officeparser 7.2.1 → 7.2.2

package/dist/utils/errorUtils.js

@@ -30,7 +30,11 @@ const ERROR_MESSAGES = {
     [types_js_1.OfficeErrorType.INVALID_SELECTOR]: (selector) => `Invalid selector: ${selector}`,
     [types_js_1.OfficeErrorType.INVALID_OUTPUT_MAPPING]: (output) => `Invalid output mapping: ${output}`,
     [types_js_1.OfficeErrorType.MISSING_EMBEDDING_FUNCTION]: `Semantic chunking requires an "embeddingFunction" to be provided in chunksConfig. This function must accept a string and return a Promise resolving to a number array (vector).`,
-    [types_js_1.OfficeErrorType.OPERATION_ABORTED]: `The operation was aborted.`
+    [types_js_1.OfficeErrorType.OPERATION_ABORTED]: `The operation was aborted.`,
+    [types_js_1.OfficeErrorType.ZIP_ENTRY_COUNT_LIMIT_EXCEEDED]: (limit) => `ZIP entry count exceeds limit (${limit})`,
+    [types_js_1.OfficeErrorType.ZIP_ENTRY_INVALID_SIZE]: `ZIP entry missing a valid declared size`,
+    [types_js_1.OfficeErrorType.ZIP_SIZE_LIMIT_EXCEEDED]: (limit) => `ZIP uncompressed size limit exceeded (${limit} bytes)`,
+    [types_js_1.OfficeErrorType.EMBEDDING_TIMEOUT]: (timeout) => `Embedding call timed out after ${timeout}ms`
 };
 /**
  * Lookup table for warning messages.

package/dist/utils/zipUtils.d.ts

@@ -13,6 +13,7 @@
  *
  * @module zipUtils
  */
+import { DecompressionLimits } from '../types.js';
 /**
  * Represents a file extracted from a ZIP archive.
  * Contains the file's path within the archive and its content as a Buffer.
@@ -69,5 +70,5 @@ interface ZipFileContent {
  *
  * @see https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT ZIP file format specification
  */
-export declare const extractFiles: (zipInput: Buffer, filterFn: (fileName: string) => boolean) => Promise<ZipFileContent[]>;
+export declare const extractFiles: (zipInput: Buffer, filterFn: (fileName: string) => boolean, limits: DecompressionLimits) => Promise<ZipFileContent[]>;
 export {};

package/dist/utils/zipUtils.js

@@ -17,6 +17,8 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.extractFiles = void 0;
 const fflate_1 = require("fflate");
+const types_js_1 = require("../types.js");
+const errorUtils_js_1 = require("./errorUtils.js");
 /**
  * Extracts files from a ZIP archive with optional filtering.
  *
@@ -56,9 +58,39 @@ const fflate_1 = require("fflate");
  *
  * @see https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT ZIP file format specification
  */
-const extractFiles = (zipInput, filterFn) => {
+const extractFiles = (zipInput, filterFn, limits) => {
+    const maxUncompressedBytes = limits?.maxUncompressedBytes !== undefined && Number.isFinite(limits.maxUncompressedBytes) && limits.maxUncompressedBytes >= 0
+        ? limits.maxUncompressedBytes
+        : 512 * 1024 * 1024;
+    const maxZipEntries = limits?.maxZipEntries !== undefined && Number.isFinite(limits.maxZipEntries) && limits.maxZipEntries >= 0
+        ? limits.maxZipEntries
+        : 10000;
     return new Promise((resolve, reject) => {
-        (0, fflate_1.unzip)(new Uint8Array(zipInput.buffer, zipInput.byteOffset, zipInput.byteLength), { filter: (file) => filterFn(file.name) }, (err, decompressed) => {
+        let totalEntryCount = 0;
+        let entryCount = 0;
+        let declaredTotal = 0;
+        (0, fflate_1.unzip)(new Uint8Array(zipInput.buffer, zipInput.byteOffset, zipInput.byteLength), {
+            filter: (file) => {
+                totalEntryCount++;
+                if (totalEntryCount > maxZipEntries) {
+                    reject((0, errorUtils_js_1.getOfficeError)(types_js_1.OfficeErrorType.ZIP_ENTRY_COUNT_LIMIT_EXCEEDED, undefined, maxZipEntries));
+                    return false;
+                }
+                if (!filterFn(file.name))
+                    return false;
+                if (typeof file.originalSize !== 'number' || !Number.isFinite(file.originalSize) || file.originalSize < 0) {
+                    reject((0, errorUtils_js_1.getOfficeError)(types_js_1.OfficeErrorType.ZIP_ENTRY_INVALID_SIZE));
+                    return false;
+                }
+                entryCount++;
+                declaredTotal += file.originalSize;
+                if (declaredTotal > maxUncompressedBytes) {
+                    reject((0, errorUtils_js_1.getOfficeError)(types_js_1.OfficeErrorType.ZIP_SIZE_LIMIT_EXCEEDED, undefined, maxUncompressedBytes));
+                    return false;
+                }
+                return true;
+            }
+        }, (err, decompressed) => {
             if (err)
                 return reject(err);
             resolve(Object.entries(decompressed).map(([path, data]) => ({

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "officeparser",
-  "version": "7.2.1",
+  "version": "7.2.2",
   "description": "A robust, strictly-typed Node.js and Browser library for parsing office files (.docx, .pptx, .xlsx, .odt, .odp, .ods, .pdf, .rtf, .csv, .md, .html) and generating high-fidelity outputs in Markdown, HTML, CSV, RTF, and RAG-focused chunks.",
   "funding": "https://github.com/sponsors/harshankur",
   "main": "dist/index.js",