Cut release 0.10.0

Use a JSON filter to avoid crashes when checking for updates (#141 )
## Summary * The JSON release data from Github contains the entire release description which can be very large * The 0.9.0 release was especially bad * Use a JSON filter to avoid deserializing anything but the necessary fields ## Additional Context * https://arduinojson.org/v7/how-to/deserialize-a-very-large-document/#filtering * Fixes https://github.com/daveallie/crosspoint-reader/issues/124
2026-04-29 10:26:52 -07:00 · 2025-12-29 02:30:27 +11:00 · 2025-12-29 02:29:41 +11:00 · 2025-12-29 02:00:42 +11:00 · 2025-12-29 01:56:05 +11:00 · 2025-12-28 23:49:51 +11:00
75 changed files with 5281 additions and 1003 deletions
@@ -12,12 +12,6 @@ jobs:
      - uses: actions/checkout@v6
        with:
          submodules: recursive
-      - uses: actions/cache@v5
-        with:
-          path: |
-            ~/.cache/pip
-            ~/.platformio/.cache
-          key: ${{ runner.os }}-pio
      - uses: actions/setup-python@v6
        with:
          python-version: '3.14'
@@ -59,14 +59,28 @@ bool EpdFont::hasPrintableChars(const char* string) const {

 const EpdGlyph* EpdFont::getGlyph(const uint32_t cp) const {
  const EpdUnicodeInterval* intervals = data->intervals;
-  for (int i = 0; i < data->intervalCount; i++) {
-    const EpdUnicodeInterval* interval = &intervals[i];
-    if (cp >= interval->first && cp <= interval->last) {
+  const int count = data->intervalCount;
+
+  if (count == 0) return nullptr;
+
+  // Binary search for O(log n) lookup instead of O(n)
+  // Critical for Korean fonts with many unicode intervals
+  int left = 0;
+  int right = count - 1;
+
+  while (left <= right) {
+    const int mid = left + (right - left) / 2;
+    const EpdUnicodeInterval* interval = &intervals[mid];
+
+    if (cp < interval->first) {
+      right = mid - 1;
+    } else if (cp > interval->last) {
+      left = mid + 1;
+    } else {
+      // Found: cp >= interval->first && cp <= interval->last
      return &data->glyph[interval->offset + (cp - interval->first)];
    }
-    if (cp < interval->first) {
-      return nullptr;
-    }
  }
+
  return nullptr;
 }
@@ -1,13 +1,11 @@
 #include "Epub.h"

+#include <FsHelpers.h>
 #include <HardwareSerial.h>
 #include <JpegToBmpConverter.h>
 #include <SD.h>
 #include <ZipFile.h>

-#include <map>
-
-#include "Epub/FsHelpers.h"
 #include "Epub/parsers/ContainerParser.h"
 #include "Epub/parsers/ContentOpfParser.h"
 #include "Epub/parsers/TocNcxParser.h"
@@ -44,7 +42,15 @@ bool Epub::findContentOpfFile(std::string* contentOpfFile) const {
  return true;
 }

-bool Epub::parseContentOpf(const std::string& contentOpfFilePath) {
+bool Epub::parseContentOpf(BookMetadataCache::BookMetadata& bookMetadata) {
+  std::string contentOpfFilePath;
+  if (!findContentOpfFile(&contentOpfFilePath)) {
+    Serial.printf("[%lu] [EBP] Could not find content.opf in zip\n", millis());
+    return false;
+  }
+
+  contentBasePath = contentOpfFilePath.substr(0, contentOpfFilePath.find_last_of('/') + 1);
+
  Serial.printf("[%lu] [EBP] Parsing content.opf: %s\n", millis(), contentOpfFilePath.c_str());

  size_t contentOpfSize;
@@ -53,7 +59,9 @@ bool Epub::parseContentOpf(const std::string& contentOpfFilePath) {
    return false;
  }

-  ContentOpfParser opfParser(getBasePath(), contentOpfSize);
+  ContentOpfParser opfParser(getCachePath(), getBasePath(), contentOpfSize, bookMetadataCache.get());
+  Serial.printf("[%lu] [MEM] Free: %d bytes, Total: %d bytes, Min Free: %d bytes\n", millis(), ESP.getFreeHeap(),
+                ESP.getHeapSize(), ESP.getMinFreeHeap());

  if (!opfParser.setup()) {
    Serial.printf("[%lu] [EBP] Could not setup content.opf parser\n", millis());
@@ -66,26 +74,20 @@ bool Epub::parseContentOpf(const std::string& contentOpfFilePath) {
  }

  // Grab data from opfParser into epub
-  title = opfParser.title;
-  if (!opfParser.coverItemId.empty() && opfParser.items.count(opfParser.coverItemId) > 0) {
-    coverImageItem = opfParser.items.at(opfParser.coverItemId);
-  }
+  bookMetadata.title = opfParser.title;
+  // TODO: Parse author
+  bookMetadata.author = "";
+  bookMetadata.coverItemHref = opfParser.coverItemHref;

  if (!opfParser.tocNcxPath.empty()) {
    tocNcxItem = opfParser.tocNcxPath;
  }

-  for (auto& spineRef : opfParser.spineRefs) {
-    if (opfParser.items.count(spineRef)) {
-      spine.emplace_back(spineRef, opfParser.items.at(spineRef));
-    }
-  }
-
  Serial.printf("[%lu] [EBP] Successfully parsed content.opf\n", millis());
  return true;
 }

-bool Epub::parseTocNcxFile() {
+bool Epub::parseTocNcxFile() const {
  // the ncx file should have been specified in the content.opf file
  if (tocNcxItem.empty()) {
    Serial.printf("[%lu] [EBP] No ncx file specified\n", millis());
@@ -95,13 +97,18 @@ bool Epub::parseTocNcxFile() {
  Serial.printf("[%lu] [EBP] Parsing toc ncx file: %s\n", millis(), tocNcxItem.c_str());

  const auto tmpNcxPath = getCachePath() + "/toc.ncx";
-  File tempNcxFile = SD.open(tmpNcxPath.c_str(), FILE_WRITE);
+  File tempNcxFile;
+  if (!FsHelpers::openFileForWrite("EBP", tmpNcxPath, tempNcxFile)) {
+    return false;
+  }
  readItemContentsToStream(tocNcxItem, tempNcxFile, 1024);
  tempNcxFile.close();
-  tempNcxFile = SD.open(tmpNcxPath.c_str(), FILE_READ);
+  if (!FsHelpers::openFileForRead("EBP", tmpNcxPath, tempNcxFile)) {
+    return false;
+  }
  const auto ncxSize = tempNcxFile.size();

-  TocNcxParser ncxParser(contentBasePath, ncxSize);
+  TocNcxParser ncxParser(contentBasePath, ncxSize, bookMetadataCache.get());

  if (!ncxParser.setup()) {
    Serial.printf("[%lu] [EBP] Could not setup toc ncx parser\n", millis());
@@ -130,9 +137,7 @@ bool Epub::parseTocNcxFile() {
  tempNcxFile.close();
  SD.remove(tmpNcxPath.c_str());

-  this->toc = std::move(ncxParser.toc);
-
-  Serial.printf("[%lu] [EBP] Parsed %d TOC items\n", millis(), this->toc.size());
+  Serial.printf("[%lu] [EBP] Parsed TOC items\n", millis());
  return true;
 }

@@ -140,48 +145,79 @@ bool Epub::parseTocNcxFile() {
 bool Epub::load() {
  Serial.printf("[%lu] [EBP] Loading ePub: %s\n", millis(), filepath.c_str());

-  std::string contentOpfFilePath;
-  if (!findContentOpfFile(&contentOpfFilePath)) {
-    Serial.printf("[%lu] [EBP] Could not find content.opf in zip\n", millis());
+  // Initialize spine/TOC cache
+  bookMetadataCache.reset(new BookMetadataCache(cachePath));
+
+  // Try to load existing cache first
+  if (bookMetadataCache->load()) {
+    Serial.printf("[%lu] [EBP] Loaded ePub: %s\n", millis(), filepath.c_str());
+    return true;
+  }
+
+  // Cache doesn't exist or is invalid, build it
+  Serial.printf("[%lu] [EBP] Cache not found, building spine/TOC cache\n", millis());
+  setupCacheDir();
+
+  // Begin building cache - stream entries to disk immediately
+  if (!bookMetadataCache->beginWrite()) {
+    Serial.printf("[%lu] [EBP] Could not begin writing cache\n", millis());
    return false;
  }

-  Serial.printf("[%lu] [EBP] Found content.opf at: %s\n", millis(), contentOpfFilePath.c_str());
-
-  contentBasePath = contentOpfFilePath.substr(0, contentOpfFilePath.find_last_of('/') + 1);
-
-  if (!parseContentOpf(contentOpfFilePath)) {
+  // OPF Pass
+  BookMetadataCache::BookMetadata bookMetadata;
+  if (!bookMetadataCache->beginContentOpfPass()) {
+    Serial.printf("[%lu] [EBP] Could not begin writing content.opf pass\n", millis());
+    return false;
+  }
+  if (!parseContentOpf(bookMetadata)) {
    Serial.printf("[%lu] [EBP] Could not parse content.opf\n", millis());
    return false;
  }
+  if (!bookMetadataCache->endContentOpfPass()) {
+    Serial.printf("[%lu] [EBP] Could not end writing content.opf pass\n", millis());
+    return false;
+  }

