From 25c681328268af038e6246b533c5fba16fdd4c3e Mon Sep 17 00:00:00 2001
From: Michele Dolfi <dol@zurich.ibm.com>
Date: Fri, 31 Jan 2025 15:24:27 +0100
Subject: [PATCH] make mathml the default

Signed-off-by: Michele Dolfi <dol@zurich.ibm.com>
---
 docling_core/types/doc/document.py | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/docling_core/types/doc/document.py b/docling_core/types/doc/document.py
index eb56e7cc..a3616a4a 100644
--- a/docling_core/types/doc/document.py
+++ b/docling_core/types/doc/document.py
@@ -2288,7 +2288,7 @@ def save_as_html(
         to_element: int = sys.maxsize,
         labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
         image_mode: ImageRefMode = ImageRefMode.PLACEHOLDER,
-        formula_to_mathml: bool = False,
+        formula_to_mathml: bool = True,
         page_no: Optional[int] = None,
         html_lang: str = "en",
         html_head: str = _HTML_DEFAULT_HEAD,
@@ -2355,7 +2355,7 @@ def export_to_html(  # noqa: C901
         to_element: int = sys.maxsize,
         labels: set[DocItemLabel] = DEFAULT_EXPORT_LABELS,
         image_mode: ImageRefMode = ImageRefMode.PLACEHOLDER,
-        formula_to_mathml: bool = False,
+        formula_to_mathml: bool = True,
         page_no: Optional[int] = None,
         html_lang: str = "en",
         html_head: str = _HTML_DEFAULT_HEAD,
@@ -2479,7 +2479,13 @@ def _prepare_tag_content(
                     annotation.text = math_formula
                     mathml = unescape(tostring(mathml_element, encoding="unicode"))
                     text = f"<div>{mathml}</div>"
-                else:
+
+                elif (
+                    item.text == ""
+                    and item.orig != ""
+                    and image_mode == ImageRefMode.EMBEDDED
+                ):
+
                     text = f"<pre>{math_formula}</pre>"
                 html_texts.append(text)