Add special handling for apostrophe handling

lib/Epub/Epub/hyphenation/HyphenationCommon.cpp

@@ -107,6 +107,17 @@ bool isPunctuation(const uint32_t cp) {
 
 bool isAsciiDigit(const uint32_t cp) { return cp >= '0' && cp <= '9'; }
 
+bool isApostrophe(const uint32_t cp) {
+  switch (cp) {
+    case '\'':
+    case 0x2018:  // left single quotation mark
+    case 0x2019:  // right single quotation mark
+      return true;
+    default:
+      return false;
+  }
+}
+
 bool isExplicitHyphen(const uint32_t cp) {
   switch (cp) {
     case '-':

lib/Epub/Epub/hyphenation/HyphenationCommon.h

@@ -19,6 +19,7 @@ bool isCyrillicLetter(uint32_t cp);
 bool isAlphabetic(uint32_t cp);
 bool isPunctuation(uint32_t cp);
 bool isAsciiDigit(uint32_t cp);
+bool isApostrophe(uint32_t cp);
 bool isExplicitHyphen(uint32_t cp);
 bool isSoftHyphen(uint32_t cp);
 void trimSurroundingPunctuationAndFootnote(std::vector<CodepointInfo>& cps);

lib/Epub/Epub/hyphenation/Hyphenator.cpp

@@ -59,6 +59,83 @@ std::vector<Hyphenator::BreakInfo> buildExplicitBreakInfos(const std::vector<Cod
   return breaks;
 }
 
+bool isSegmentSeparator(const uint32_t cp) { return isExplicitHyphen(cp) || isApostrophe(cp); }
+
+void appendSegmentPatternBreaks(const std::vector<CodepointInfo>& cps, const LanguageHyphenator& hyphenator,
+                                const bool includeFallback, std::vector<Hyphenator::BreakInfo>& outBreaks) {
+  size_t segStart = 0;
+
+  for (size_t i = 0; i <= cps.size(); ++i) {
+    const bool atEnd = i == cps.size();
+    const bool atSeparator = !atEnd && isSegmentSeparator(cps[i].value);
+    if (!atEnd && !atSeparator) {
+      continue;
+    }
+
+    if (i > segStart) {
+      std::vector<CodepointInfo> segment(cps.begin() + segStart, cps.begin() + i);
+      auto segIndexes = hyphenator.breakIndexes(segment);
+
+      if (includeFallback && segIndexes.empty()) {
+        const size_t minPrefix = hyphenator.minPrefix();
+        const size_t minSuffix = hyphenator.minSuffix();
+        for (size_t idx = minPrefix; idx + minSuffix <= segment.size(); ++idx) {
+          segIndexes.push_back(idx);
+        }
+      }
+
+      for (const size_t idx : segIndexes) {
+        const size_t cpIdx = segStart + idx;
+        if (cpIdx < cps.size()) {
+          outBreaks.push_back({cps[cpIdx].byteOffset, true});
+        }
+      }
+    }
+
+    segStart = i + 1;
+  }
+}
+
+void appendApostropheContractionBreaks(const std::vector<CodepointInfo>& cps,
+                                       std::vector<Hyphenator::BreakInfo>& outBreaks) {
+  size_t segmentStart = 0;
+
+  for (size_t i = 0; i < cps.size(); ++i) {
+    if (isSegmentSeparator(cps[i].value)) {
+      if (isApostrophe(cps[i].value) && i > 0 && i + 1 < cps.size() && isAlphabetic(cps[i - 1].value) &&
+          isAlphabetic(cps[i + 1].value)) {
+        size_t leftPrefixLen = 0;
+        for (size_t j = segmentStart; j < i; ++j) {
+          if (isAlphabetic(cps[j].value)) {
+            ++leftPrefixLen;
+          }
+        }
+
+        // Avoid stranding very short clitics like "l'" or "d'" at line end.
+        if (leftPrefixLen >= 3) {
+          outBreaks.push_back({cps[i + 1].byteOffset, false});
+        }
+      }
+      segmentStart = i + 1;
+    }
+  }
+}
+
+void sortAndDedupeBreakInfos(std::vector<Hyphenator::BreakInfo>& infos) {
+  std::sort(infos.begin(), infos.end(), [](const Hyphenator::BreakInfo& a, const Hyphenator::BreakInfo& b) {
+    if (a.byteOffset != b.byteOffset) {
+      return a.byteOffset < b.byteOffset;
+    }
+    return a.requiresInsertedHyphen < b.requiresInsertedHyphen;
+  });
+
+  infos.erase(std::unique(infos.begin(), infos.end(),
+                          [](const Hyphenator::BreakInfo& a, const Hyphenator::BreakInfo& b) {
+                            return a.byteOffset == b.byteOffset && a.requiresInsertedHyphen == b.requiresInsertedHyphen;
+                          }),
+              infos.end());
+}
+
 }  // namespace
 