+  // TOC Pass
+  if (!bookMetadataCache->beginTocPass()) {
+    Serial.printf("[%lu] [EBP] Could not begin writing toc pass\n", millis());
+    return false;
+  }
  if (!parseTocNcxFile()) {
    Serial.printf("[%lu] [EBP] Could not parse toc\n", millis());
    return false;
  }
-
-  initializeSpineItemSizes();
-  Serial.printf("[%lu] [EBP] Loaded ePub: %s\n", millis(), filepath.c_str());
-
-  return true;
-}
-
-void Epub::initializeSpineItemSizes() {
-  Serial.printf("[%lu] [EBP] Calculating book size\n", millis());
-
-  const size_t spineItemsCount = getSpineItemsCount();
-  size_t cumSpineItemSize = 0;
-  const ZipFile zip("/sd" + filepath);
-
-  for (size_t i = 0; i < spineItemsCount; i++) {
-    std::string spineItem = getSpineItem(i);
-    size_t s = 0;
-    getItemSize(zip, spineItem, &s);
-    cumSpineItemSize += s;
-    cumulativeSpineItemSize.emplace_back(cumSpineItemSize);
+  if (!bookMetadataCache->endTocPass()) {
+    Serial.printf("[%lu] [EBP] Could not end writing toc pass\n", millis());
+    return false;
  }

-  Serial.printf("[%lu] [EBP] Book size: %lu\n", millis(), cumSpineItemSize);
+  // Close the cache files
+  if (!bookMetadataCache->endWrite()) {
+    Serial.printf("[%lu] [EBP] Could not end writing cache\n", millis());
+    return false;
+  }
+
+  // Build final book.bin
+  if (!bookMetadataCache->buildBookBin(filepath, bookMetadata)) {
+    Serial.printf("[%lu] [EBP] Could not update mappings and sizes\n", millis());
+    return false;
+  }
+
+  if (!bookMetadataCache->cleanupTmpFiles()) {
+    Serial.printf("[%lu] [EBP] Could not cleanup tmp files - ignoring\n", millis());
+  }
+
+  // Reload the cache from disk so it's in the correct state
+  bookMetadataCache.reset(new BookMetadataCache(cachePath));
+  if (!bookMetadataCache->load()) {
+    Serial.printf("[%lu] [EBP] Failed to reload cache after writing\n", millis());
+    return false;
+  }
+
+  Serial.printf("[%lu] [EBP] Loaded ePub: %s\n", millis(), filepath.c_str());
+  return true;
 }

 bool Epub::clearCache() const {
@@ -217,7 +253,14 @@ const std::string& Epub::getCachePath() const { return cachePath; }

 const std::string& Epub::getPath() const { return filepath; }

-const std::string& Epub::getTitle() const { return title; }
+const std::string& Epub::getTitle() const {
+  static std::string blank;
+  if (!bookMetadataCache || !bookMetadataCache->isLoaded()) {
+    return blank;
+  }
+
+  return bookMetadataCache->coreMetadata.title;
+}

 std::string Epub::getCoverBmpPath() const { return cachePath + "/cover.bmp"; }

@@ -227,24 +270,42 @@ bool Epub::generateCoverBmp() const {
    return true;
  }

-  if (coverImageItem.empty()) {
+  if (!bookMetadataCache || !bookMetadataCache->isLoaded()) {
+    Serial.printf("[%lu] [EBP] Cannot generate cover BMP, cache not loaded\n", millis());
+    return false;
+  }
+
+  const auto coverImageHref = bookMetadataCache->coreMetadata.coverItemHref;
+  if (coverImageHref.empty()) {
    Serial.printf("[%lu] [EBP] No known cover image\n", millis());
    return false;
  }

-  if (coverImageItem.substr(coverImageItem.length() - 4) == ".jpg" ||
-      coverImageItem.substr(coverImageItem.length() - 5) == ".jpeg") {
+  if (coverImageHref.substr(coverImageHref.length() - 4) == ".jpg" ||
+      coverImageHref.substr(coverImageHref.length() - 5) == ".jpeg") {
    Serial.printf("[%lu] [EBP] Generating BMP from JPG cover image\n", millis());
-    File coverJpg = SD.open((getCachePath() + "/.cover.jpg").c_str(), FILE_WRITE, true);
-    readItemContentsToStream(coverImageItem, coverJpg, 1024);
+    const auto coverJpgTempPath = getCachePath() + "/.cover.jpg";
+
+    File coverJpg;
+    if (!FsHelpers::openFileForWrite("EBP", coverJpgTempPath, coverJpg)) {
+      return false;
+    }
+    readItemContentsToStream(coverImageHref, coverJpg, 1024);
    coverJpg.close();

-    coverJpg = SD.open((getCachePath() + "/.cover.jpg").c_str(), FILE_READ);
-    File coverBmp = SD.open(getCoverBmpPath().c_str(), FILE_WRITE, true);
+    if (!FsHelpers::openFileForRead("EBP", coverJpgTempPath, coverJpg)) {
+      return false;
+    }
+
+    File coverBmp;
+    if (!FsHelpers::openFileForWrite("EBP", getCoverBmpPath(), coverBmp)) {
+      coverJpg.close();
+      return false;
+    }
    const bool success = JpegToBmpConverter::jpegFileToBmpStream(coverJpg, coverBmp);
    coverJpg.close();
    coverBmp.close();
-    SD.remove((getCachePath() + "/.cover.jpg").c_str());
+    SD.remove(coverJpgTempPath.c_str());

    if (!success) {
      Serial.printf("[%lu] [EBP] Failed to generate BMP from JPG cover image\n", millis());
@@ -259,45 +320,9 @@ bool Epub::generateCoverBmp() const {
  return false;
 }

-std::string normalisePath(const std::string& path) {
-  std::vector<std::string> components;
-  std::string component;
-
-  for (const auto c : path) {
-    if (c == '/') {
-      if (!component.empty()) {
-        if (component == "..") {
-          if (!components.empty()) {
-            components.pop_back();
-          }
-        } else {
-          components.push_back(component);
-        }
-        component.clear();
-      }
-    } else {
-      component += c;
-    }
-  }
-
-  if (!component.empty()) {
-    components.push_back(component);
-  }
-
-  std::string result;
-  for (const auto& c : components) {
-    if (!result.empty()) {
-      result += "/";
-    }
-    result += c;
-  }
-
-  return result;
-}
-
-uint8_t* Epub::readItemContentsToBytes(const std::string& itemHref, size_t* size, bool trailingNullByte) const {
+uint8_t* Epub::readItemContentsToBytes(const std::string& itemHref, size_t* size, const bool trailingNullByte) const {
  const ZipFile zip("/sd" + filepath);
-  const std::string path = normalisePath(itemHref);
+  const std::string path = FsHelpers::normalisePath(itemHref);

  const auto content = zip.readFileToMemory(path.c_str(), size, trailingNullByte);
  if (!content) {
@@ -310,7 +335,7 @@ uint8_t* Epub::readItemContentsToBytes(const std::string& itemHref, size_t* size

 bool Epub::readItemContentsToStream(const std::string& itemHref, Print& out, const size_t chunkSize) const {
  const ZipFile zip("/sd" + filepath);
-  const std::string path = normalisePath(itemHref);
+  const std::string path = FsHelpers::normalisePath(itemHref);

  return zip.readFileToStream(path.c_str(), out, chunkSize);
 }
@@ -321,103 +346,93 @@ bool Epub::getItemSize(const std::string& itemHref, size_t* size) const {
 }

 bool Epub::getItemSize(const ZipFile& zip, const std::string& itemHref, size_t* size) {
-  const std::string path = normalisePath(itemHref);
+  const std::string path = FsHelpers::normalisePath(itemHref);
  return zip.getInflatedFileSize(path.c_str(), size);
 }

-int Epub::getSpineItemsCount() const { return spine.size(); }
-
-size_t Epub::getCumulativeSpineItemSize(const int spineIndex) const {
-  if (spineIndex < 0 || spineIndex >= static_cast<int>(cumulativeSpineItemSize.size())) {
-    Serial.printf("[%lu] [EBP] getCumulativeSpineItemSize index:%d is out of range\n", millis(), spineIndex);
+int Epub::getSpineItemsCount() const {
+  if (!bookMetadataCache || !bookMetadataCache->isLoaded()) {
    return 0;
  }
-  return cumulativeSpineItemSize.at(spineIndex);
+  return bookMetadataCache->getSpineCount();
 }

-std::string& Epub::getSpineItem(const int spineIndex) {
-  static std::string emptyString;
-  if (spine.empty()) {
-    Serial.printf("[%lu] [EBP] getSpineItem called but spine is empty\n", millis());
-    return emptyString;
+size_t Epub::getCumulativeSpineItemSize(const int spineIndex) const { return getSpineItem(spineIndex).cumulativeSize; }
+
+BookMetadataCache::SpineEntry Epub::getSpineItem(const int spineIndex) const {
+  if (!bookMetadataCache || !bookMetadataCache->isLoaded()) {
+    Serial.printf("[%lu] [EBP] getSpineItem called but cache not loaded\n", millis());
+    return {};
  }
-  if (spineIndex < 0 || spineIndex >= static_cast<int>(spine.size())) {
+
+  if (spineIndex < 0 || spineIndex >= bookMetadataCache->getSpineCount()) {
    Serial.printf("[%lu] [EBP] getSpineItem index:%d is out of range\n", millis(), spineIndex);
-    return spine.at(0).second;
+    return bookMetadataCache->getSpineEntry(0);
  }

-  return spine.at(spineIndex).second;
+  return bookMetadataCache->getSpineEntry(spineIndex);
 }

