fix: Document should handle empty pageContent (fixes langchain-ai#5884)

glorat · Jun 25, 2024 · bb23668 · bb23668
1 parent a4a2d11
commit bb23668
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 4 deletions.
diff --git a/langchain-core/src/documents/document.ts b/langchain-core/src/documents/document.ts
@@ -20,18 +20,18 @@ export interface DocumentInterface<
  * Interface for interacting with a document.
  */
 export class Document<
-  // eslint-disable-next-line @typescript-eslint/no-explicit-any
-  Metadata extends Record<string, any> = Record<string, any>
+    // eslint-disable-next-line @typescript-eslint/no-explicit-any
+    Metadata extends Record<string, any> = Record<string, any>
 > implements DocumentInput, DocumentInterface
 {
   pageContent: string;
 
   metadata: Metadata;
 
   constructor(fields: DocumentInput<Metadata>) {
-    this.pageContent = fields.pageContent
+    this.pageContent = fields.pageContent !== undefined
       ? fields.pageContent.toString()
-      : this.pageContent;
+      : "";
     this.metadata = fields.metadata ?? ({} as Metadata);
   }
 }
diff --git a/langchain-core/src/tests/document.test.ts b/langchain-core/src/tests/document.test.ts
@@ -0,0 +1,7 @@
+import {test, expect} from '@jest/globals'
+import {Document} from '../documents/document.js'
+
+test('Document should handle empty pageContent', () => {
+  const doc = new Document({pageContent: ''})
+  expect(doc.pageContent).toEqual('')
+})