@peaceroad/markdown-it-cjk-breaks-mod 0.1.2 → 0.1.4

package/index.js

@@ -6,6 +6,13 @@ const ASCII_PRINTABLE_MAX = 0x7E;
 const IDEOGRAPHIC_SPACE = '\u3000';
 const DEFAULT_PUNCTUATION_TARGETS = ['！', '？', '⁉', '！？', '？！', '!?', '?!', '.', ':'];
 const DEFAULT_PUNCTUATION_CONFIG = create_punctuation_config(DEFAULT_PUNCTUATION_TARGETS);
+/* eslint-disable max-len */
+// require('unicode-10.0.0/Script/Hangul/regex')
+const HANGUL_RE = /[\u1100-\u11FF\u302E\u302F\u3131-\u318E\u3200-\u321E\u3260-\u327E\uA960-\uA97C\uAC00-\uD7A3\uD7B0-\uD7C6\uD7CB-\uD7FB\uFFA0-\uFFBE\uFFC2-\uFFC7\uFFCA-\uFFCF\uFFD2-\uFFD7\uFFDA-\uFFDC]/;
+/* eslint-enable max-len */
+const WHITESPACE_RE = /\s/;
+const WHITESPACE_LEAD_RE = /^\s/;
+const WHITESPACE_TRAIL_RE = /\s$/;
 
 
 function is_surrogate(c1, c2) {
@@ -14,10 +21,7 @@ function is_surrogate(c1, c2) {
 
 
 function is_hangul(c) {
-  // require('unicode-10.0.0/Script/Hangul/regex')
-  /* eslint-disable max-len */
-  return /[\u1100-\u11FF\u302E\u302F\u3131-\u318E\u3200-\u321E\u3260-\u327E\uA960-\uA97C\uAC00-\uD7A3\uD7B0-\uD7C6\uD7CB-\uD7FB\uFFA0-\uFFBE\uFFC2-\uFFC7\uFFCA-\uFFCF\uFFD2-\uFFD7\uFFDA-\uFFDC]/.test(c);
-  /* eslint-enable max-len */
+  return HANGUL_RE.test(c);
 }
 
 
@@ -51,10 +55,15 @@ function resolve_punctuation_targets(opts) {
   if (!opts) return DEFAULT_PUNCTUATION_CONFIG;
 
   var hasCustomTargets = Object.prototype.hasOwnProperty.call(opts, 'spaceAfterPunctuationTargets');
+  var addTargets = opts.spaceAfterPunctuationTargetsAdd;
+  var removeTargets = opts.spaceAfterPunctuationTargetsRemove;
+  if (!hasCustomTargets && addTargets === undefined && removeTargets === undefined) {
+    return DEFAULT_PUNCTUATION_CONFIG;
+  }
   var baseTargets;
 
   if (!hasCustomTargets) {
-    baseTargets = DEFAULT_PUNCTUATION_TARGETS.slice();
+    baseTargets = DEFAULT_PUNCTUATION_TARGETS;
   } else {
     var customTargets = opts.spaceAfterPunctuationTargets;
     if (customTargets === null || customTargets === false) return null;
@@ -63,13 +72,12 @@ function resolve_punctuation_targets(opts) {
       baseTargets = [ customTargets ];
     } else if (Array.isArray(customTargets)) {
       if (customTargets.length === 0) return null;
-      baseTargets = customTargets.slice();
+      baseTargets = customTargets;
     } else {
-      baseTargets = DEFAULT_PUNCTUATION_TARGETS.slice();
+      baseTargets = DEFAULT_PUNCTUATION_TARGETS;
     }
   }
 
-  var addTargets = opts.spaceAfterPunctuationTargetsAdd;
   if (addTargets !== undefined) {
     var addList = [];
     if (typeof addTargets === 'string') {
@@ -82,7 +90,6 @@ function resolve_punctuation_targets(opts) {
     }
   }
 
-  var removeTargets = opts.spaceAfterPunctuationTargetsRemove;
   if (removeTargets !== undefined) {
     var removeList = [];
     if (typeof removeTargets === 'string') {
@@ -140,15 +147,15 @@ function get_cjk_width_class(ch) {
   return width === 'F' || width === 'W' || width === 'H' ? width : '';
 }
 