-EpubTocEntry& Epub::getTocItem(const int tocTndex) {
-  static EpubTocEntry emptyEntry = {};
-  if (toc.empty()) {
-    Serial.printf("[%lu] [EBP] getTocItem called but toc is empty\n", millis());
-    return emptyEntry;
-  }
-  if (tocTndex < 0 || tocTndex >= static_cast<int>(toc.size())) {
-    Serial.printf("[%lu] [EBP] getTocItem index:%d is out of range\n", millis(), tocTndex);
-    return toc.at(0);
+BookMetadataCache::TocEntry Epub::getTocItem(const int tocIndex) const {
+  if (!bookMetadataCache || !bookMetadataCache->isLoaded()) {
+    Serial.printf("[%lu] [EBP] getTocItem called but cache not loaded\n", millis());
+    return {};
  }

-  return toc.at(tocTndex);
+  if (tocIndex < 0 || tocIndex >= bookMetadataCache->getTocCount()) {
+    Serial.printf("[%lu] [EBP] getTocItem index:%d is out of range\n", millis(), tocIndex);
+    return {};
+  }
+
+  return bookMetadataCache->getTocEntry(tocIndex);
 }

-int Epub::getTocItemsCount() const { return toc.size(); }
+int Epub::getTocItemsCount() const {
+  if (!bookMetadataCache || !bookMetadataCache->isLoaded()) {
+    return 0;
+  }
+
+  return bookMetadataCache->getTocCount();
+}

 // work out the section index for a toc index
 int Epub::getSpineIndexForTocIndex(const int tocIndex) const {
-  if (tocIndex < 0 || tocIndex >= toc.size()) {
+  if (!bookMetadataCache || !bookMetadataCache->isLoaded()) {
+    Serial.printf("[%lu] [EBP] getSpineIndexForTocIndex called but cache not loaded\n", millis());
+    return 0;
+  }
+
+  if (tocIndex < 0 || tocIndex >= bookMetadataCache->getTocCount()) {
    Serial.printf("[%lu] [EBP] getSpineIndexForTocIndex: tocIndex %d out of range\n", millis(), tocIndex);
    return 0;
  }

-  // the toc entry should have an href that matches the spine item
-  // so we can find the spine index by looking for the href
-  for (int i = 0; i < spine.size(); i++) {
-    if (spine[i].second == toc[tocIndex].href) {
-      return i;
-    }
+  const int spineIndex = bookMetadataCache->getTocEntry(tocIndex).spineIndex;
+  if (spineIndex < 0) {
+    Serial.printf("[%lu] [EBP] Section not found for TOC index %d\n", millis(), tocIndex);
+    return 0;
  }

-  Serial.printf("[%lu] [EBP] Section not found\n", millis());
-  // not found - default to the start of the book
-  return 0;
+  return spineIndex;
 }

-int Epub::getTocIndexForSpineIndex(const int spineIndex) const {
-  if (spineIndex < 0 || spineIndex >= spine.size()) {
-    Serial.printf("[%lu] [EBP] getTocIndexForSpineIndex: spineIndex %d out of range\n", millis(), spineIndex);
-    return -1;
-  }
-
-  // the toc entry should have an href that matches the spine item
-  // so we can find the toc index by looking for the href
-  for (int i = 0; i < toc.size(); i++) {
-    if (toc[i].href == spine[spineIndex].second) {
-      return i;
-    }
-  }
-
-  Serial.printf("[%lu] [EBP] TOC item not found\n", millis());
-  return -1;
-}
+int Epub::getTocIndexForSpineIndex(const int spineIndex) const { return getSpineItem(spineIndex).tocIndex; }

 size_t Epub::getBookSize() const {
-  if (spine.empty()) {
+  if (!bookMetadataCache || !bookMetadataCache->isLoaded() || bookMetadataCache->getSpineCount() == 0) {
    return 0;
  }
  return getCumulativeSpineItemSize(getSpineItemsCount() - 1);
 }

 // Calculate progress in book
-uint8_t Epub::calculateProgress(const int currentSpineIndex, const float currentSpineRead) {
-  size_t bookSize = getBookSize();
+uint8_t Epub::calculateProgress(const int currentSpineIndex, const float currentSpineRead) const {
+  const size_t bookSize = getBookSize();
  if (bookSize == 0) {
    return 0;
  }
-  size_t prevChapterSize = (currentSpineIndex >= 1) ? getCumulativeSpineItemSize(currentSpineIndex - 1) : 0;
-  size_t curChapterSize = getCumulativeSpineItemSize(currentSpineIndex) - prevChapterSize;
-  size_t sectionProgSize = currentSpineRead * curChapterSize;
+  const size_t prevChapterSize = (currentSpineIndex >= 1) ? getCumulativeSpineItemSize(currentSpineIndex - 1) : 0;
+  const size_t curChapterSize = getCumulativeSpineItemSize(currentSpineIndex) - prevChapterSize;
+  const size_t sectionProgSize = currentSpineRead * curChapterSize;
  return round(static_cast<float>(prevChapterSize + sectionProgSize) / bookSize * 100.0);
 }
@@ -1,38 +1,29 @@
 #pragma once
-#include <Print.h>

+#include <memory>
 #include <string>
 #include <unordered_map>
 #include <vector>

-#include "Epub/EpubTocEntry.h"
+#include "Epub/BookMetadataCache.h"

 class ZipFile;

 class Epub {
-  // the title read from the EPUB meta data
-  std::string title;
-  // the cover image
-  std::string coverImageItem;
  // the ncx file
  std::string tocNcxItem;
  // where is the EPUBfile?
  std::string filepath;
-  // the spine of the EPUB file
-  std::vector<std::pair<std::string, std::string>> spine;
-  // the file size of the spine items (proxy to book progress)
-  std::vector<size_t> cumulativeSpineItemSize;
-  // the toc of the EPUB file
-  std::vector<EpubTocEntry> toc;
  // the base path for items in the EPUB file
  std::string contentBasePath;
  // Uniq cache key based on filepath
  std::string cachePath;
+  // Spine and TOC cache
+  std::unique_ptr<BookMetadataCache> bookMetadataCache;

  bool findContentOpfFile(std::string* contentOpfFile) const;
-  bool parseContentOpf(const std::string& contentOpfFilePath);
-  bool parseTocNcxFile();
-  void initializeSpineItemSizes();
+  bool parseContentOpf(BookMetadataCache::BookMetadata& bookMetadata);
+  bool parseTocNcxFile() const;
  static bool getItemSize(const ZipFile& zip, const std::string& itemHref, size_t* size);

 public:
@@ -54,14 +45,14 @@ class Epub {
                                   bool trailingNullByte = false) const;
  bool readItemContentsToStream(const std::string& itemHref, Print& out, size_t chunkSize) const;
  bool getItemSize(const std::string& itemHref, size_t* size) const;
-  std::string& getSpineItem(int spineIndex);
+  BookMetadataCache::SpineEntry getSpineItem(int spineIndex) const;
+  BookMetadataCache::TocEntry getTocItem(int tocIndex) const;
  int getSpineItemsCount() const;
-  size_t getCumulativeSpineItemSize(const int spineIndex) const;
-  EpubTocEntry& getTocItem(int tocIndex);
  int getTocItemsCount() const;
  int getSpineIndexForTocIndex(int tocIndex) const;
  int getTocIndexForSpineIndex(int spineIndex) const;
+  size_t getCumulativeSpineItemSize(int spineIndex) const;

