Consolidate chapter page data into single file (#144)

## Summary

* Consolidate chapter page data into single file
* Header structure of the file stays the same, following the page count,
we now put a LUT offset
   * The page data is all then appended to this file
* Finally the LUT is appended to the end of the file, and the page count
is updated
* This will also significantly improve the duration of cache cleanup
which takes a while to scan the directory and cleanup content
* Remove page file version as it's all tied up into the section file now
* Bumped section file version to 7
* Moved section content into sub directory
* Updated docs

## Additional Context

* Benchmarks:
  * Generating 74 pages of content from a chapter in Jade Legacy took:
    * master: 6,229ms
    * this PR: 1,305ms
    * Speedup of 79%
  * Generating 207 pages of content from Livesuit book:
    * With progress bar UI updates:
      * master: 24,250ms
      * this PR: 8,063ms
      * Speedup of 67%
    * Without progress bar UI updates:
      * master: 13,055ms
      * this PR: 3,600ms
      * Speedup of 72%

lib/Epub/Epub/Page.cpp

@@ -3,20 +3,16 @@
 #include <HardwareSerial.h>
 #include <Serialization.h>
 
-namespace {
-constexpr uint8_t PAGE_FILE_VERSION = 3;
-}
-
 void PageLine::render(GfxRenderer& renderer, const int fontId, const int xOffset, const int yOffset) {
   block->render(renderer, fontId, xPos + xOffset, yPos + yOffset);
 }
 
-void PageLine::serialize(File& file) {
+bool PageLine::serialize(File& file) {
   serialization::writePod(file, xPos);
   serialization::writePod(file, yPos);
 
   // serialize TextBlock pointed to by PageLine
-  block->serialize(file);
+  return block->serialize(file);
 }
 
 std::unique_ptr<PageLine> PageLine::deserialize(File& file) {
@@ -35,27 +31,22 @@ void Page::render(GfxRenderer& renderer, const int fontId, const int xOffset, co
   }
 }
 
-void Page::serialize(File& file) const {
-  serialization::writePod(file, PAGE_FILE_VERSION);
-
+bool Page::serialize(File& file) const {
   const uint32_t count = elements.size();
   serialization::writePod(file, count);
 
   for (const auto& el : elements) {
     // Only PageLine exists currently
     serialization::writePod(file, static_cast<uint8_t>(TAG_PageLine));
-    el->serialize(file);
+    if (!el->serialize(file)) {
+      return false;
+    }
   }
+
+  return true;
 }
 
 std::unique_ptr<Page> Page::deserialize(File& file) {
-  uint8_t version;
-  serialization::readPod(file, version);
-  if (version != PAGE_FILE_VERSION) {
-    Serial.printf("[%lu] [PGE] Deserialization failed: Unknown version %u\n", millis(), version);
-    return nullptr;
-  }
-
   auto page = std::unique_ptr<Page>(new Page());
 
   uint32_t count;

lib/Epub/Epub/Page.h

@@ -18,7 +18,7 @@ class PageElement {
   explicit PageElement(const int16_t xPos, const int16_t yPos) : xPos(xPos), yPos(yPos) {}
   virtual ~PageElement() = default;
   virtual void render(GfxRenderer& renderer, int fontId, int xOffset, int yOffset) = 0;
-  virtual void serialize(File& file) = 0;
+  virtual bool serialize(File& file) = 0;
 };
 
 // a line from a block element
@@ -29,7 +29,7 @@ class PageLine final : public PageElement {
   PageLine(std::shared_ptr<TextBlock> block, const int16_t xPos, const int16_t yPos)
       : PageElement(xPos, yPos), block(std::move(block)) {}
   void render(GfxRenderer& renderer, int fontId, int xOffset, int yOffset) override;
-  void serialize(File& file) override;
+  bool serialize(File& file) override;
   static std::unique_ptr<PageLine> deserialize(File& file);
 };
 
