crosspoint-reader/lib/Epub/Epub/parsers/ContentOpfParser.cpp

#include "ContentOpfParser.h"

#include <FsHelpers.h>
#include <HardwareSerial.h>
#include <Serialization.h>
#include <ZipFile.h>

#include "../BookMetadataCache.h"

namespace {
constexpr char MEDIA_TYPE_NCX[] = "application/x-dtbncx+xml";
constexpr char itemCacheFile[] = "/.items.bin";
}  // namespace

bool ContentOpfParser::setup() {
  parser = XML_ParserCreate(nullptr);
  if (!parser) {
    Serial.printf("[%lu] [COF] Couldn't allocate memory for parser\n", millis());
    return false;
  }

  XML_SetUserData(parser, this);
  XML_SetElementHandler(parser, startElement, endElement);
  XML_SetCharacterDataHandler(parser, characterData);
  return true;
}

ContentOpfParser::~ContentOpfParser() {
  if (parser) {
    XML_StopParser(parser, XML_FALSE);                // Stop any pending processing
    XML_SetElementHandler(parser, nullptr, nullptr);  // Clear callbacks
    XML_SetCharacterDataHandler(parser, nullptr);
    XML_ParserFree(parser);
    parser = nullptr;
  }
  if (tempItemStore) {
    tempItemStore.close();
  }
  if (SD.exists((cachePath + itemCacheFile).c_str())) {
    SD.remove((cachePath + itemCacheFile).c_str());
  }
}

size_t ContentOpfParser::write(const uint8_t data) { return write(&data, 1); }

size_t ContentOpfParser::write(const uint8_t* buffer, const size_t size) {
  if (!parser) return 0;

  const uint8_t* currentBufferPos = buffer;
  auto remainingInBuffer = size;

  while (remainingInBuffer > 0) {
    void* const buf = XML_GetBuffer(parser, 1024);

    if (!buf) {
      Serial.printf("[%lu] [COF] 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);
      parser = nullptr;
      return 0;
    }

    const auto toRead = remainingInBuffer < 1024 ? remainingInBuffer : 1024;
    memcpy(buf, currentBufferPos, toRead);

    if (XML_ParseBuffer(parser, static_cast<int>(toRead), remainingSize == toRead) == XML_STATUS_ERROR) {
      Serial.printf("[%lu] [COF] Parse error at line %lu: %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);
      parser = nullptr;
      return 0;
    }

    currentBufferPos += toRead;
    remainingInBuffer -= toRead;
    remainingSize -= toRead;
  }

  return size;
}

void XMLCALL ContentOpfParser::startElement(void* userData, const XML_Char* name, const XML_Char** atts) {
  auto* self = static_cast<ContentOpfParser*>(userData);
  (void)atts;

  if (self->state == START && (strcmp(name, "package") == 0 || strcmp(name, "opf:package") == 0)) {
    self->state = IN_PACKAGE;
    return;
  }

  if (self->state == IN_PACKAGE && (strcmp(name, "metadata") == 0 || strcmp(name, "opf:metadata") == 0)) {
    self->state = IN_METADATA;
    return;
  }

  if (self->state == IN_METADATA && strcmp(name, "dc:title") == 0) {
    self->state = IN_BOOK_TITLE;
    return;
  }

  if (self->state == IN_PACKAGE && (strcmp(name, "manifest") == 0 || strcmp(name, "opf:manifest") == 0)) {
    self->state = IN_MANIFEST;
    if (!FsHelpers::openFileForWrite("COF", self->cachePath + itemCacheFile, self->tempItemStore)) {
      Serial.printf(
          "[%lu] [COF] Couldn't open temp items file for writing. This is probably going to be a fatal error.\n",
          millis());
    }
    return;
  }

  if (self->state == IN_PACKAGE && (strcmp(name, "spine") == 0 || strcmp(name, "opf:spine") == 0)) {
    self->state = IN_SPINE;
    if (!FsHelpers::openFileForRead("COF", self->cachePath + itemCacheFile, self->tempItemStore)) {
      Serial.printf(
          "[%lu] [COF] Couldn't open temp items file for reading. This is probably going to be a fatal error.\n",
          millis());
    }
    return;
  }