  size_t getBookSize() const;
-  uint8_t calculateProgress(const int currentSpineIndex, const float currentSpineRead);
+  uint8_t calculateProgress(const int currentSpineIndex, const float currentSpineRead) const;
 };
@@ -0,0 +1,326 @@
+#include "BookMetadataCache.h"
+
+#include <HardwareSerial.h>
+#include <SD.h>
+#include <Serialization.h>
+#include <ZipFile.h>
+
+#include <vector>
+
+#include "FsHelpers.h"
+
+namespace {
+constexpr uint8_t BOOK_CACHE_VERSION = 1;
+constexpr char bookBinFile[] = "/book.bin";
+constexpr char tmpSpineBinFile[] = "/spine.bin.tmp";
+constexpr char tmpTocBinFile[] = "/toc.bin.tmp";
+}  // namespace
+
+/* ============= WRITING / BUILDING FUNCTIONS ================ */
+
+bool BookMetadataCache::beginWrite() {
+  buildMode = true;
+  spineCount = 0;
+  tocCount = 0;
+  Serial.printf("[%lu] [BMC] Entering write mode\n", millis());
+  return true;
+}
+
+bool BookMetadataCache::beginContentOpfPass() {
+  Serial.printf("[%lu] [BMC] Beginning content opf pass\n", millis());
+
+  // Open spine file for writing
+  return FsHelpers::openFileForWrite("BMC", cachePath + tmpSpineBinFile, spineFile);
+}
+
+bool BookMetadataCache::endContentOpfPass() {
+  spineFile.close();
+  return true;
+}
+
+bool BookMetadataCache::beginTocPass() {
+  Serial.printf("[%lu] [BMC] Beginning toc pass\n", millis());
+
+  // Open spine file for reading
+  if (!FsHelpers::openFileForRead("BMC", cachePath + tmpSpineBinFile, spineFile)) {
+    return false;
+  }
+  if (!FsHelpers::openFileForWrite("BMC", cachePath + tmpTocBinFile, tocFile)) {
+    spineFile.close();
+    return false;
+  }
+  return true;
+}
+
+bool BookMetadataCache::endTocPass() {
+  tocFile.close();
+  spineFile.close();
+  return true;
+}
+
+bool BookMetadataCache::endWrite() {
+  if (!buildMode) {
+    Serial.printf("[%lu] [BMC] endWrite called but not in build mode\n", millis());
+    return false;
+  }
+
+  buildMode = false;
+  Serial.printf("[%lu] [BMC] Wrote %d spine, %d TOC entries\n", millis(), spineCount, tocCount);
+  return true;
+}
+
+bool BookMetadataCache::buildBookBin(const std::string& epubPath, const BookMetadata& metadata) {
+  // Open all three files, writing to meta, reading from spine and toc
+  if (!FsHelpers::openFileForWrite("BMC", cachePath + bookBinFile, bookFile)) {
+    return false;
+  }
+
+  if (!FsHelpers::openFileForRead("BMC", cachePath + tmpSpineBinFile, spineFile)) {
+    bookFile.close();
+    return false;
+  }
+
+  if (!FsHelpers::openFileForRead("BMC", cachePath + tmpTocBinFile, tocFile)) {
+    bookFile.close();
+    spineFile.close();
+    return false;
+  }
+
+  constexpr size_t headerASize =
+      sizeof(BOOK_CACHE_VERSION) + /* LUT Offset */ sizeof(size_t) + sizeof(spineCount) + sizeof(tocCount);
+  const size_t metadataSize =
+      metadata.title.size() + metadata.author.size() + metadata.coverItemHref.size() + sizeof(uint32_t) * 3;
+  const size_t lutSize = sizeof(size_t) * spineCount + sizeof(size_t) * tocCount;
+  const size_t lutOffset = headerASize + metadataSize;
+
+  // Header A
+  serialization::writePod(bookFile, BOOK_CACHE_VERSION);
+  serialization::writePod(bookFile, lutOffset);
+  serialization::writePod(bookFile, spineCount);
+  serialization::writePod(bookFile, tocCount);
+  // Metadata
+  serialization::writeString(bookFile, metadata.title);
+  serialization::writeString(bookFile, metadata.author);
+  serialization::writeString(bookFile, metadata.coverItemHref);
+
+  // Loop through spine entries, writing LUT positions
+  spineFile.seek(0);
+  for (int i = 0; i < spineCount; i++) {
+    auto pos = spineFile.position();
+    auto spineEntry = readSpineEntry(spineFile);
+    serialization::writePod(bookFile, pos + lutOffset + lutSize);
+  }
+
+  // Loop through toc entries, writing LUT positions
+  tocFile.seek(0);
+  for (int i = 0; i < tocCount; i++) {
+    auto pos = tocFile.position();
+    auto tocEntry = readTocEntry(tocFile);
+    serialization::writePod(bookFile, pos + lutOffset + lutSize + spineFile.position());
+  }
+
+  // LUTs complete
+  // Loop through spines from spine file matching up TOC indexes, calculating cumulative size and writing to book.bin
+
+  const ZipFile zip("/sd" + epubPath);
+  size_t cumSize = 0;
+  spineFile.seek(0);
+  for (int i = 0; i < spineCount; i++) {
+    auto spineEntry = readSpineEntry(spineFile);
+
+    tocFile.seek(0);
+    for (int j = 0; j < tocCount; j++) {
+      auto tocEntry = readTocEntry(tocFile);
+      if (tocEntry.spineIndex == i) {
+        spineEntry.tocIndex = j;
+        break;
+      }
+    }
+
+    // Not a huge deal if we don't fine a TOC entry for the spine entry, this is expected behaviour for EPUBs
+    // Logging here is for debugging
+    if (spineEntry.tocIndex == -1) {
+      Serial.printf("[%lu] [BMC] Warning: Could not find TOC entry for spine item %d: %s\n", millis(), i,
+                    spineEntry.href.c_str());
+    }
+
+    // Calculate size for cumulative size
+    size_t itemSize = 0;
+    const std::string path = FsHelpers::normalisePath(spineEntry.href);
+    if (zip.getInflatedFileSize(path.c_str(), &itemSize)) {
+      cumSize += itemSize;
+      spineEntry.cumulativeSize = cumSize;
+    } else {
+      Serial.printf("[%lu] [BMC] Warning: Could not get size for spine item: %s\n", millis(), path.c_str());
+    }
+
+    // Write out spine data to book.bin
+    writeSpineEntry(bookFile, spineEntry);
+  }
+
+  // Loop through toc entries from toc file writing to book.bin
+  tocFile.seek(0);
+  for (int i = 0; i < tocCount; i++) {
+    auto tocEntry = readTocEntry(tocFile);
+    writeTocEntry(bookFile, tocEntry);
+  }
+
+  bookFile.close();
+  spineFile.close();
+  tocFile.close();
+
+  Serial.printf("[%lu] [BMC] Successfully built book.bin\n", millis());
+  return true;
+}
+
+bool BookMetadataCache::cleanupTmpFiles() const {
+  if (SD.exists((cachePath + tmpSpineBinFile).c_str())) {
+    SD.remove((cachePath + tmpSpineBinFile).c_str());
+  }
+  if (SD.exists((cachePath + tmpTocBinFile).c_str())) {
+    SD.remove((cachePath + tmpTocBinFile).c_str());
+  }
+  return true;
+}
+
+size_t BookMetadataCache::writeSpineEntry(File& file, const SpineEntry& entry) const {
+  const auto pos = file.position();
+  serialization::writeString(file, entry.href);
+  serialization::writePod(file, entry.cumulativeSize);
+  serialization::writePod(file, entry.tocIndex);
+  return pos;
+}
+
+size_t BookMetadataCache::writeTocEntry(File& file, const TocEntry& entry) const {
+  const auto pos = file.position();
+  serialization::writeString(file, entry.title);
+  serialization::writeString(file, entry.href);
+  serialization::writeString(file, entry.anchor);
+  serialization::writePod(file, entry.level);
+  serialization::writePod(file, entry.spineIndex);
+  return pos;
+}
+
+// Note: for the LUT to be accurate, this **MUST** be called for all spine items before `addTocEntry` is ever called
+// this is because in this function we're marking positions of the items
+void BookMetadataCache::createSpineEntry(const std::string& href) {
+  if (!buildMode || !spineFile) {
+    Serial.printf("[%lu] [BMC] createSpineEntry called but not in build mode\n", millis());
+    return;
+  }
+
+  const SpineEntry entry(href, 0, -1);
+  writeSpineEntry(spineFile, entry);
+  spineCount++;
+}
+
+void BookMetadataCache::createTocEntry(const std::string& title, const std::string& href, const std::string& anchor,
+                                       const uint8_t level) {
+  if (!buildMode || !tocFile || !spineFile) {
+    Serial.printf("[%lu] [BMC] createTocEntry called but not in build mode\n", millis());
+    return;
+  }
+
+  int spineIndex = -1;
+  // find spine index
+  spineFile.seek(0);
+  for (int i = 0; i < spineCount; i++) {
+    auto spineEntry = readSpineEntry(spineFile);
+    if (spineEntry.href == href) {
+      spineIndex = i;
+      break;
+    }
+  }
+
+  if (spineIndex == -1) {
+    Serial.printf("[%lu] [BMC] addTocEntry: Could not find spine item for TOC href %s\n", millis(), href.c_str());
+  }
+
+  const TocEntry entry(title, href, anchor, level, spineIndex);
+  writeTocEntry(tocFile, entry);
+  tocCount++;
+}
+
+/* ============= READING / LOADING FUNCTIONS ================ */
+
+bool BookMetadataCache::load() {
+  if (!FsHelpers::openFileForRead("BMC", cachePath + bookBinFile, bookFile)) {
+    return false;
+  }
+
+  uint8_t version;
+  serialization::readPod(bookFile, version);
+  if (version != BOOK_CACHE_VERSION) {
+    Serial.printf("[%lu] [BMC] Cache version mismatch: expected %d, got %d\n", millis(), BOOK_CACHE_VERSION, version);
+    bookFile.close();
+    return false;
+  }
+
+  serialization::readPod(bookFile, lutOffset);
+  serialization::readPod(bookFile, spineCount);
+  serialization::readPod(bookFile, tocCount);
+
+  serialization::readString(bookFile, coreMetadata.title);
+  serialization::readString(bookFile, coreMetadata.author);
+  serialization::readString(bookFile, coreMetadata.coverItemHref);
+
+  loaded = true;
+  Serial.printf("[%lu] [BMC] Loaded cache data: %d spine, %d TOC entries\n", millis(), spineCount, tocCount);
+  return true;
+}
+
+BookMetadataCache::SpineEntry BookMetadataCache::getSpineEntry(const int index) {
+  if (!loaded) {
+    Serial.printf("[%lu] [BMC] getSpineEntry called but cache not loaded\n", millis());
+    return {};
+  }
+
+  if (index < 0 || index >= static_cast<int>(spineCount)) {
+    Serial.printf("[%lu] [BMC] getSpineEntry index %d out of range\n", millis(), index);
+    return {};
+  }
+
+  // Seek to spine LUT item, read from LUT and get out data
+  bookFile.seek(lutOffset + sizeof(size_t) * index);
+  size_t spineEntryPos;
+  serialization::readPod(bookFile, spineEntryPos);
+  bookFile.seek(spineEntryPos);
+  return readSpineEntry(bookFile);
+}
+
+BookMetadataCache::TocEntry BookMetadataCache::getTocEntry(const int index) {
+  if (!loaded) {
+    Serial.printf("[%lu] [BMC] getTocEntry called but cache not loaded\n", millis());
+    return {};
+  }
+
+  if (index < 0 || index >= static_cast<int>(tocCount)) {
+    Serial.printf("[%lu] [BMC] getTocEntry index %d out of range\n", millis(), index);
+    return {};
+  }
+
+  // Seek to TOC LUT item, read from LUT and get out data
+  bookFile.seek(lutOffset + sizeof(size_t) * spineCount + sizeof(size_t) * index);
+  size_t tocEntryPos;
+  serialization::readPod(bookFile, tocEntryPos);
+  bookFile.seek(tocEntryPos);
+  return readTocEntry(bookFile);
+}
+
+BookMetadataCache::SpineEntry BookMetadataCache::readSpineEntry(File& file) const {
+  SpineEntry entry;
+  serialization::readString(file, entry.href);
+  serialization::readPod(file, entry.cumulativeSize);
+  serialization::readPod(file, entry.tocIndex);
+  return entry;
+}
+
+BookMetadataCache::TocEntry BookMetadataCache::readTocEntry(File& file) const {
+  TocEntry entry;
+  serialization::readString(file, entry.title);
+  serialization::readString(file, entry.href);
+  serialization::readString(file, entry.anchor);
+  serialization::readPod(file, entry.level);
+  serialization::readPod(file, entry.spineIndex);
+  return entry;
+}
@@ -0,0 +1,87 @@
+#pragma once
+
+#include <SD.h>
+
+#include <string>
+
+class BookMetadataCache {
+ public:
+  struct BookMetadata {
+    std::string title;
+    std::string author;
+    std::string coverItemHref;
+  };
+
+  struct SpineEntry {
+    std::string href;
+    size_t cumulativeSize;
+    int16_t tocIndex;
+
+    SpineEntry() : cumulativeSize(0), tocIndex(-1) {}
+    SpineEntry(std::string href, const size_t cumulativeSize, const int16_t tocIndex)
+        : href(std::move(href)), cumulativeSize(cumulativeSize), tocIndex(tocIndex) {}
+  };
+
+  struct TocEntry {
+    std::string title;
+    std::string href;
+    std::string anchor;
+    uint8_t level;
+    int16_t spineIndex;
+
+    TocEntry() : level(0), spineIndex(-1) {}
+    TocEntry(std::string title, std::string href, std::string anchor, const uint8_t level, const int16_t spineIndex)
+        : title(std::move(title)),
+          href(std::move(href)),
+          anchor(std::move(anchor)),
+          level(level),
+          spineIndex(spineIndex) {}
+  };
+
+ private:
+  std::string cachePath;
+  size_t lutOffset;
+  uint16_t spineCount;
+  uint16_t tocCount;
+  bool loaded;
+  bool buildMode;
+
+  File bookFile;
+  // Temp file handles during build
+  File spineFile;
+  File tocFile;
+
+  size_t writeSpineEntry(File& file, const SpineEntry& entry) const;
+  size_t writeTocEntry(File& file, const TocEntry& entry) const;
+  SpineEntry readSpineEntry(File& file) const;
+  TocEntry readTocEntry(File& file) const;
+
+ public:
+  BookMetadata coreMetadata;
+
+  explicit BookMetadataCache(std::string cachePath)
+      : cachePath(std::move(cachePath)), lutOffset(0), spineCount(0), tocCount(0), loaded(false), buildMode(false) {}
+  ~BookMetadataCache() = default;
+
+  // Building phase (stream to disk immediately)
+  bool beginWrite();
+  bool beginContentOpfPass();
+  void createSpineEntry(const std::string& href);
+  bool endContentOpfPass();
+  bool beginTocPass();
+  void createTocEntry(const std::string& title, const std::string& href, const std::string& anchor, uint8_t level);
+  bool endTocPass();
+  bool endWrite();
+  bool cleanupTmpFiles() const;
+
+  // Post-processing to update mappings and sizes
+  bool buildBookBin(const std::string& epubPath, const BookMetadata& metadata);
+
+  // Reading phase (read mode)
+  bool load();
+  SpineEntry getSpineEntry(int index);
+  TocEntry getTocEntry(int index);
+  int getSpineCount() const { return spineCount; }
+  int getTocCount() const { return tocCount; }
+  bool isLoaded() const { return loaded; }
+};
@@ -1,10 +0,0 @@
-#pragma once
-
-#include <string>
-
-struct EpubTocEntry {
-  std::string title;
-  std::string href;
-  std::string anchor;
-  uint8_t level;
-};
@@ -2,6 +2,26 @@