 std::vector<Hyphenator::BreakInfo> Hyphenator::breakOffsets(const std::string& word, const bool includeFallback) {
@@ -89,31 +166,34 @@ std::vector<Hyphenator::BreakInfo> Hyphenator::breakOffsets(const std::string& w
     //                                            @16 Satellitensys|tems  (+hyphen)
     //   Result: 6 sorted break points; the line-breaker picks the widest prefix that fits.
     if (hyphenator) {
-      size_t segStart = 0;
+      appendSegmentPatternBreaks(cps, *hyphenator, /*includeFallback=*/false, explicitBreakInfos);
-      for (size_t i = 0; i <= cps.size(); ++i) {
-        const bool atEnd = (i == cps.size());
-        const bool atHyphen = !atEnd && isExplicitHyphen(cps[i].value);
-        if (atEnd || atHyphen) {
-          if (i > segStart) {
-            std::vector<CodepointInfo> segment(cps.begin() + segStart, cps.begin() + i);
-            auto segIndexes = hyphenator->breakIndexes(segment);
-            for (const size_t idx : segIndexes) {
-              const size_t cpIdx = segStart + idx;
-              if (cpIdx < cps.size()) {
-                explicitBreakInfos.push_back({cps[cpIdx].byteOffset, true});
-              }
-            }
-          }
-          segStart = i + 1;
-        }
-      }
       // Merge explicit and pattern breaks into ascending byte-offset order.
-      std::sort(explicitBreakInfos.begin(), explicitBreakInfos.end(),
+      sortAndDedupeBreakInfos(explicitBreakInfos);
-                [](const BreakInfo& a, const BreakInfo& b) { return a.byteOffset < b.byteOffset; });
     }
     return explicitBreakInfos;
   }
 
+  // Apostrophe-like separators split compounds into alphabetic segments; run Liang on each segment.
+  // This allows words like "all'improvviso" to hyphenate within "improvviso" instead of becoming
+  // completely unsplittable due to the apostrophe punctuation.
+  bool hasApostropheLikeSeparator = false;
+  for (const auto& cp : cps) {
+    if (isApostrophe(cp.value)) {
+      hasApostropheLikeSeparator = true;
+      break;
+    }
+  }
+
+  if (hyphenator && hasApostropheLikeSeparator) {
+    std::vector<BreakInfo> segmentedBreaks;
+    appendSegmentPatternBreaks(cps, *hyphenator, includeFallback, segmentedBreaks);
+    appendApostropheContractionBreaks(cps, segmentedBreaks);
+    if (!segmentedBreaks.empty()) {
+      sortAndDedupeBreakInfos(segmentedBreaks);
+      return segmentedBreaks;
+    }
+  }
+
   // Ask language hyphenator for legal break points.
   std::vector<size_t> indexes;
   if (hyphenator) {

lib/Epub/Epub/hyphenation/Hyphenator.h

@@ -11,7 +11,8 @@ class Hyphenator {
   struct BreakInfo {
     size_t byteOffset;            // Byte position inside the UTF-8 word where a break may occur.
     bool requiresInsertedHyphen;  // true = a visible '-' must be rendered at the break (pattern/fallback breaks).
-                                  // false = the word already contains a hyphen at this position (explicit '-').
+                                  // false = break occurs at an existing visible separator boundary
+                                  //         (explicit '-' or eligible apostrophe contraction boundary).
   };
 
   // Returns byte offsets where the word may be hyphenated.
@@ -19,12 +20,16 @@ class Hyphenator {
   // Break sources (in priority order):
   //   1. Explicit hyphens already present in the word (e.g. '-' or soft-hyphen U+00AD).
   //      When found, language patterns are additionally run on each alphabetic segment
-  //      between hyphens so compound words can break within their parts.
+  //      between separators so compound words can break within their parts.
   //      Example: "US-Satellitensystems" yields breaks after "US-" (no inserted hyphen)
   //               plus pattern breaks inside "Satellitensystems" (Sa|tel|li|ten|sys|tems).
-  //   2. Language-specific Liang patterns (e.g. German de_patterns).
+  //   2. Apostrophe contractions between letters (e.g. all'improvviso).
+  //      Liang patterns are run per alphabetic segment around apostrophes.
+  //      A direct break at the apostrophe boundary is allowed only when the left
+  //      segment has at least 3 letters to avoid stranding short clitics (e.g. l', d').
+  //   3. Language-specific Liang patterns (e.g. German de_patterns).
   //      Example: "Quadratkilometer" -> Qua|drat|ki|lo|me|ter.
-  //   3. Fallback every-N-chars splitting (only when includeFallback is true AND no
+  //   4. Fallback every-N-chars splitting (only when includeFallback is true AND no
   //      pattern breaks were found). Used as a last resort to prevent a single oversized
   //      word from overflowing the page width.
   static std::vector<BreakInfo> breakOffsets(const std::string& word, bool includeFallback);