  if (self->state == IN_METADATA && (strcmp(name, "meta") == 0 || strcmp(name, "opf:meta") == 0)) {
    bool isCover = false;
    std::string coverItemId;

    for (int i = 0; atts[i]; i += 2) {
      if (strcmp(atts[i], "name") == 0 && strcmp(atts[i + 1], "cover") == 0) {
        isCover = true;
      } else if (strcmp(atts[i], "content") == 0) {
        coverItemId = atts[i + 1];
      }
    }

    if (isCover) {
      self->coverItemId = coverItemId;
    }
    return;
  }

  if (self->state == IN_MANIFEST && (strcmp(name, "item") == 0 || strcmp(name, "opf:item") == 0)) {
    std::string itemId;
    std::string href;
    std::string mediaType;

    for (int i = 0; atts[i]; i += 2) {
      if (strcmp(atts[i], "id") == 0) {
        itemId = atts[i + 1];
      } else if (strcmp(atts[i], "href") == 0) {
        href = self->baseContentPath + atts[i + 1];
      } else if (strcmp(atts[i], "media-type") == 0) {
        mediaType = atts[i + 1];
      }
    }

    // Write items down to SD card
    serialization::writeString(self->tempItemStore, itemId);
    serialization::writeString(self->tempItemStore, href);

    if (itemId == self->coverItemId) {
      self->coverItemHref = href;
    }

    if (mediaType == MEDIA_TYPE_NCX) {
      if (self->tocNcxPath.empty()) {
        self->tocNcxPath = href;
      } else {
        Serial.printf("[%lu] [COF] Warning: Multiple NCX files found in manifest. Ignoring duplicate: %s\n", millis(),
                      href.c_str());
      }
    }
    return;
  }

  // NOTE: This relies on spine appearing after item manifest (which is pretty safe as it's part of the EPUB spec)
  // Only run the spine parsing if there's a cache to add it to
  if (self->cache) {
    if (self->state == IN_SPINE && (strcmp(name, "itemref") == 0 || strcmp(name, "opf:itemref") == 0)) {
      for (int i = 0; atts[i]; i += 2) {
        if (strcmp(atts[i], "idref") == 0) {
          const std::string idref = atts[i + 1];
          // Resolve the idref to href using items map
          self->tempItemStore.seek(0);
          std::string itemId;
          std::string href;
          while (self->tempItemStore.available()) {
            serialization::readString(self->tempItemStore, itemId);
            serialization::readString(self->tempItemStore, href);
            if (itemId == idref) {
              self->cache->createSpineEntry(href);
              break;
            }
          }
        }
      }
      return;
    }
  }
}

void XMLCALL ContentOpfParser::characterData(void* userData, const XML_Char* s, const int len) {
  auto* self = static_cast<ContentOpfParser*>(userData);

  if (self->state == IN_BOOK_TITLE) {
    self->title.append(s, len);
    return;
  }
}

void XMLCALL ContentOpfParser::endElement(void* userData, const XML_Char* name) {
  auto* self = static_cast<ContentOpfParser*>(userData);
  (void)name;

  if (self->state == IN_SPINE && (strcmp(name, "spine") == 0 || strcmp(name, "opf:spine") == 0)) {
    self->state = IN_PACKAGE;
    self->tempItemStore.close();
    return;
  }

  if (self->state == IN_MANIFEST && (strcmp(name, "manifest") == 0 || strcmp(name, "opf:manifest") == 0)) {
    self->state = IN_PACKAGE;
    self->tempItemStore.close();
    return;
  }

  if (self->state == IN_BOOK_TITLE && strcmp(name, "dc:title") == 0) {
    self->state = IN_METADATA;
    return;
  }

  if (self->state == IN_METADATA && (strcmp(name, "metadata") == 0 || strcmp(name, "opf:metadata") == 0)) {
    self->state = IN_PACKAGE;
    return;
  }

  if (self->state == IN_PACKAGE && (strcmp(name, "package") == 0 || strcmp(name, "opf:package") == 0)) {
    self->state = START;
    return;
  }
}