 #include <SD.h>

+#include <vector>
+
+bool FsHelpers::openFileForRead(const char* moduleName, const std::string& path, File& file) {
+  file = SD.open(path.c_str(), FILE_READ);
+  if (!file) {
+    Serial.printf("[%lu] [%s] Failed to open file for reading: %s\n", millis(), moduleName, path.c_str());
+    return false;
+  }
+  return true;
+}
+
+bool FsHelpers::openFileForWrite(const char* moduleName, const std::string& path, File& file) {
+  file = SD.open(path.c_str(), FILE_WRITE, true);
+  if (!file) {
+    Serial.printf("[%lu] [%s] Failed to open file for writing: %s\n", millis(), moduleName, path.c_str());
+    return false;
+  }
+  return true;
+}
+
 bool FsHelpers::removeDir(const char* path) {
  // 1. Open the directory
  File dir = SD.open(path);
@@ -34,3 +54,39 @@ bool FsHelpers::removeDir(const char* path) {

  return SD.rmdir(path);
 }
+
+std::string FsHelpers::normalisePath(const std::string& path) {
+  std::vector<std::string> components;
+  std::string component;
+
+  for (const auto c : path) {
+    if (c == '/') {
+      if (!component.empty()) {
+        if (component == "..") {
+          if (!components.empty()) {
+            components.pop_back();
+          }
+        } else {
+          components.push_back(component);
+        }
+        component.clear();
+      }
+    } else {
+      component += c;
+    }
+  }
+
+  if (!component.empty()) {
+    components.push_back(component);
+  }
+
+  std::string result;
+  for (const auto& c : components) {
+    if (!result.empty()) {
+      result += "/";
+    }
+    result += c;
+  }
+
+  return result;
+}
@@ -1,6 +1,12 @@
 #pragma once
+#include <FS.h>
+
+#include <string>

 class FsHelpers {
 public:
+  static bool openFileForRead(const char* moduleName, const std::string& path, File& file);
+  static bool openFileForWrite(const char* moduleName, const std::string& path, File& file);
  static bool removeDir(const char* path);
+  static std::string normalisePath(const std::string& path);
 };
@@ -7,48 +7,50 @@ namespace {
 constexpr uint8_t PAGE_FILE_VERSION = 3;
 }

-void PageLine::render(GfxRenderer& renderer, const int fontId) { block->render(renderer, fontId, xPos, yPos); }
-
-void PageLine::serialize(std::ostream& os) {
-  serialization::writePod(os, xPos);
-  serialization::writePod(os, yPos);
-
-  // serialize TextBlock pointed to by PageLine
-  block->serialize(os);
+void PageLine::render(GfxRenderer& renderer, const int fontId, const int xOffset, const int yOffset) {
+  block->render(renderer, fontId, xPos + xOffset, yPos + yOffset);
 }

-std::unique_ptr<PageLine> PageLine::deserialize(std::istream& is) {
+void PageLine::serialize(File& file) {
+  serialization::writePod(file, xPos);
+  serialization::writePod(file, yPos);
+
+  // serialize TextBlock pointed to by PageLine
+  block->serialize(file);
+}
+
+std::unique_ptr<PageLine> PageLine::deserialize(File& file) {
  int16_t xPos;
  int16_t yPos;
-  serialization::readPod(is, xPos);
-  serialization::readPod(is, yPos);
+  serialization::readPod(file, xPos);
+  serialization::readPod(file, yPos);

-  auto tb = TextBlock::deserialize(is);
+  auto tb = TextBlock::deserialize(file);
  return std::unique_ptr<PageLine>(new PageLine(std::move(tb), xPos, yPos));
 }

-void Page::render(GfxRenderer& renderer, const int fontId) const {
+void Page::render(GfxRenderer& renderer, const int fontId, const int xOffset, const int yOffset) const {
  for (auto& element : elements) {
-    element->render(renderer, fontId);
+    element->render(renderer, fontId, xOffset, yOffset);
  }
 }

-void Page::serialize(std::ostream& os) const {
-  serialization::writePod(os, PAGE_FILE_VERSION);
+void Page::serialize(File& file) const {
+  serialization::writePod(file, PAGE_FILE_VERSION);

  const uint32_t count = elements.size();
-  serialization::writePod(os, count);
+  serialization::writePod(file, count);

  for (const auto& el : elements) {
    // Only PageLine exists currently
-    serialization::writePod(os, static_cast<uint8_t>(TAG_PageLine));
-    el->serialize(os);
+    serialization::writePod(file, static_cast<uint8_t>(TAG_PageLine));
+    el->serialize(file);
  }
 }

-std::unique_ptr<Page> Page::deserialize(std::istream& is) {
+std::unique_ptr<Page> Page::deserialize(File& file) {
  uint8_t version;
-  serialization::readPod(is, version);
+  serialization::readPod(file, version);
  if (version != PAGE_FILE_VERSION) {
    Serial.printf("[%lu] [PGE] Deserialization failed: Unknown version %u\n", millis(), version);
    return nullptr;
@@ -57,14 +59,14 @@ std::unique_ptr<Page> Page::deserialize(std::istream& is) {
  auto page = std::unique_ptr<Page>(new Page());

  uint32_t count;
-  serialization::readPod(is, count);
+  serialization::readPod(file, count);

  for (uint32_t i = 0; i < count; i++) {
    uint8_t tag;
-    serialization::readPod(is, tag);
+    serialization::readPod(file, tag);

    if (tag == TAG_PageLine) {
-      auto pl = PageLine::deserialize(is);
+      auto pl = PageLine::deserialize(file);
      page->elements.push_back(std::move(pl));
    } else {
      Serial.printf("[%lu] [PGE] Deserialization failed: Unknown tag %u\n", millis(), tag);
@@ -1,4 +1,6 @@
 #pragma once
+#include <FS.h>
+
 #include <utility>
 #include <vector>

@@ -15,8 +17,8 @@ class PageElement {
  int16_t yPos;
  explicit PageElement(const int16_t xPos, const int16_t yPos) : xPos(xPos), yPos(yPos) {}
  virtual ~PageElement() = default;
-  virtual void render(GfxRenderer& renderer, int fontId) = 0;
-  virtual void serialize(std::ostream& os) = 0;
+  virtual void render(GfxRenderer& renderer, int fontId, int xOffset, int yOffset) = 0;
+  virtual void serialize(File& file) = 0;
 };

