From 436d0fcca9c55e07a3e58c884e59186340ba3423 Mon Sep 17 00:00:00 2001
From: James Villarrubia <james.villarrubia@gmail.com>
Date: Wed, 12 Jun 2024 09:31:25 -0400
Subject: [PATCH] Additional header required to get style tags for visual
 processing.

---
 nlm_ingestor/file_parser/tika_parser.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/nlm_ingestor/file_parser/tika_parser.py b/nlm_ingestor/file_parser/tika_parser.py
index fb10955..97ad788 100644
--- a/nlm_ingestor/file_parser/tika_parser.py
+++ b/nlm_ingestor/file_parser/tika_parser.py
@@ -23,6 +23,7 @@ def parse_to_html(self, filepath, do_ocr=False):
                 "X-Tika-OCRskipOcr": "false",
                 "X-Tika-OCRoutputType": "hocr",
                 "X-Tika-Timeout-Millis": str(100 * timeout),
+                "X-Tika-PDFOcrStrategy": "ocr_only",
                 "X-Tika-OCRtimeoutSeconds": str(timeout),
             }