fix: pdf loader

1d1c56b5 · John Wang · 049ea8e7 · 1d1c56b5
Commit 1d1c56b5 authored Jun 20, 2023 by John Wang
Hide whitespace changes
Inline Side-by-side

Showing with 2 additions and 1 deletion

pdf.py api/core/data_loader/loader/pdf.py +2 -1

No files found.
--- a/api/core/data_loader/loader/pdf.py
+++ b/api/core/data_loader/loader/pdf.py
@@ -38,7 +38,8 @@ class PdfLoader(BaseLoader):
                try:
                    text = storage.load(plaintext_file_key).decode('utf-8')
                    plaintext_file_exists = True
-                    return text
+                    metadata = {"source": self._file_path}
+                    return [Document(page_content=text, metadata=metadata)]
                except FileNotFoundError:
                    pass