@@ -38,6 +38,6 @@ class Page {
   // the list of block index and line numbers on this page
   std::vector<std::shared_ptr<PageElement>> elements;
   void render(GfxRenderer& renderer, int fontId, int xOffset, int yOffset) const;
-  void serialize(File& file) const;
+  bool serialize(File& file) const;
   static std::unique_ptr<Page> deserialize(File& file);
 };

lib/Epub/Epub/Section.cpp

@@ -8,54 +8,60 @@
 #include "parsers/ChapterHtmlSlimParser.h"
 
 namespace {
-constexpr uint8_t SECTION_FILE_VERSION = 6;
+constexpr uint8_t SECTION_FILE_VERSION = 7;
+constexpr size_t HEADER_SIZE = sizeof(uint8_t) + sizeof(int) + sizeof(float) + sizeof(bool) + sizeof(int) +
+                               sizeof(int) + sizeof(int) + sizeof(size_t);
 }  // namespace
 
-void Section::onPageComplete(std::unique_ptr<Page> page) {
-  const auto filePath = cachePath + "/page_" + std::to_string(pageCount) + ".bin";
-
-  File outputFile;
-  if (!FsHelpers::openFileForWrite("SCT", filePath, outputFile)) {
-    return;
+size_t Section::onPageComplete(std::unique_ptr<Page> page) {
+  if (!file) {
+    Serial.printf("[%lu] [SCT] File not open for writing page %d\n", millis(), pageCount);
+    return 0;
   }
-  page->serialize(outputFile);
-  outputFile.close();
 
+  const auto position = file.position();
+  if (!page->serialize(file)) {
+    Serial.printf("[%lu] [SCT] Failed to serialize page %d\n", millis(), pageCount);
+    return 0;
+  }
   Serial.printf("[%lu] [SCT] Page %d processed\n", millis(), pageCount);
 
   pageCount++;
+  return position;
 }
 
-void Section::writeCacheMetadata(const int fontId, const float lineCompression, const bool extraParagraphSpacing,
-                                 const int viewportWidth, const int viewportHeight) const {
-  File outputFile;
-  if (!FsHelpers::openFileForWrite("SCT", cachePath + "/section.bin", outputFile)) {
+void Section::writeSectionFileHeader(const int fontId, const float lineCompression, const bool extraParagraphSpacing,
+                                     const int viewportWidth, const int viewportHeight) {
+  if (!file) {
+    Serial.printf("[%lu] [SCT] File not open for writing header\n", millis());
     return;
   }
-  serialization::writePod(outputFile, SECTION_FILE_VERSION);
-  serialization::writePod(outputFile, fontId);
-  serialization::writePod(outputFile, lineCompression);
-  serialization::writePod(outputFile, extraParagraphSpacing);
-  serialization::writePod(outputFile, viewportWidth);
-  serialization::writePod(outputFile, viewportHeight);
-  serialization::writePod(outputFile, pageCount);
-  outputFile.close();
+  static_assert(HEADER_SIZE == sizeof(SECTION_FILE_VERSION) + sizeof(fontId) + sizeof(lineCompression) +
+                                   sizeof(extraParagraphSpacing) + sizeof(viewportWidth) + sizeof(viewportHeight) +
+                                   sizeof(pageCount) + sizeof(size_t),
+                "Header size mismatch");
+  serialization::writePod(file, SECTION_FILE_VERSION);
+  serialization::writePod(file, fontId);
+  serialization::writePod(file, lineCompression);
+  serialization::writePod(file, extraParagraphSpacing);
+  serialization::writePod(file, viewportWidth);
+  serialization::writePod(file, viewportHeight);
+  serialization::writePod(file, pageCount);  // Placeholder for page count (will be initially 0 when written)
+  serialization::writePod(file, static_cast<size_t>(0));  // Placeholder for LUT offset
 }
 
-bool Section::loadCacheMetadata(const int fontId, const float lineCompression, const bool extraParagraphSpacing,
-                                const int viewportWidth, const int viewportHeight) {
-  const auto sectionFilePath = cachePath + "/section.bin";
-  File inputFile;
-  if (!FsHelpers::openFileForRead("SCT", sectionFilePath, inputFile)) {
+bool Section::loadSectionFile(const int fontId, const float lineCompression, const bool extraParagraphSpacing,
+                              const int viewportWidth, const int viewportHeight) {
+  if (!FsHelpers::openFileForRead("SCT", filePath, file)) {
     return false;
   }
 
