stegdoc 5.7.0 → 6.1.0

package/src/lib/file-utils.js

@@ -1,160 +0,0 @@
-const fs = require('fs');
-const path = require('path');
-const { parseMetadata } = require('./metadata');
-const { parseFilename } = require('./utils');
-
-/**
- * Extract content and metadata based on format
- * @param {object} readResult - Result from readFile
- * @param {string} format - File format ('xlsx' or 'docx')
- * @returns {object} { encryptedContent, encryptionMeta, metadata }
- */
-function extractContent(readResult, format) {
-  // v5 log-embed format returns metadata already parsed
-  if (readResult.formatVersion === 'v5') {
-    return {
-      encryptedContent: null, // v5 uses payloadBuffer instead
-      encryptionMeta: readResult.encryptionMeta,
-      metadata: readResult.metadata,
-      payloadBuffer: readResult.payloadBuffer,
-    };
-  }
-
-  if (format === 'xlsx') {
-    return {
-      encryptedContent: readResult.base64Content,
-      encryptionMeta: readResult.encryptionMeta,
-      metadata: parseMetadata(readResult.metadata),
-    };
-  } else {
-    // DOCX: encryption meta is embedded in content with ||| separator
-    const { base64Content, metadata } = readResult;
-
-    // Check if this is a v2+ encrypted file
-    if (base64Content.includes('|||')) {
-      const [encryptionMeta, encryptedContent] = base64Content.split('|||');
-      return {
-        encryptedContent,
-        encryptionMeta,
-        metadata,
-      };
-    }
-
-    // Legacy unencrypted DOCX
-    return {
-      encryptedContent: base64Content,
-      encryptionMeta: null,
-      metadata,
-    };
-  }
-}
-
-/**
- * Find all parts of a multi-part file in a directory
- * @param {string} dirPath - Directory to search
- * @param {string} hash - Original hash from metadata
- * @param {string} format - File format ('xlsx' or 'docx')
- * @param {number} [expectedParts] - Expected total parts (optional, for validation)
- * @returns {Array<{path: string, partNumber: number, filename: string}>} Array of parts sorted by part number
- */
-function findMultiPartFiles(dirPath, hash, format, expectedParts = null) {
-  const files = fs.readdirSync(dirPath);
-  const parts = [];
-  const ext = format === 'docx' ? '.docx' : '.xlsx';
-
-  // Also support legacy hex filenames for backward compatibility
-  const legacyBaseHash = hash.length >= 16 ? hash.slice(0, 16) : hash;
-
-  for (const file of files) {
-    if (!file.toLowerCase().endsWith(ext)) continue;
-
-    const parsed = parseFilename(file);
-    if (!parsed || parsed.partNumber === null) continue;
-
-    // Match by new realistic filename pattern
-    // Check if file matches the expected pattern (same reportId from hash)
-    if (parsed.reportId) {
-      // New realistic format - match by reportId (last 4 chars of hash)
-      // This is deterministic and doesn't depend on current date
-      const expectedReportId = hash.slice(-4).toUpperCase();
-      if (parsed.reportId === expectedReportId) {
-        parts.push({
-          path: path.join(dirPath, file),
-          partNumber: parsed.partNumber,
-          filename: file,
-          dateStr: parsed.dateStr,
-          timeStr: parsed.timeStr,
-        });
-      }
-    } else if (parsed.baseHash === legacyBaseHash) {
-      // Legacy hex format - match by base hash
-      parts.push({
-        path: path.join(dirPath, file),
-        partNumber: parsed.partNumber,
-        filename: file,
-      });
-    }
-  }
-
-  // Sort by part number
-  parts.sort((a, b) => a.partNumber - b.partNumber);
-
-  // For realistic filenames, ensure all parts have the same date/time pattern
-  // This handles edge cases where multiple file sets might share the same reportId
-  if (parts.length > 0 && parts[0].dateStr) {
-    const refDateStr = parts[0].dateStr;
-    const refTimeStr = parts[0].timeStr;
-    const filteredParts = parts.filter(
-      (p) => p.dateStr === refDateStr && p.timeStr === refTimeStr
-    );
-    // If filtering removed some parts, use the filtered set
-    if (filteredParts.length !== parts.length) {
-      parts.length = 0;
-      parts.push(...filteredParts);
-    }
-  }
-
-  // Validate sequential parts if expectedParts is provided
-  if (expectedParts !== null && parts.length === expectedParts) {
-    for (let i = 0; i < expectedParts; i++) {
-      if (parts[i].partNumber !== i + 1) {
-        throw new Error(`Missing part ${i + 1}. Parts must be sequential.`);
-      }
-    }
-  }
-
-  return parts;
-}
-
-/**
- * Check if a directory is writable
- * @param {string} dirPath - Directory path to check
- * @returns {boolean} True if writable
- */
-function isDirectoryWritable(dirPath) {
-  try {
-    if (!fs.existsSync(dirPath)) {
-      fs.mkdirSync(dirPath, { recursive: true });
-    }
-    fs.accessSync(dirPath, fs.constants.W_OK);
-    return true;
-  } catch {
-    return false;
-  }
-}
-
-/**
- * Merge base64 chunks back into a single string
- * @param {Array<string>} chunks - Array of base64 chunks
- * @returns {string} Merged base64 string
- */
-function mergeBase64Chunks(chunks) {
-  return chunks.join('');
-}
-
-module.exports = {
-  extractContent,
-  findMultiPartFiles,
-  isDirectoryWritable,
-  mergeBase64Chunks,
-};

