From 9bb636402a1c1d2eedf458e78deb98fc9739c848 Mon Sep 17 00:00:00 2001
From: Calixte Denizet <calixte.denizet@gmail.com>
Date: Fri, 7 Jan 2022 14:02:28 +0100
Subject: [PATCH 1/2] Use the correct dimension to know if we have to add an
 EOL in vertical mode

---
 src/core/evaluator.js | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/core/evaluator.js b/src/core/evaluator.js
index 3e0dd8492..6a0d54035 100644
--- a/src/core/evaluator.js
+++ b/src/core/evaluator.js
@@ -2476,7 +2476,7 @@ class PartialEvaluator {
           return;
         }
 
-        if (Math.abs(advanceX) > textContentItem.height) {
+        if (Math.abs(advanceX) > textContentItem.width) {
           appendEOL();
           return;
         }

From 9dae421a0d2e060fd0b066cc8ae4d5e58e70266f Mon Sep 17 00:00:00 2001
From: Calixte Denizet <calixte.denizet@gmail.com>
Date: Fri, 7 Jan 2022 21:20:53 +0100
Subject: [PATCH 2/2] Handle all the whitespaces the same way when creating
 text chunks

---
 src/core/evaluator.js | 13 +++++++------
 src/core/fonts.js     | 15 ++++++++++-----
 src/core/unicode.js   | 10 ++++++++++
 3 files changed, 27 insertions(+), 11 deletions(-)

diff --git a/src/core/evaluator.js b/src/core/evaluator.js
index 6a0d54035..a189281e0 100644
--- a/src/core/evaluator.js
+++ b/src/core/evaluator.js
@@ -2171,7 +2171,6 @@ class PartialEvaluator {
     stateManager = stateManager || new StateManager(new TextState());
 
     const WhitespaceRegexp = /\s/g;
-    const DiacriticRegExp = new RegExp("^\\p{Mn}$", "u");
     const NormalizedUnicodes = getNormalizedUnicodes();
 
     const textContent = {
@@ -2572,6 +2571,7 @@ class PartialEvaluator {
 
       const glyphs = font.charsToGlyphs(chars);
       const scale = textState.fontMatrix[0] * textState.fontSize;
+
       for (let i = 0, ii = glyphs.length; i < ii; i++) {
         const glyph = glyphs[i];
         let charSpacing =
@@ -2583,13 +2583,12 @@ class PartialEvaluator {
         }
         let scaledDim = glyphWidth * scale;
 
-        let glyphUnicode = glyph.unicode;
         if (
-          glyphUnicode === " " &&
+          glyph.isWhitespace &&
           (i === 0 ||
             i + 1 === ii ||
-            glyphs[i - 1].unicode === " " ||
-            glyphs[i + 1].unicode === " " ||
+            glyphs[i - 1].isWhitespace ||
+            glyphs[i + 1].isWhitespace ||
             extraSpacing)
         ) {
           // Don't push a " " in the textContentItem
@@ -2612,10 +2611,12 @@ class PartialEvaluator {
 
         compareWithLastPosition();
 
+        let glyphUnicode = glyph.unicode;
+
         // Must be called after compareWithLastPosition because
         // the textContentItem could have been flushed.
         const textChunk = ensureTextContentItem();
-        if (DiacriticRegExp.test(glyph.unicode)) {
+        if (glyph.isDiacritic) {
           scaledDim = 0;
         }
 
diff --git a/src/core/fonts.js b/src/core/fonts.js
index 5946b6d49..3e20bbc18 100644
--- a/src/core/fonts.js
+++ b/src/core/fonts.js
@@ -34,6 +34,12 @@ import {
   recoverGlyphName,
   SEAC_ANALYSIS_ENABLED,
 } from "./fonts_utils.js";
+import {
+  getCharUnicodeCategory,
+  getUnicodeForGlyph,
+  getUnicodeRangeFor,
+  mapSpecialUnicodeValues,
+} from "./unicode.js";
 import { getDingbatsGlyphsUnicode, getGlyphsUnicode } from "./glyphlist.js";
 import {
   getEncoding,
@@ -50,11 +56,6 @@ import {
   getSupplementalGlyphMapForArialBlack,
   getSupplementalGlyphMapForCalibri,
 } from "./standard_fonts.js";
-import {
-  getUnicodeForGlyph,
-  getUnicodeRangeFor,
-  mapSpecialUnicodeValues,
-} from "./unicode.js";
 import { IdentityToUnicodeMap, ToUnicodeMap } from "./to_unicode_map.js";
 import { CFFFont } from "./cff_font.js";
 import { FontRendererFactory } from "./font_renderer.js";
@@ -212,6 +213,10 @@ class Glyph {
     this.operatorListId = operatorListId;
     this.isSpace = isSpace;
     this.isInFont = isInFont;
+
+    const category = getCharUnicodeCategory(unicode);
+    this.isWhitespace = category.isWhitespace;
+    this.isDiacritic = category.isDiacritic;
   }
 
   matchesForCache(
diff --git a/src/core/unicode.js b/src/core/unicode.js
index b9623308c..71c1c8d4f 100644
--- a/src/core/unicode.js
+++ b/src/core/unicode.js
@@ -1640,7 +1640,17 @@ function reverseIfRtl(chars) {
   return buf.join("");
 }
 
+const SpecialCharRegExp = new RegExp("^(\\s)|(\\p{Mn})$", "u");
+function getCharUnicodeCategory(char) {
+  const groups = char.match(SpecialCharRegExp);
+  return {
+    isWhitespace: !!(groups && groups[1]),
+    isDiacritic: !!(groups && groups[2]),
+  };
+}
+
 export {
+  getCharUnicodeCategory,
   getNormalizedUnicodes,
   getUnicodeForGlyph,
   getUnicodeRangeFor,