   // Match parameters
   {
     uint8_t version;
-    serialization::readPod(inputFile, version);
+    serialization::readPod(file, version);
     if (version != SECTION_FILE_VERSION) {
-      inputFile.close();
+      file.close();
       Serial.printf("[%lu] [SCT] Deserialization failed: Unknown version %u\n", millis(), version);
       clearCache();
       return false;
@@ -64,41 +70,36 @@ bool Section::loadCacheMetadata(const int fontId, const float lineCompression, c
     int fileFontId, fileViewportWidth, fileViewportHeight;
     float fileLineCompression;
     bool fileExtraParagraphSpacing;
-    serialization::readPod(inputFile, fileFontId);
-    serialization::readPod(inputFile, fileLineCompression);
-    serialization::readPod(inputFile, fileExtraParagraphSpacing);
-    serialization::readPod(inputFile, fileViewportWidth);
-    serialization::readPod(inputFile, fileViewportHeight);
+    serialization::readPod(file, fileFontId);
+    serialization::readPod(file, fileLineCompression);
+    serialization::readPod(file, fileExtraParagraphSpacing);
+    serialization::readPod(file, fileViewportWidth);
+    serialization::readPod(file, fileViewportHeight);
 
     if (fontId != fileFontId || lineCompression != fileLineCompression ||
         extraParagraphSpacing != fileExtraParagraphSpacing || viewportWidth != fileViewportWidth ||
         viewportHeight != fileViewportHeight) {
-      inputFile.close();
+      file.close();
       Serial.printf("[%lu] [SCT] Deserialization failed: Parameters do not match\n", millis());
       clearCache();
       return false;
     }
   }
 
-  serialization::readPod(inputFile, pageCount);
-  inputFile.close();
+  serialization::readPod(file, pageCount);
+  file.close();
   Serial.printf("[%lu] [SCT] Deserialization succeeded: %d pages\n", millis(), pageCount);
   return true;
 }
 
-void Section::setupCacheDir() const {
-  epub->setupCacheDir();
-  SD.mkdir(cachePath.c_str());
-}
-
 // Your updated class method (assuming you are using the 'SD' object, which is a wrapper for a specific filesystem)
 bool Section::clearCache() const {
-  if (!SD.exists(cachePath.c_str())) {
+  if (!SD.exists(filePath.c_str())) {
     Serial.printf("[%lu] [SCT] Cache does not exist, no action needed\n", millis());
     return true;
   }
 
-  if (!FsHelpers::removeDir(cachePath.c_str())) {
+  if (!SD.remove(filePath.c_str())) {
     Serial.printf("[%lu] [SCT] Failed to clear cache\n", millis());
     return false;
   }
@@ -107,10 +108,10 @@ bool Section::clearCache() const {
   return true;
 }
 
-bool Section::persistPageDataToSD(const int fontId, const float lineCompression, const bool extraParagraphSpacing,
-                                  const int viewportWidth, const int viewportHeight,
-                                  const std::function<void()>& progressSetupFn,
-                                  const std::function<void(int)>& progressFn) {
+bool Section::createSectionFile(const int fontId, const float lineCompression, const bool extraParagraphSpacing,
+                                const int viewportWidth, const int viewportHeight,
+                                const std::function<void()>& progressSetupFn,
+                                const std::function<void(int)>& progressFn) {
   constexpr size_t MIN_SIZE_FOR_PROGRESS = 50 * 1024;  // 50KB
   const auto localPath = epub->getSpineItem(spineIndex).href;
   const auto tmpHtmlPath = epub->getCachePath() + "/.tmp_" + std::to_string(spineIndex) + ".html";
@@ -156,30 +157,66 @@ bool Section::persistPageDataToSD(const int fontId, const float lineCompression,
     progressSetupFn();
   }
 