 // a line from a block element
@@ -26,16 +28,16 @@ class PageLine final : public PageElement {
 public:
  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) override;
-  void serialize(std::ostream& os) override;
-  static std::unique_ptr<PageLine> deserialize(std::istream& is);
+  void render(GfxRenderer& renderer, int fontId, int xOffset, int yOffset) override;
+  void serialize(File& file) override;
+  static std::unique_ptr<PageLine> deserialize(File& file);
 };

 class Page {
 public:
  // the list of block index and line numbers on this page
  std::vector<std::shared_ptr<PageElement>> elements;
-  void render(GfxRenderer& renderer, int fontId) const;
-  void serialize(std::ostream& os) const;
-  static std::unique_ptr<Page> deserialize(std::istream& is);
+  void render(GfxRenderer& renderer, int fontId, int xOffset, int yOffset) const;
+  void serialize(File& file) const;
+  static std::unique_ptr<Page> deserialize(File& file);
 };
@@ -18,14 +18,14 @@ void ParsedText::addWord(std::string word, const EpdFontStyle fontStyle) {
 }

 // Consumes data to minimize memory usage
-void ParsedText::layoutAndExtractLines(const GfxRenderer& renderer, const int fontId, const int horizontalMargin,
+void ParsedText::layoutAndExtractLines(const GfxRenderer& renderer, const int fontId, const int viewportWidth,
                                       const std::function<void(std::shared_ptr<TextBlock>)>& processLine,
                                       const bool includeLastLine) {
  if (words.empty()) {
    return;
  }

-  const int pageWidth = renderer.getScreenWidth() - horizontalMargin;
+  const int pageWidth = viewportWidth;
  const int spaceWidth = renderer.getSpaceWidth(fontId);
  const auto wordWidths = calculateWordWidths(renderer, fontId);
  const auto lineBreakIndices = computeLineBreaks(pageWidth, spaceWidth, wordWidths);
@@ -106,21 +106,34 @@ std::vector<size_t> ParsedText::computeLineBreaks(const int pageWidth, const int
        ans[i] = j;  // j is the index of the last word in this optimal line
      }
    }
+
+    // Handle oversized word: if no valid configuration found, force single-word line
+    // This prevents cascade failure where one oversized word breaks all preceding words
+    if (dp[i] == MAX_COST) {
+      ans[i] = i;  // Just this word on its own line
+      // Inherit cost from next word to allow subsequent words to find valid configurations
+      if (i + 1 < static_cast<int>(totalWordCount)) {
+        dp[i] = dp[i + 1];
+      } else {
+        dp[i] = 0;
+      }
+    }
  }

  // Stores the index of the word that starts the next line (last_word_index + 1)
  std::vector<size_t> lineBreakIndices;
  size_t currentWordIndex = 0;
-  constexpr size_t MAX_LINES = 1000;

  while (currentWordIndex < totalWordCount) {
-    if (lineBreakIndices.size() >= MAX_LINES) {
-      break;
+    size_t nextBreakIndex = ans[currentWordIndex] + 1;
+
+    // Safety check: prevent infinite loop if nextBreakIndex doesn't advance
+    if (nextBreakIndex <= currentWordIndex) {
+      // Force advance by at least one word to avoid infinite loop
+      nextBreakIndex = currentWordIndex + 1;
    }

-    size_t nextBreakIndex = ans[currentWordIndex] + 1;
    lineBreakIndices.push_back(nextBreakIndex);
-
    currentWordIndex = nextBreakIndex;
  }

@@ -34,7 +34,7 @@ class ParsedText {
  TextBlock::BLOCK_STYLE getStyle() const { return style; }
  size_t size() const { return words.size(); }
  bool isEmpty() const { return words.empty(); }
-  void layoutAndExtractLines(const GfxRenderer& renderer, int fontId, int horizontalMargin,
+  void layoutAndExtractLines(const GfxRenderer& renderer, int fontId, int viewportWidth,
                             const std::function<void(std::shared_ptr<TextBlock>)>& processLine,
                             bool includeLastLine = true);
 };
@@ -1,22 +1,23 @@
 #include "Section.h"

+#include <FsHelpers.h>
 #include <SD.h>
 #include <Serialization.h>

-#include <fstream>
-
-#include "FsHelpers.h"
 #include "Page.h"
 #include "parsers/ChapterHtmlSlimParser.h"

 namespace {
-constexpr uint8_t SECTION_FILE_VERSION = 5;
-}
+constexpr uint8_t SECTION_FILE_VERSION = 6;
+}  // namespace

 void Section::onPageComplete(std::unique_ptr<Page> page) {
  const auto filePath = cachePath + "/page_" + std::to_string(pageCount) + ".bin";

-  std::ofstream outputFile("/sd" + filePath);
+  File outputFile;
+  if (!FsHelpers::openFileForWrite("SCT", filePath, outputFile)) {
+    return;
+  }
  page->serialize(outputFile);
  outputFile.close();

@@ -25,36 +26,30 @@ void Section::onPageComplete(std::unique_ptr<Page> page) {
  pageCount++;
 }

-void Section::writeCacheMetadata(const int fontId, const float lineCompression, const int marginTop,
-                                 const int marginRight, const int marginBottom, const int marginLeft,
-                                 const bool extraParagraphSpacing) const {
-  std::ofstream outputFile(("/sd" + cachePath + "/section.bin").c_str());
+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)) {
+    return;
+  }
  serialization::writePod(outputFile, SECTION_FILE_VERSION);
  serialization::writePod(outputFile, fontId);
  serialization::writePod(outputFile, lineCompression);
-  serialization::writePod(outputFile, marginTop);
-  serialization::writePod(outputFile, marginRight);
-  serialization::writePod(outputFile, marginBottom);
-  serialization::writePod(outputFile, marginLeft);
  serialization::writePod(outputFile, extraParagraphSpacing);
+  serialization::writePod(outputFile, viewportWidth);
+  serialization::writePod(outputFile, viewportHeight);
  serialization::writePod(outputFile, pageCount);
  outputFile.close();
 }

-bool Section::loadCacheMetadata(const int fontId, const float lineCompression, const int marginTop,
-                                const int marginRight, const int marginBottom, const int marginLeft,
-                                const bool extraParagraphSpacing) {
-  if (!SD.exists(cachePath.c_str())) {
-    return false;
-  }
-
+bool Section::loadCacheMetadata(const int fontId, const float lineCompression, const bool extraParagraphSpacing,
+                                const int viewportWidth, const int viewportHeight) {
  const auto sectionFilePath = cachePath + "/section.bin";
-  if (!SD.exists(sectionFilePath.c_str())) {
+  File inputFile;
+  if (!FsHelpers::openFileForRead("SCT", sectionFilePath, inputFile)) {
    return false;
  }

-  std::ifstream inputFile(("/sd" + sectionFilePath).c_str());
-
  // Match parameters
  {
    uint8_t version;
@@ -66,20 +61,18 @@ bool Section::loadCacheMetadata(const int fontId, const float lineCompression, c
      return false;
    }

-    int fileFontId, fileMarginTop, fileMarginRight, fileMarginBottom, fileMarginLeft;
+    int fileFontId, fileViewportWidth, fileViewportHeight;
    float fileLineCompression;
    bool fileExtraParagraphSpacing;
    serialization::readPod(inputFile, fileFontId);
    serialization::readPod(inputFile, fileLineCompression);
-    serialization::readPod(inputFile, fileMarginTop);
-    serialization::readPod(inputFile, fileMarginRight);
-    serialization::readPod(inputFile, fileMarginBottom);
-    serialization::readPod(inputFile, fileMarginLeft);
    serialization::readPod(inputFile, fileExtraParagraphSpacing);
+    serialization::readPod(inputFile, fileViewportWidth);
+    serialization::readPod(inputFile, fileViewportHeight);

-    if (fontId != fileFontId || lineCompression != fileLineCompression || marginTop != fileMarginTop ||
-        marginRight != fileMarginRight || marginBottom != fileMarginBottom || marginLeft != fileMarginLeft ||
-        extraParagraphSpacing != fileExtraParagraphSpacing) {
+    if (fontId != fileFontId || lineCompression != fileLineCompression ||
+        extraParagraphSpacing != fileExtraParagraphSpacing || viewportWidth != fileViewportWidth ||
+        viewportHeight != fileViewportHeight) {
      inputFile.close();
      Serial.printf("[%lu] [SCT] Deserialization failed: Parameters do not match\n", millis());
      clearCache();
@@ -114,31 +107,58 @@ bool Section::clearCache() const {
  return true;
 }