-function build_next_text_info(tokens) {
+function build_next_text_info(tokens, trackSkippedEmpty) {
   var nextTextIndex = new Array(tokens.length);
-  var nextSkippedEmpty = new Array(tokens.length);
+  var nextSkippedEmpty = trackSkippedEmpty ? new Array(tokens.length) : null;
   var nextNonEmpty = -1;
   var sawEmpty = false;
 
   for (var idx = tokens.length - 1; idx >= 0; idx--) {
     nextTextIndex[idx] = nextNonEmpty;
-    nextSkippedEmpty[idx] = sawEmpty;
+    if (trackSkippedEmpty) nextSkippedEmpty[idx] = sawEmpty;
 
     var token = tokens[idx];
     if (!token || token.type !== 'text') continue;
@@ -170,7 +177,7 @@ function build_next_text_info(tokens) {
 
 
 function process_inlines(tokens, state, ctx, inlineToken) {
-  var i, last, trailing, next, c1, c2, remove_break;
+  var i, last, next, c1, c2, remove_break;
   var either = ctx.either;
   var normalizeSoftBreaks = ctx.normalizeSoftBreaks;
   var punctuationSpace = ctx.punctuationSpace;
@@ -179,15 +186,27 @@ function process_inlines(tokens, state, ctx, inlineToken) {
   var considerInlineBoundaries = ctx.considerInlineBoundaries;
   var needsPunctuation = punctuationSpace && punctuationConfig && maxPunctuationLength > 0;
 
+  if (!tokens || tokens.length === 0) return;
   if (normalizeSoftBreaks) normalize_text_tokens(tokens);
 
-  var nextInfo = build_next_text_info(tokens);
-  var nextTextIndex = nextInfo.nextTextIndex;
-  var nextSkippedEmpty = nextInfo.nextSkippedEmpty;
-
-  var widthCache = Object.create(null);
+  var nextTextIndex = null;
+  var nextSkippedEmpty = null;
+  var widthCache = null;
   function get_cached_width_class(ch) {
     if (!ch) return '';
+    if (!widthCache) {
+      var firstWidth = get_cjk_width_class(ch);
+      if (firstWidth === '') {
+        var codePoint = ch.codePointAt(0);
+        if (codePoint !== undefined && codePoint <= ASCII_PRINTABLE_MAX) return '';
+        widthCache = Object.create(null);
+        widthCache[ch] = '';
+        return '';
+      }
+      widthCache = Object.create(null);
+      widthCache[ch] = firstWidth;
+      return firstWidth;
+    }
     var cached = widthCache[ch];
     if (cached !== undefined) return cached;
     var width = get_cjk_width_class(ch);
@@ -204,22 +223,24 @@ function process_inlines(tokens, state, ctx, inlineToken) {
     var isSoftbreakToken = token.type === 'softbreak';
     var isTextBreakToken = token.type === 'text' && token.content === '\n';
     if (isSoftbreakToken || isTextBreakToken) {
+      if (!nextTextIndex) {
+        var nextInfo = build_next_text_info(tokens, considerInlineBoundaries);
+        nextTextIndex = nextInfo.nextTextIndex;
+        nextSkippedEmpty = nextInfo.nextSkippedEmpty;
+      }
       // default last/next character to space
       last = next = ' ';
-      trailing = '';
-      var trailingMatchesPunctuation = false;
-
-      var skippedEmptyBefore = sawEmptySinceLast;
-      var skippedEmptyAfter = nextSkippedEmpty[i];
+      var skippedEmptyBefore = false;
+      var skippedEmptyAfter = false;
+      if (considerInlineBoundaries) {
+        skippedEmptyBefore = sawEmptySinceLast;
+        skippedEmptyAfter = nextSkippedEmpty ? nextSkippedEmpty[i] : false;
+      }
 
       if (hasLastText) {
         c1 = lastTextContent.charCodeAt(lastTextContent.length - 2);
         c2 = lastTextContent.charCodeAt(lastTextContent.length - 1);
         last = lastTextContent.slice(is_surrogate(c1, c2) ? -2 : -1);
-        if (needsPunctuation) {
-          trailing = lastTextContent.slice(-maxPunctuationLength);
-          trailingMatchesPunctuation = matches_punctuation_sequence(trailing, punctuationConfig);
-        }
       }
 
       var nextIdx = nextTextIndex[i];
@@ -233,46 +254,44 @@ function process_inlines(tokens, state, ctx, inlineToken) {
       }
 
       remove_break = false;
-
-      // remove newline if it's adjacent to ZWSP
-      if (last === '\u200b' || next === '\u200b') remove_break = true;
-
-      var lastWidthClass = '';
       var nextWidthClass = '';
+      var nextWidthComputed = false;
 
