🫙 fix: Surface Document Extraction Failures (#15651)

Co-authored-by: aeyeopsdev <275853971+aeyeopsdev@users.noreply.github.com>

Author: aeyeopsdev

api/server/services/Files/process.js

@@ -889,6 +889,7 @@ const processAgentFileUpload = async ({ req, res, metadata, sseStream }) => {
           `[processAgentFileUpload] Document parser failed for ${extractionFileLabel}:`,
           errorMetadata,
         );
+        throw err;
       }
     };
 

api/server/services/Files/process.spec.js

@@ -622,7 +622,7 @@ describe('processAgentFileUpload', () => {
 
       await expect(
         processAgentFileUpload({ req, res: mockRes, metadata: makeMetadata() }),
-      ).rejects.toThrow(/image-based and requires an OCR service/);
+      ).rejects.toThrow('No text found in document');
 
       expect(parseText).not.toHaveBeenCalled();
     });
@@ -632,8 +632,11 @@ describe('processAgentFileUpload', () => {
         handleFileUpload: jest.fn().mockRejectedValue(new Error('PRIVATE parser failure')),
       });
       extractInspectableFileText.mockImplementationOnce(async ({ extract }) => {
-        await extract();
-        throw makeUninspectableExtractedTextError();
+        try {
+          await extract();
+        } catch {
+          throw makeUninspectableExtractedTextError();
+        }
       });
       const req = makeReq({
         mimetype: PDF_MIME,
@@ -771,7 +774,7 @@ describe('processAgentFileUpload', () => {
 
       await expect(
         processAgentFileUpload({ req, res: mockRes, metadata: makeMetadata() }),
-      ).rejects.toThrow(/image-based and requires an OCR service/);
+      ).rejects.toThrow('failure');
 
       expect(parseText).not.toHaveBeenCalled();
     });
@@ -862,8 +865,11 @@ describe('processAgentFileUpload', () => {
         handleFileUpload: jest.fn().mockRejectedValue(new Error('PRIVATE parser failure')),
       });
       extractInspectableFileText.mockImplementationOnce(async ({ extract }) => {
-        await extract();
-        throw makeUninspectableExtractedTextError();
+        try {
+          await extract();
+        } catch {
+          throw makeUninspectableExtractedTextError();
+        }
       });
       const req = makeReq({
         mimetype: DOCX_MIME,

packages/api/src/utils/files.ts

@@ -7,6 +7,12 @@ const USER_FACING_UPLOAD_ERRORS = [
   ['Invalid file format', 'Invalid file format'],
   ['exceeds token limit', 'File content exceeds token limit'],
   ['Unable to extract text from', 'Unable to extract text from file'],
+  ['No text found in document', 'No text found in document'],
+  ['MB document parser limit', 'File exceeds the document parser size limit'],
+  ['MB per-entry decompressed cap', 'Document entry exceeds the decompressed size limit'],
+  ['total decompressed size exceeds the', 'Document exceeds the total decompressed size limit'],
+  ['MB decompressed limit', 'Document exceeds the decompressed size limit'],
+  ['MB storage limit', 'Extracted text exceeds the storage size limit'],
 ] as const;
 
 const ASCII_FILENAME_SAFE_PATTERN = /^[a-zA-Z0-9._-]$/;