+  if (!FsHelpers::openFileForWrite("SCT", filePath, file)) {
+    return false;
+  }
+  writeSectionFileHeader(fontId, lineCompression, extraParagraphSpacing, viewportWidth, viewportHeight);
+  std::vector<size_t> lut = {};
+
   ChapterHtmlSlimParser visitor(
       tmpHtmlPath, renderer, fontId, lineCompression, extraParagraphSpacing, viewportWidth, viewportHeight,
-      [this](std::unique_ptr<Page> page) { this->onPageComplete(std::move(page)); }, progressFn);
+      [this, &lut](std::unique_ptr<Page> page) { lut.emplace_back(this->onPageComplete(std::move(page))); },
+      progressFn);
   success = visitor.parseAndBuildPages();
 
   SD.remove(tmpHtmlPath.c_str());
   if (!success) {
     Serial.printf("[%lu] [SCT] Failed to parse XML and build pages\n", millis());
+    file.close();
+    SD.remove(filePath.c_str());
     return false;
   }
 
-  writeCacheMetadata(fontId, lineCompression, extraParagraphSpacing, viewportWidth, viewportHeight);
+  const auto lutOffset = file.position();
+  bool hasFailedLutRecords = false;
+  // Write LUT
+  for (const auto& pos : lut) {
+    if (pos == 0) {
+      hasFailedLutRecords = true;
+      break;
+    }
+    serialization::writePod(file, pos);
+  }
 
+  if (hasFailedLutRecords) {
+    Serial.printf("[%lu] [SCT] Failed to write LUT due to invalid page positions\n", millis());
+    file.close();
+    SD.remove(filePath.c_str());
+    return false;
+  }
+
+  // Go back and write LUT offset
+  file.seek(HEADER_SIZE - sizeof(size_t) - sizeof(pageCount));
+  serialization::writePod(file, pageCount);
+  serialization::writePod(file, lutOffset);
+  file.close();
   return true;
 }
 
-std::unique_ptr<Page> Section::loadPageFromSD() const {
-  const auto filePath = cachePath + "/page_" + std::to_string(currentPage) + ".bin";
-
-  File inputFile;
-  if (!FsHelpers::openFileForRead("SCT", filePath, inputFile)) {
+std::unique_ptr<Page> Section::loadPageFromSectionFile() {
+  if (!FsHelpers::openFileForRead("SCT", filePath, file)) {
     return nullptr;
   }
-  auto page = Page::deserialize(inputFile);
-  inputFile.close();
+
+  file.seek(HEADER_SIZE - sizeof(size_t));
+  size_t lutOffset;
+  serialization::readPod(file, lutOffset);
+  file.seek(lutOffset + sizeof(size_t) * currentPage);
+  size_t pagePos;
+  serialization::readPod(file, pagePos);
+  file.seek(pagePos);
+
+  auto page = Page::deserialize(file);
+  file.close();
   return page;
 }

lib/Epub/Epub/Section.h

@@ -11,11 +11,12 @@ class Section {
   std::shared_ptr<Epub> epub;
   const int spineIndex;
   GfxRenderer& renderer;
-  std::string cachePath;
+  std::string filePath;
+  File file;
 
-  void writeCacheMetadata(int fontId, float lineCompression, bool extraParagraphSpacing, int viewportWidth,
-                          int viewportHeight) const;
-  void onPageComplete(std::unique_ptr<Page> page);
+  void writeSectionFileHeader(int fontId, float lineCompression, bool extraParagraphSpacing, int viewportWidth,
+                              int viewportHeight);
+  size_t onPageComplete(std::unique_ptr<Page> page);
 
  public:
   int pageCount = 0;
@@ -25,14 +26,13 @@ class Section {
       : epub(epub),
         spineIndex(spineIndex),
         renderer(renderer),
-        cachePath(epub->getCachePath() + "/" + std::to_string(spineIndex)) {}
+        filePath(epub->getCachePath() + "/sections/" + std::to_string(spineIndex) + ".bin") {}
   ~Section() = default;
-  bool loadCacheMetadata(int fontId, float lineCompression, bool extraParagraphSpacing, int viewportWidth,
-                         int viewportHeight);
-  void setupCacheDir() const;
+  bool loadSectionFile(int fontId, float lineCompression, bool extraParagraphSpacing, int viewportWidth,
+                       int viewportHeight);
   bool clearCache() const;
-  bool persistPageDataToSD(int fontId, float lineCompression, bool extraParagraphSpacing, int viewportWidth,
-                           int viewportHeight, const std::function<void()>& progressSetupFn = nullptr,
-                           const std::function<void(int)>& progressFn = nullptr);
-  std::unique_ptr<Page> loadPageFromSD() const;
+  bool createSectionFile(int fontId, float lineCompression, bool extraParagraphSpacing, int viewportWidth,
+                         int viewportHeight, const std::function<void()>& progressSetupFn = nullptr,
+                         const std::function<void(int)>& progressFn = nullptr);
+  std::unique_ptr<Page> loadPageFromSectionFile();
 };

