,
+        this.nowPoint,
+        this.sliceSerialize
+      );
+    }
+    return this.cachedValue;
+  }
+};
 function tryGetGenuineNextCode(code, nowPoint, sliceSerialize) {
   const nextCandidate = sliceSerialize({
     start: nowPoint,
@@ -167,15 +226,18 @@ function tryGetGenuineNextCode(code, nowPoint, sliceSerialize) {
   return nextCandidate && nextCandidate >= 65536 ? nextCandidate : code;
 }
 export {
+  TwoPreviousCode,
   classifyCharacter,
+  classifyPrecedingCharacter,
   constantsEx,
   isCjk,
+  isCjkOrIvs,
   isCodeHighSurrogate,
   isCodeLowSurrogate,
   isIvs,
   isNonCjkPunctuation,
+  isNonEmojiGeneralUseVS,
   isSpaceOrPunctuation,
-  isSvsFollowingCjk,
   isUnicodeWhitespace,
   tryGetCodeTwoBefore,
   tryGetGenuineNextCode,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "micromark-extension-cjk-friendly-util",
-  "version": "1.1.0",
+  "version": "2.0.0-rc.2",
   "type": "module",
   "exports": {
     ".": {