wip

lib/md060.mjs

@@ -7,8 +7,19 @@ import { filterByTypesCached } from "./cache.mjs";
 /** @typedef {import("markdownlint").MicromarkToken} MicromarkToken */
 /** @typedef {import("markdownlint").RuleOnErrorInfo} RuleOnErrorInfo */
 
-// See https://unicode.org/reports/tr51/
-const defaultWideCharacterReString = "\\p{RGI_Emoji}";
+const regExpFlags = "gv";
+const anyCharacterRe = new RegExp("[\\s\\S]", regExpFlags);
+// See:
+// https://www.unicode.org/reports/tr11/
+// https://unicode.org/reports/tr24/
+// https://unicode.org/reports/tr51/
+// https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5B%3AEast_Asian_Width%3DFullwidth%3A%5D&abb=on&esc=on&g=&i=
+// Notes:
+// The East_Asian_Width property is not supported (seemingly at all) by JavaScript, so East_Asian_Width=Fullwidth ranges are matched directly:
+// https://github.com/tc39/proposal-regexp-unicode-property-escapes/issues/28
+// As an alternative to matching by Script names, consider matching East_Asian_Width=Wide (Wide is a superset of Fullwidth) directly as well:
+// https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5B%3AEast_Asian_Width%3DWide%3A%5D&abb=on&esc=on&g=&i=
+const defaultWideCharacterReString = "[\\p{RGI_Emoji}\\p{Script=Han}\\p{Script=Hiragana}\\p{Script=Katakana}\\p{Script=Hangul}\\u3000\\uFF01-\\uFF60\\uFFE0-\\uFFE6]";
 
 /**
  * @typedef Column
@@ -21,12 +32,14 @@ const defaultWideCharacterReString = "\\p{RGI_Emoji}";
  *
  * @param {string} line Line of text.
  * @param {number} column Actual column (1-based).
- * @param {RegExp} wideRe Wide character RegExp.
+ * @param {RegExp} wideCharacterRe Wide character RegExp.
  * @returns {number} Effective column (1-based).
  */
-function effectiveColumn(line, column, wideRe) {
-  const wideCharacterCount = (line.slice(0, column - 1).match(wideRe) || []).length;
-  return column + wideCharacterCount;
+function effectiveColumn(line, column, wideCharacterRe) {
+  const span = line.slice(0, column - 1);
+  const totalCharacterCount = (span.match(anyCharacterRe) || []).length;
+  const wideCharacterCount = (span.match(wideCharacterRe) || []).length;
+  return totalCharacterCount + wideCharacterCount;
 }
 
 /**
@@ -34,13 +47,13 @@ function effectiveColumn(line, column, wideRe) {
  *
  * @param {readonly string[]} lines File/string lines.
  * @param {MicromarkToken} row Micromark row token.
- * @param {RegExp} wideRe Wide character RegExp.
+ * @param {RegExp} wideCharacterRe Wide character RegExp.
  * @returns {Column[]} Divider columns.
  */
-function getTableDividerColumns(lines, row, wideRe) {
+function getTableDividerColumns(lines, row, wideCharacterRe) {
   return filterByTypes(
     row.children,
-    [ "tableCellDivider" ]).map((divider) => ({ "actual": divider.startColumn, "effective": effectiveColumn(lines[row.startLine - 1], divider.startColumn, wideRe) })
+    [ "tableCellDivider" ]).map((divider) => ({ "actual": divider.startColumn, "effective": effectiveColumn(lines[row.startLine - 1], divider.startColumn, wideCharacterRe) })
   );
 }
 
@@ -71,9 +84,7 @@ export default {
     const styleAlignedAllowed = (style === "any") || (style === "aligned");
     const styleCompactAllowed = (style === "any") || (style === "compact");
     const styleTightAllowed = (style === "any") || (style === "tight");
-    const wideCharacter = params.config.wide_character;
-    const wideCharacterReString = (wideCharacter === undefined) ? defaultWideCharacterReString : wideCharacter;
-    const wideCharacterRe = new RegExp(wideCharacterReString, "gv");
+    const wideCharacterRe = new RegExp(params.config.wide_character || defaultWideCharacterReString, regExpFlags);
 
     // Scan all tables/rows
     const tables = filterByTypesCached([ "table" ]);