lib/Epub/Epub/blocks/TextBlock.cpp

@@ -24,34 +24,33 @@ void TextBlock::render(const GfxRenderer& renderer, const int fontId, const int
   }
 }
 
-void TextBlock::serialize(File& file) const {
-  // words
-  const uint32_t wc = words.size();
-  serialization::writePod(file, wc);
+bool TextBlock::serialize(File& file) const {
+  if (words.size() != wordXpos.size() || words.size() != wordStyles.size()) {
+    Serial.printf("[%lu] [TXB] Serialization failed: size mismatch (words=%u, xpos=%u, styles=%u)\n", millis(),
+                  words.size(), wordXpos.size(), wordStyles.size());
+    return false;
+  }
+
+  // Word data
+  serialization::writePod(file, static_cast<uint32_t>(words.size()));
   for (const auto& w : words) serialization::writeString(file, w);
-
-  // wordXpos
-  const uint32_t xc = wordXpos.size();
-  serialization::writePod(file, xc);
   for (auto x : wordXpos) serialization::writePod(file, x);
-
-  // wordStyles
-  const uint32_t sc = wordStyles.size();
-  serialization::writePod(file, sc);
   for (auto s : wordStyles) serialization::writePod(file, s);
 
-  // style
+  // Block style
   serialization::writePod(file, style);
+
+  return true;
 }
 
 std::unique_ptr<TextBlock> TextBlock::deserialize(File& file) {
-  uint32_t wc, xc, sc;
+  uint32_t wc;
   std::list<std::string> words;
   std::list<uint16_t> wordXpos;
   std::list<EpdFontStyle> wordStyles;
   BLOCK_STYLE style;
 
-  // words
+  // Word count
   serialization::readPod(file, wc);
 
   // Sanity check: prevent allocation of unreasonably large lists (max 10000 words per block)
@@ -60,27 +59,15 @@ std::unique_ptr<TextBlock> TextBlock::deserialize(File& file) {
     return nullptr;
   }
 
+  // Word data
   words.resize(wc);
+  wordXpos.resize(wc);
+  wordStyles.resize(wc);
   for (auto& w : words) serialization::readString(file, w);
-
-  // wordXpos
-  serialization::readPod(file, xc);
-  wordXpos.resize(xc);
   for (auto& x : wordXpos) serialization::readPod(file, x);
-
-  // wordStyles
-  serialization::readPod(file, sc);
-  wordStyles.resize(sc);
   for (auto& s : wordStyles) serialization::readPod(file, s);
 
-  // Validate data consistency: all three lists must have the same size
-  if (wc != xc || wc != sc) {
-    Serial.printf("[%lu] [TXB] Deserialization failed: size mismatch (words=%u, xpos=%u, styles=%u)\n", millis(), wc,
-                  xc, sc);
-    return nullptr;
-  }
-
-  // style
+  // Block style
   serialization::readPod(file, style);
 
   return std::unique_ptr<TextBlock>(new TextBlock(std::move(words), std::move(wordXpos), std::move(wordStyles), style));

lib/Epub/Epub/blocks/TextBlock.h

@@ -36,6 +36,6 @@ class TextBlock final : public Block {
   // given a renderer works out where to break the words into lines
   void render(const GfxRenderer& renderer, int fontId, int x, int y) const;
   BlockType getType() override { return TEXT_BLOCK; }
-  void serialize(File& file) const;
+  bool serialize(File& file) const;
   static std::unique_ptr<TextBlock> deserialize(File& file);
 };