@peaceroad/markdown-it-cjk-breaks-mod 0.1.9 → 0.1.10

package/index.js

@@ -350,10 +350,7 @@ function normalize_text_tokens(tokens) {
       normalized = tokens.slice(0, idx);
     }
 
-    var replacement = split_text_token(token);
-    for (var r = 0; r < replacement.length; r++) {
-      normalized.push(replacement[r]);
-    }
+    append_split_text_token(normalized, token);
   }
 
   if (normalized) {
@@ -365,9 +362,8 @@ function normalize_text_tokens(tokens) {
 }
 
 
-function split_text_token(token) {
+function append_split_text_token(target, token) {
   var TokenConstructor = token.constructor;
-  var parts = [];
   var content = token.content;
   var start = 0;
   var reusedToken = false;
@@ -376,11 +372,11 @@ function split_text_token(token) {
     if (!text) return;
     if (!reusedToken) {
       token.content = text;
-      parts.push(token);
+      target.push(token);
       reusedToken = true;
       return;
     }
-    parts.push(clone_text_token(TokenConstructor, token, text));
+    target.push(clone_text_token(TokenConstructor, token, text));
   }
 
   for (var pos = 0; pos < content.length; pos++) {
@@ -388,13 +384,11 @@ function split_text_token(token) {
 
     if (pos > start) push_text_part(content.slice(start, pos));
 
-    parts.push(create_softbreak_token(TokenConstructor, token));
+    target.push(create_softbreak_token(TokenConstructor, token));
     start = pos + 1;
   }
 
   if (start < content.length) push_text_part(content.slice(start));
-
-  return parts;
 }
 
 
@@ -475,25 +469,24 @@ function apply_missing_punctuation_spacing(tokens, inlineToken, punctuationSpace
 function raw_boundary_includes_newline(source, beforeFragment, betweenFragment, afterFragment, state) {
   if (!source || !afterFragment) return false;
   if (!beforeFragment) return false;
-  betweenFragment = betweenFragment || '';
-  if (Array.isArray(afterFragment)) {
-    for (var i = 0; i < afterFragment.length; i++) {
-      var fragment = afterFragment[i];
-      if (!fragment) continue;
-      var candidate = beforeFragment + betweenFragment + '\n' + fragment;
-      var startPos = source.indexOf(candidate, state.pos);
-      if (startPos === -1) continue;
-      state.pos = startPos + candidate.length - fragment.length;
-      return true;
-    }
-    return false;
+  var beforeBoundary = betweenFragment ? beforeFragment + betweenFragment : beforeFragment;
+  var newlinePositions = get_newline_positions(source, state);
+  var startIndex = state.newlineIndex || 0;
+
+  while (startIndex < newlinePositions.length && newlinePositions[startIndex] < state.pos) {
+    startIndex++;
+  }
+  state.newlineIndex = startIndex;
+
+  for (var idx = startIndex; idx < newlinePositions.length; idx++) {
+    var newlinePos = newlinePositions[idx];
+    if (!matches_raw_newline_boundary(source, newlinePos, beforeBoundary, afterFragment)) continue;
+    state.pos = newlinePos + 1;
+    state.newlineIndex = idx + 1;
+    return true;
   }
-  var fragment = afterFragment;
-  var candidate = beforeFragment + betweenFragment + '\n' + fragment;
-  var startPos = source.indexOf(candidate, state.pos);
-  if (startPos === -1) return false;
-  state.pos = startPos + candidate.length - fragment.length;
-  return true;
+
+  return false;
 }
 
 
@@ -517,6 +510,33 @@ function find_next_visible_token(tokens, startIdx) {
 }
 
 
+function get_newline_positions(source, state) {
+  if (state.newlinePositions) return state.newlinePositions;
+  var newlinePositions = [];
+  for (var idx = 0; idx < source.length; idx++) {
+    if (source.charCodeAt(idx) === 0x0A) newlinePositions.push(idx);
+  }
+  state.newlinePositions = newlinePositions;
+  return newlinePositions;
+}
+
+
+function matches_raw_newline_boundary(source, newlinePos, beforeBoundary, afterFragment) {
+  var beforeStart = newlinePos - beforeBoundary.length;
+  if (beforeStart < 0) return false;
+  if (!source.startsWith(beforeBoundary, beforeStart)) return false;
+  var afterStart = newlinePos + 1;
+  if (Array.isArray(afterFragment)) {
+    for (var i = 0; i < afterFragment.length; i++) {
+      var fragment = afterFragment[i];
+      if (fragment && source.startsWith(fragment, afterStart)) return true;
+    }
+    return false;
+  }
+  return source.startsWith(afterFragment, afterStart);
+}
+
+
 function derive_after_fragment(token) {
   if (!token) return '';
   if (token.type === 'text' || token.type === 'html_inline' || token.type === 'code_inline') {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@peaceroad/markdown-it-cjk-breaks-mod",
-  "version": "0.1.9",
+  "version": "0.1.10",
   "type": "module",
   "description": "Suppress linebreaks between east asian (Especially Japanese) characters",
   "repository": {