-      // remove newline if both characters AND/OR fullwidth (F), wide (W) or
-      // halfwidth (H), but not Hangul
-      var tLast = false;
-      var tNext = false;
-
-      var needsWidthForRemoval = !remove_break;
-      var needsWidthForPunctuation = punctuationSpace && trailingMatchesPunctuation && last && next && next !== '\u200b';
-
-      if (needsWidthForRemoval || needsWidthForPunctuation) {
-        if (needsWidthForRemoval) {
-          lastWidthClass = get_cached_width_class(last);
+      // remove newline if it's adjacent to ZWSP
+      if (last === '\u200b' || next === '\u200b') {
+        remove_break = true;
+      } else {
+        // remove newline if both characters AND/OR fullwidth (F), wide (W) or
+        // halfwidth (H), but not Hangul
+        var lastWidthClass = get_cached_width_class(last);
+        if (either || lastWidthClass) {
+          nextWidthClass = get_cached_width_class(next);
+          nextWidthComputed = true;
         }
-        nextWidthClass = get_cached_width_class(next);
 
-        if (needsWidthForRemoval) {
-          tLast = lastWidthClass !== '';
-          tNext = nextWidthClass !== '';
+        var tLast = lastWidthClass !== '';
+        var tNext = nextWidthComputed ? nextWidthClass !== '' : false;
 
-          if (considerInlineBoundaries && (skippedEmptyBefore || skippedEmptyAfter) && tLast && tNext) {
-            tLast = false;
-            tNext = false;
-          }
-          if (either ? tLast || tNext : tLast && tNext) {
-            if (!is_hangul(last) && !is_hangul(next)) remove_break = true;
-          }
+        if (considerInlineBoundaries && (skippedEmptyBefore || skippedEmptyAfter) && tLast && tNext) {
+          tLast = false;
+          tNext = false;
+        }
+        if (either ? tLast || tNext : tLast && tNext) {
+          if (!is_hangul(last) && !is_hangul(next)) remove_break = true;
         }
       }
 
       if (remove_break) {
         var insertPunctuationSpace = false;
-        if (punctuationSpace && punctuationConfig && trailingMatchesPunctuation && last && next && next !== '\u200b') {
-          var nextIsFullwidthOrWide = nextWidthClass === 'F' || nextWidthClass === 'W';
-          if (is_printable_ascii(next) || nextIsFullwidthOrWide) insertPunctuationSpace = true;
+        if (needsPunctuation && last && next && next !== '\u200b') {
+          var trailing = hasLastText ? lastTextContent.slice(-maxPunctuationLength) : '';
+          if (matches_punctuation_sequence(trailing, punctuationConfig)) {
+            if (!nextWidthComputed) {
+              nextWidthClass = get_cached_width_class(next);
+            }
+            var nextIsFullwidthOrWide = nextWidthClass === 'F' || nextWidthClass === 'W';
+            if (nextIsFullwidthOrWide || is_printable_ascii(next)) insertPunctuationSpace = true;
+          }
         }
         token.type    = 'text';
         token.content = insertPunctuationSpace ? punctuationSpace : '';
@@ -281,16 +300,16 @@ function process_inlines(tokens, state, ctx, inlineToken) {
 
     if (token.type === 'text') {
       if (!token.content) {
-        sawEmptySinceLast = true;
+        if (considerInlineBoundaries) sawEmptySinceLast = true;
       } else {
         lastTextContent = token.content;
         hasLastText = true;
-        sawEmptySinceLast = false;
+        if (considerInlineBoundaries) sawEmptySinceLast = false;
       }
     }
   }
 
-  if (punctuationSpace && punctuationConfig) {
+  if (needsPunctuation) {
     apply_missing_punctuation_spacing(tokens, inlineToken, punctuationSpace, punctuationConfig);
   }
 }
@@ -386,6 +405,8 @@ function apply_missing_punctuation_spacing(tokens, inlineToken, punctuationSpace
   if (!inlineToken || !inlineToken.content) return;
   if (inlineToken.content.indexOf('\n') === -1) return;
   if (!tokens || tokens.length === 0) return;
+  var maxPunctuationLength = punctuationConfig.maxLength;
+  if (maxPunctuationLength <= 0) return;
 
   var rawSearchState = { pos: 0 };
 
