LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows
Python ai-agentsdata-engineeringdocument-aigenerative-aiidpjson-extractionllmmcp-serverocrpdf-extractionprompt-engineeringstructured-output
[FIX] Updated pdf js worker to load from cdn (#1718)
vishnuszipstack09f62be01d40
diff --git a/frontend/.gitignore b/frontend/.gitignoreindex 16b4aa25e..f287e6415 100644--- a/frontend/.gitignore+++ b/frontend/.gitignore@@ -11,9 +11,6 @@# production/build-# pdfjs worker file (copied from node_modules via postinstall)-/public/pdf.worker.min.js-# misc.DS_Store.env.localdiff --git a/frontend/package.json b/frontend/package.jsonindex 8d11f6c47..535c95fd2 100644--- a/frontend/package.json+++ b/frontend/package.json@@ -57,7 +57,6 @@"zustand": "^4.3.8"},"scripts": {- "postinstall": "cp node_modules/pdfjs-dist/build/pdf.worker.min.js public/pdf.worker.min.js","start": "react-scripts start","build": "react-scripts build","test": "react-scripts test",diff --git a/frontend/src/helpers/pdfWorkerConfig.js b/frontend/src/helpers/pdfWorkerConfig.jsindex 47a972925..5c9e5f12b 100644--- a/frontend/src/helpers/pdfWorkerConfig.js+++ b/frontend/src/helpers/pdfWorkerConfig.js@@ -1,4 +1,4 @@-// Use the worker file from the public folder (served as static asset)-// This avoids Webpack processing which incorrectly bundles it with require() statements-// Note: The ?url import suffix only works in Vite, not in CRA (Webpack)-export const PDF_WORKER_URL = "/pdf.worker.min.js";+// Use CDN to load the worker file+// Version must match the installed pdfjs-dist version (3.11.174 via @react-pdf-viewer/core)+export const PDF_WORKER_URL =+ "https://unpkg.com/[email protected]/build/pdf.worker.min.js";