package/src/lib/xml-utils.js

@@ -1,115 +0,0 @@
-/**
- * Shared XML parsing utilities for DOCX and XLSX handlers
- * Handles namespace prefix variations (w:, ns0:, ns1:, etc.)
- */
-
-const AdmZip = require('adm-zip');
-const { XMLParser } = require('fast-xml-parser');
-
-// Reusable XML parser configured to strip namespace prefixes
-const xmlParser = new XMLParser({
-  ignoreAttributes: false,
-  attributeNamePrefix: '@_',
-  removeNSPrefix: true, // Strips ns0:, ns1:, w:, etc.
-});
-
-/**
- * Parse XML string with namespace-agnostic parser
- * @param {string} xmlString - XML content to parse
- * @returns {object} Parsed XML as JavaScript object
- */
-function parseXml(xmlString) {
-  return xmlParser.parse(xmlString);
-}
-
-/**
- * Read and parse an XML file from a ZIP archive (DOCX/XLSX)
- * @param {string|Buffer} zipPath - Path to ZIP file or Buffer
- * @param {string} entryPath - Path to XML file within ZIP
- * @returns {object|null} Parsed XML or null if entry not found
- */
-function parseXmlFromZip(zipPath, entryPath) {
-  const zip = new AdmZip(zipPath);
-  const entry = zip.getEntry(entryPath);
-
-  if (!entry) {
-    return null;
-  }
-
-  const xmlString = entry.getData().toString('utf8');
-  return parseXml(xmlString);
-}
-
-/**
- * Get raw XML string from a ZIP archive
- * @param {string|Buffer} zipPath - Path to ZIP file or Buffer
- * @param {string} entryPath - Path to XML file within ZIP
- * @returns {string|null} XML string or null if entry not found
- */
-function getXmlStringFromZip(zipPath, entryPath) {
-  const zip = new AdmZip(zipPath);
-  const entry = zip.getEntry(entryPath);
-
-  if (!entry) {
-    return null;
-  }
-
-  return entry.getData().toString('utf8');
-}
-
-/**
- * List all entries in a ZIP file
- * @param {string|Buffer} zipPath - Path to ZIP file or Buffer
- * @returns {string[]} Array of entry paths
- */
-function listZipEntries(zipPath) {
-  const zip = new AdmZip(zipPath);
-  return zip.getEntries().map(e => e.entryName);
-}
-
-/**
- * Ensure value is an array (handles single item vs array in XML parsing)
- * @param {*} value - Value that might be an array or single item
- * @returns {Array} Always returns an array
- */
-function ensureArray(value) {
-  if (value === undefined || value === null) return [];
-  return Array.isArray(value) ? value : [value];
-}
-
-/**
- * Safely get nested property from object
- * @param {object} obj - Object to traverse
- * @param {string} path - Dot-separated path (e.g., 'worksheet.sheetData.row')
- * @returns {*} Value at path or undefined
- */
-function getNestedValue(obj, path) {
-  return path.split('.').reduce((current, key) => {
-    return current && current[key] !== undefined ? current[key] : undefined;
-  }, obj);
-}
-
-/**
- * Extract text content from a parsed XML text node
- * Handles both simple strings and objects with #text
- * @param {*} textNode - Text node from parsed XML
- * @returns {string} Extracted text
- */
-function extractTextContent(textNode) {
-  if (typeof textNode === 'string') return textNode;
-  if (typeof textNode === 'number') return String(textNode);
-  if (textNode && typeof textNode === 'object') {
-    if (textNode['#text'] !== undefined) return String(textNode['#text']);
-  }
-  return '';
-}
-
-module.exports = {
-  parseXml,
-  parseXmlFromZip,
-  getXmlStringFromZip,
-  listZipEntries,
-  ensureArray,
-  getNestedValue,
-  extractTextContent,
-};