@@ -393,15 +414,13 @@ function apply_missing_punctuation_spacing(tokens, inlineToken, punctuationSpace
     var current = tokens[idx];
     if (!current || current.type !== 'text' || !current.content) continue;
 
-    var trailing = punctuationConfig.maxLength > 0 ?
-      current.content.slice(-punctuationConfig.maxLength) :
-      current.content.slice(-1);
+    var trailing = current.content.slice(-maxPunctuationLength);
     if (!matches_punctuation_sequence(trailing, punctuationConfig)) continue;
-    if (/\s$/.test(current.content)) continue;
+    if (WHITESPACE_TRAIL_RE.test(current.content)) continue;
 
     var nextInfo = find_next_visible_token(tokens, idx + 1);
     if (!nextInfo) continue;
-    if (nextInfo.token.type === 'text' && /^\s/.test(nextInfo.token.content || '')) continue;
+    if (nextInfo.token.type === 'text' && WHITESPACE_LEAD_RE.test(nextInfo.token.content || '')) continue;
 
     if (!raw_boundary_includes_newline(inlineToken.content, tokens, idx, nextInfo.index, nextInfo.fragment, rawSearchState)) {
       continue;
@@ -411,7 +430,9 @@ function apply_missing_punctuation_spacing(tokens, inlineToken, punctuationSpace
     idx = nextInfo.index;
   }
 
-  apply_single_text_token_spacing(tokens, inlineToken, punctuationSpace, punctuationConfig);
+  if (tokens.length === 1) {
+    apply_single_text_token_spacing(tokens, inlineToken, punctuationSpace, punctuationConfig);
+  }
 }
 
 
@@ -481,6 +502,8 @@ function apply_single_text_token_spacing(tokens, inlineToken, punctuationSpace,
   if (inlineToken.content.indexOf('\n') === -1) return;
   var token = tokens[0];
   if (!token || token.type !== 'text' || !token.content) return;
+  var maxPunctuationLength = punctuationConfig.maxLength;
+  if (maxPunctuationLength <= 0) return;
 
   var segments = inlineToken.content.split('\n');
   if (segments.length < 2) return;
@@ -490,7 +513,7 @@ function apply_single_text_token_spacing(tokens, inlineToken, punctuationSpace,
   for (var segIdx = 0; segIdx < segments.length - 1; segIdx++) {
     var leftRaw = segments[segIdx];
     var rightRaw = segments[segIdx + 1];
-    var tail = extract_visible_tail(leftRaw, punctuationConfig.maxLength);
+    var tail = extract_visible_tail(leftRaw, maxPunctuationLength);
     var nextChar = extract_visible_head(rightRaw);
     var shouldInsert = tail &&
       matches_punctuation_sequence(tail, punctuationConfig) &&
@@ -501,7 +524,7 @@ function apply_single_text_token_spacing(tokens, inlineToken, punctuationSpace,
       var splitIndex = cumulativeLength + leftRaw.length + offsetDelta;
       if (splitIndex >= 0 && splitIndex <= updatedContent.length) {
         var existingChar = updatedContent.charAt(splitIndex);
-        if (existingChar && /\s/.test(existingChar)) {
+        if (existingChar && WHITESPACE_RE.test(existingChar)) {
           // already has whitespace at this boundary
           cumulativeLength += leftRaw.length;
           continue;
@@ -528,7 +551,7 @@ function extract_visible_tail(raw, maxLength) {
     var charLen = code > 0xFFFF ? 2 : 1;
     var ch = raw.slice(pos - charLen, pos);
     pos -= charLen;
-    if (/\s/.test(ch)) continue;
+    if (WHITESPACE_RE.test(ch)) continue;
     if (is_markup_closer_char(ch)) continue;
     result = ch + result;
   }
@@ -543,7 +566,7 @@ function extract_visible_head(raw) {
     var charLen = code > 0xFFFF ? 2 : 1;
     var ch = raw.slice(pos, pos + charLen);
     pos += charLen;
-    if (/\s/.test(ch)) continue;
+    if (WHITESPACE_RE.test(ch)) continue;
     return ch;
   }
   return '';

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@peaceroad/markdown-it-cjk-breaks-mod",
-  "version": "0.1.2",
+  "version": "0.1.4",
   "type": "module",
   "description": "Suppress linebreaks between east asian (Especially Japanese) characters",
   "repository": {