-bool Section::persistPageDataToSD(const int fontId, const float lineCompression, const int marginTop,
-                                  const int marginRight, const int marginBottom, const int marginLeft,
-                                  const bool extraParagraphSpacing) {
-  const auto localPath = epub->getSpineItem(spineIndex);
-
-  // TODO: Should we get rid of this file all together?
-  //       It currently saves us a bit of memory by allowing for all the inflation bits to be released
-  //       before loading the XML parser
+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) {
+  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";
-  File f = SD.open(tmpHtmlPath.c_str(), FILE_WRITE, true);
-  bool success = epub->readItemContentsToStream(localPath, f, 1024);
-  f.close();
+
+  // Retry logic for SD card timing issues
+  bool success = false;
+  size_t fileSize = 0;
+  for (int attempt = 0; attempt < 3 && !success; attempt++) {
+    if (attempt > 0) {
+      Serial.printf("[%lu] [SCT] Retrying stream (attempt %d)...\n", millis(), attempt + 1);
+      delay(50);  // Brief delay before retry
+    }
+
+    // Remove any incomplete file from previous attempt before retrying
+    if (SD.exists(tmpHtmlPath.c_str())) {
+      SD.remove(tmpHtmlPath.c_str());
+    }
+
+    File tmpHtml;
+    if (!FsHelpers::openFileForWrite("SCT", tmpHtmlPath, tmpHtml)) {
+      continue;
+    }
+    success = epub->readItemContentsToStream(localPath, tmpHtml, 1024);
+    fileSize = tmpHtml.size();
+    tmpHtml.close();
+
+    // If streaming failed, remove the incomplete file immediately
+    if (!success && SD.exists(tmpHtmlPath.c_str())) {
+      SD.remove(tmpHtmlPath.c_str());
+      Serial.printf("[%lu] [SCT] Removed incomplete temp file after failed attempt\n", millis());
+    }
+  }

  if (!success) {
-    Serial.printf("[%lu] [SCT] Failed to stream item contents to temp file\n", millis());
+    Serial.printf("[%lu] [SCT] Failed to stream item contents to temp file after retries\n", millis());
    return false;
  }

-  Serial.printf("[%lu] [SCT] Streamed temp HTML to %s\n", millis(), tmpHtmlPath.c_str());
+  Serial.printf("[%lu] [SCT] Streamed temp HTML to %s (%d bytes)\n", millis(), tmpHtmlPath.c_str(), fileSize);

-  const auto sdTmpHtmlPath = "/sd" + tmpHtmlPath;
+  // Only show progress bar for larger chapters where rendering overhead is worth it
+  if (progressSetupFn && fileSize >= MIN_SIZE_FOR_PROGRESS) {
+    progressSetupFn();
+  }

-  ChapterHtmlSlimParser visitor(sdTmpHtmlPath.c_str(), renderer, fontId, lineCompression, marginTop, marginRight,
-                                marginBottom, marginLeft, extraParagraphSpacing,
-                                [this](std::unique_ptr<Page> page) { this->onPageComplete(std::move(page)); });
+  ChapterHtmlSlimParser visitor(
+      tmpHtmlPath, renderer, fontId, lineCompression, extraParagraphSpacing, viewportWidth, viewportHeight,
+      [this](std::unique_ptr<Page> page) { this->onPageComplete(std::move(page)); }, progressFn);
  success = visitor.parseAndBuildPages();

  SD.remove(tmpHtmlPath.c_str());
@@ -147,19 +167,18 @@ bool Section::persistPageDataToSD(const int fontId, const float lineCompression,
    return false;
  }

-  writeCacheMetadata(fontId, lineCompression, marginTop, marginRight, marginBottom, marginLeft, extraParagraphSpacing);
+  writeCacheMetadata(fontId, lineCompression, extraParagraphSpacing, viewportWidth, viewportHeight);

  return true;
 }

 std::unique_ptr<Page> Section::loadPageFromSD() const {
-  const auto filePath = "/sd" + cachePath + "/page_" + std::to_string(currentPage) + ".bin";
-  if (!SD.exists(filePath.c_str() + 3)) {
-    Serial.printf("[%lu] [SCT] Page file does not exist: %s\n", millis(), filePath.c_str());
+  const auto filePath = cachePath + "/page_" + std::to_string(currentPage) + ".bin";
+
+  File inputFile;
+  if (!FsHelpers::openFileForRead("SCT", filePath, inputFile)) {
    return nullptr;
  }
-
-  std::ifstream inputFile(filePath);
  auto page = Page::deserialize(inputFile);
  inputFile.close();
  return page;
@@ -1,4 +1,5 @@
 #pragma once
+#include <functional>
 #include <memory>

 #include "Epub.h"
@@ -12,8 +13,8 @@ class Section {
  GfxRenderer& renderer;
  std::string cachePath;

-  void writeCacheMetadata(int fontId, float lineCompression, int marginTop, int marginRight, int marginBottom,
-                          int marginLeft, bool extraParagraphSpacing) const;
+  void writeCacheMetadata(int fontId, float lineCompression, bool extraParagraphSpacing, int viewportWidth,
+                          int viewportHeight) const;
  void onPageComplete(std::unique_ptr<Page> page);

 public:
@@ -26,11 +27,12 @@ class Section {
        renderer(renderer),
        cachePath(epub->getCachePath() + "/" + std::to_string(spineIndex)) {}
  ~Section() = default;
-  bool loadCacheMetadata(int fontId, float lineCompression, int marginTop, int marginRight, int marginBottom,
-                         int marginLeft, bool extraParagraphSpacing);
+  bool loadCacheMetadata(int fontId, float lineCompression, bool extraParagraphSpacing, int viewportWidth,
+                         int viewportHeight);
  void setupCacheDir() const;
  bool clearCache() const;
-  bool persistPageDataToSD(int fontId, float lineCompression, int marginTop, int marginRight, int marginBottom,
-                           int marginLeft, bool extraParagraphSpacing);
+  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;
 };
@@ -4,11 +4,18 @@
 #include <Serialization.h>

 void TextBlock::render(const GfxRenderer& renderer, const int fontId, const int x, const int y) const {
+  // Validate iterator bounds before rendering
+  if (words.size() != wordXpos.size() || words.size() != wordStyles.size()) {
+    Serial.printf("[%lu] [TXB] Render skipped: size mismatch (words=%u, xpos=%u, styles=%u)\n", millis(),
+                  (uint32_t)words.size(), (uint32_t)wordXpos.size(), (uint32_t)wordStyles.size());
+    return;
+  }
+
  auto wordIt = words.begin();
  auto wordStylesIt = wordStyles.begin();
  auto wordXposIt = wordXpos.begin();

-  for (int i = 0; i < words.size(); i++) {
+  for (size_t i = 0; i < words.size(); i++) {
    renderer.drawText(fontId, *wordXposIt + x, y, wordIt->c_str(), true, *wordStylesIt);

    std::advance(wordIt, 1);
@@ -17,27 +24,27 @@ void TextBlock::render(const GfxRenderer& renderer, const int fontId, const int
  }
 }

-void TextBlock::serialize(std::ostream& os) const {
+void TextBlock::serialize(File& file) const {
  // words
  const uint32_t wc = words.size();
-  serialization::writePod(os, wc);
-  for (const auto& w : words) serialization::writeString(os, w);
+  serialization::writePod(file, wc);
+  for (const auto& w : words) serialization::writeString(file, w);

  // wordXpos
  const uint32_t xc = wordXpos.size();
-  serialization::writePod(os, xc);
-  for (auto x : wordXpos) serialization::writePod(os, x);
+  serialization::writePod(file, xc);
+  for (auto x : wordXpos) serialization::writePod(file, x);

  // wordStyles
  const uint32_t sc = wordStyles.size();
-  serialization::writePod(os, sc);
-  for (auto s : wordStyles) serialization::writePod(os, s);
+  serialization::writePod(file, sc);
+  for (auto s : wordStyles) serialization::writePod(file, s);

  // style
-  serialization::writePod(os, style);
+  serialization::writePod(file, style);
 }

-std::unique_ptr<TextBlock> TextBlock::deserialize(std::istream& is) {
+std::unique_ptr<TextBlock> TextBlock::deserialize(File& file) {
  uint32_t wc, xc, sc;
  std::list<std::string> words;
  std::list<uint16_t> wordXpos;
@@ -45,22 +52,36 @@ std::unique_ptr<TextBlock> TextBlock::deserialize(std::istream& is) {
  BLOCK_STYLE style;

  // words
-  serialization::readPod(is, wc);
+  serialization::readPod(file, wc);
+
+  // Sanity check: prevent allocation of unreasonably large lists (max 10000 words per block)
+  if (wc > 10000) {
+    Serial.printf("[%lu] [TXB] Deserialization failed: word count %u exceeds maximum\n", millis(), wc);
+    return nullptr;
+  }
+
  words.resize(wc);
-  for (auto& w : words) serialization::readString(is, w);
+  for (auto& w : words) serialization::readString(file, w);

  // wordXpos
-  serialization::readPod(is, xc);
+  serialization::readPod(file, xc);
  wordXpos.resize(xc);
-  for (auto& x : wordXpos) serialization::readPod(is, x);
+  for (auto& x : wordXpos) serialization::readPod(file, x);

  // wordStyles
-  serialization::readPod(is, sc);
+  serialization::readPod(file, sc);
  wordStyles.resize(sc);
-  for (auto& s : wordStyles) serialization::readPod(is, s);
+  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
-  serialization::readPod(is, style);
+  serialization::readPod(file, style);

  return std::unique_ptr<TextBlock>(new TextBlock(std::move(words), std::move(wordXpos), std::move(wordStyles), style));
 }
@@ -1,5 +1,6 @@
 #pragma once
 #include <EpdFontFamily.h>
+#include <FS.h>

 #include <list>
 #include <memory>
@@ -35,6 +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(std::ostream& os) const;
-  static std::unique_ptr<TextBlock> deserialize(std::istream& is);
+  void serialize(File& file) const;
+  static std::unique_ptr<TextBlock> deserialize(File& file);
 };
@@ -1,5 +1,6 @@
 #include "ChapterHtmlSlimParser.h"

+#include <FsHelpers.h>
 #include <GfxRenderer.h>
 #include <HardwareSerial.h>
 #include <expat.h>
@@ -10,13 +11,16 @@
 const char* HEADER_TAGS[] = {"h1", "h2", "h3", "h4", "h5", "h6"};
 constexpr int NUM_HEADER_TAGS = sizeof(HEADER_TAGS) / sizeof(HEADER_TAGS[0]);

-const char* BLOCK_TAGS[] = {"p", "li", "div", "br"};
+// Minimum file size (in bytes) to show progress bar - smaller chapters don't benefit from it
+constexpr size_t MIN_SIZE_FOR_PROGRESS = 50 * 1024;  // 50KB
+
+const char* BLOCK_TAGS[] = {"p", "li", "div", "br", "blockquote"};
 constexpr int NUM_BLOCK_TAGS = sizeof(BLOCK_TAGS) / sizeof(BLOCK_TAGS[0]);

-const char* BOLD_TAGS[] = {"b"};
+const char* BOLD_TAGS[] = {"b", "strong"};
 constexpr int NUM_BOLD_TAGS = sizeof(BOLD_TAGS) / sizeof(BOLD_TAGS[0]);

-const char* ITALIC_TAGS[] = {"i"};
+const char* ITALIC_TAGS[] = {"i", "em"};
 constexpr int NUM_ITALIC_TAGS = sizeof(ITALIC_TAGS) / sizeof(ITALIC_TAGS[0]);

 const char* IMAGE_TAGS[] = {"img"};
@@ -151,7 +155,7 @@ void XMLCALL ChapterHtmlSlimParser::characterData(void* userData, const XML_Char
  if (self->currentTextBlock->size() > 750) {
    Serial.printf("[%lu] [EHP] Text block too long, splitting into multiple pages\n", millis());
    self->currentTextBlock->layoutAndExtractLines(
-        self->renderer, self->fontId, self->marginLeft + self->marginRight,
+        self->renderer, self->fontId, self->viewportWidth,
        [self](const std::shared_ptr<TextBlock>& textBlock) { self->addLineToPage(textBlock); }, false);
  }
 }
@@ -214,48 +218,75 @@ bool ChapterHtmlSlimParser::parseAndBuildPages() {
    return false;
  }

-  XML_SetUserData(parser, this);
-  XML_SetElementHandler(parser, startElement, endElement);
-  XML_SetCharacterDataHandler(parser, characterData);
-
-  FILE* file = fopen(filepath, "r");
-  if (!file) {
-    Serial.printf("[%lu] [EHP] Couldn't open file %s\n", millis(), filepath);
+  File file;
+  if (!FsHelpers::openFileForRead("EHP", filepath, file)) {
    XML_ParserFree(parser);
    return false;
  }

+  // Get file size for progress calculation
+  const size_t totalSize = file.size();
+  size_t bytesRead = 0;
+  int lastProgress = -1;
+
+  XML_SetUserData(parser, this);
+  XML_SetElementHandler(parser, startElement, endElement);
+  XML_SetCharacterDataHandler(parser, characterData);
+
  do {
    void* const buf = XML_GetBuffer(parser, 1024);
    if (!buf) {
      Serial.printf("[%lu] [EHP] Couldn't allocate memory for buffer\n", millis());
+      XML_StopParser(parser, XML_FALSE);                // Stop any pending processing
+      XML_SetElementHandler(parser, nullptr, nullptr);  // Clear callbacks
+      XML_SetCharacterDataHandler(parser, nullptr);
      XML_ParserFree(parser);
-      fclose(file);
+      file.close();
      return false;
    }

-    const size_t len = fread(buf, 1, 1024, file);
+    const size_t len = file.read(static_cast<uint8_t*>(buf), 1024);

-    if (ferror(file)) {
+    if (len == 0) {
      Serial.printf("[%lu] [EHP] File read error\n", millis());
+      XML_StopParser(parser, XML_FALSE);                // Stop any pending processing
+      XML_SetElementHandler(parser, nullptr, nullptr);  // Clear callbacks
+      XML_SetCharacterDataHandler(parser, nullptr);
      XML_ParserFree(parser);
-      fclose(file);
+      file.close();
      return false;
    }

-    done = feof(file);
+    // Update progress (call every 10% change to avoid too frequent updates)
+    // Only show progress for larger chapters where rendering overhead is worth it
+    bytesRead += len;
+    if (progressFn && totalSize >= MIN_SIZE_FOR_PROGRESS) {
+      const int progress = static_cast<int>((bytesRead * 100) / totalSize);
+      if (lastProgress / 10 != progress / 10) {
+        lastProgress = progress;
+        progressFn(progress);
+      }
+    }
+
+    done = file.available() == 0;

    if (XML_ParseBuffer(parser, static_cast<int>(len), done) == XML_STATUS_ERROR) {
      Serial.printf("[%lu] [EHP] Parse error at line %lu:\n%s\n", millis(), XML_GetCurrentLineNumber(parser),
                    XML_ErrorString(XML_GetErrorCode(parser)));
+      XML_StopParser(parser, XML_FALSE);                // Stop any pending processing
+      XML_SetElementHandler(parser, nullptr, nullptr);  // Clear callbacks
+      XML_SetCharacterDataHandler(parser, nullptr);
      XML_ParserFree(parser);
-      fclose(file);
+      file.close();
      return false;
    }
  } while (!done);

+  XML_StopParser(parser, XML_FALSE);                // Stop any pending processing
+  XML_SetElementHandler(parser, nullptr, nullptr);  // Clear callbacks
+  XML_SetCharacterDataHandler(parser, nullptr);
  XML_ParserFree(parser);
-  fclose(file);
+  file.close();

  // Process last page if there is still text
  if (currentTextBlock) {
@@ -270,15 +301,14 @@ bool ChapterHtmlSlimParser::parseAndBuildPages() {

 void ChapterHtmlSlimParser::addLineToPage(std::shared_ptr<TextBlock> line) {
  const int lineHeight = renderer.getLineHeight(fontId) * lineCompression;
-  const int pageHeight = GfxRenderer::getScreenHeight() - marginTop - marginBottom;

-  if (currentPageNextY + lineHeight > pageHeight) {
+  if (currentPageNextY + lineHeight > viewportHeight) {
    completePageFn(std::move(currentPage));
    currentPage.reset(new Page());
-    currentPageNextY = marginTop;
+    currentPageNextY = 0;
  }

-  currentPage->elements.push_back(std::make_shared<PageLine>(line, marginLeft, currentPageNextY));
+  currentPage->elements.push_back(std::make_shared<PageLine>(line, 0, currentPageNextY));
  currentPageNextY += lineHeight;
 }

@@ -290,12 +320,12 @@ void ChapterHtmlSlimParser::makePages() {

  if (!currentPage) {
    currentPage.reset(new Page());
-    currentPageNextY = marginTop;
+    currentPageNextY = 0;
  }

  const int lineHeight = renderer.getLineHeight(fontId) * lineCompression;
  currentTextBlock->layoutAndExtractLines(
-      renderer, fontId, marginLeft + marginRight,
+      renderer, fontId, viewportWidth,
      [this](const std::shared_ptr<TextBlock>& textBlock) { addLineToPage(textBlock); });
  // Extra paragraph spacing if enabled
  if (extraParagraphSpacing) {
@@ -15,9 +15,10 @@ class GfxRenderer;
 #define MAX_WORD_SIZE 200

 class ChapterHtmlSlimParser {
-  const char* filepath;
+  const std::string& filepath;
  GfxRenderer& renderer;
  std::function<void(std::unique_ptr<Page>)> completePageFn;
+  std::function<void(int)> progressFn;  // Progress callback (0-100)
  int depth = 0;
  int skipUntilDepth = INT_MAX;
  int boldUntilDepth = INT_MAX;
@@ -31,11 +32,9 @@ class ChapterHtmlSlimParser {
  int16_t currentPageNextY = 0;
  int fontId;
  float lineCompression;
-  int marginTop;
-  int marginRight;
-  int marginBottom;
-  int marginLeft;
  bool extraParagraphSpacing;
+  int viewportWidth;
+  int viewportHeight;

  void startNewTextBlock(TextBlock::BLOCK_STYLE style);
  void makePages();
@@ -45,20 +44,20 @@ class ChapterHtmlSlimParser {
  static void XMLCALL endElement(void* userData, const XML_Char* name);

 public:
-  explicit ChapterHtmlSlimParser(const char* filepath, GfxRenderer& renderer, const int fontId,
-                                 const float lineCompression, const int marginTop, const int marginRight,
-                                 const int marginBottom, const int marginLeft, const bool extraParagraphSpacing,
-                                 const std::function<void(std::unique_ptr<Page>)>& completePageFn)
+  explicit ChapterHtmlSlimParser(const std::string& filepath, GfxRenderer& renderer, const int fontId,
+                                 const float lineCompression, const bool extraParagraphSpacing, const int viewportWidth,
+                                 const int viewportHeight,
+                                 const std::function<void(std::unique_ptr<Page>)>& completePageFn,
+                                 const std::function<void(int)>& progressFn = nullptr)
      : filepath(filepath),
        renderer(renderer),
        fontId(fontId),
        lineCompression(lineCompression),
-        marginTop(marginTop),
-        marginRight(marginRight),
-        marginBottom(marginBottom),
-        marginLeft(marginLeft),
        extraParagraphSpacing(extraParagraphSpacing),
-        completePageFn(completePageFn) {}
+        viewportWidth(viewportWidth),
+        viewportHeight(viewportHeight),
+        completePageFn(completePageFn),
+        progressFn(progressFn) {}
  ~ChapterHtmlSlimParser() = default;
  bool parseAndBuildPages();
  void addLineToPage(std::shared_ptr<TextBlock> line);
@@ -16,6 +16,8 @@ bool ContainerParser::setup() {

 ContainerParser::~ContainerParser() {
  if (parser) {
+    XML_StopParser(parser, XML_FALSE);                // Stop any pending processing
+    XML_SetElementHandler(parser, nullptr, nullptr);  // Clear callbacks
    XML_ParserFree(parser);
    parser = nullptr;
  }
--- a/Show More
+++ b/Show More