From 8891c6653615768996b64e463dd26eed4a6bbe18 Mon Sep 17 00:00:00 2001
From: Peter Staar <taa@zurich.ibm.com>
Date: Thu, 10 Apr 2025 13:58:59 +0200
Subject: [PATCH] updated the cli to output html in split-page mode

Signed-off-by: Peter Staar <taa@zurich.ibm.com>
---
 docling/cli/main.py              | 13 ++++++++++++-
 docling/datamodel/base_models.py |  1 +
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/docling/cli/main.py b/docling/cli/main.py
index e0f0cbd8..4f266638 100644
--- a/docling/cli/main.py
+++ b/docling/cli/main.py
@@ -153,6 +153,7 @@ def export_documents(
     output_dir: Path,
     export_json: bool,
     export_html: bool,
+    export_html_split_page: bool,
     export_md: bool,
     export_txt: bool,
     export_doctags: bool,
@@ -180,7 +181,15 @@ def export_documents(
                 fname = output_dir / f"{doc_filename}.html"
                 _log.info(f"writing HTML output to {fname}")
                 conv_res.document.save_as_html(
-                    filename=fname, image_mode=image_export_mode
+                    filename=fname, image_mode=image_export_mode, split_page_view=False
+                )
+
+            # Export HTML format:
+            if export_html_split_page:
+                fname = output_dir / f"{doc_filename}.html"
+                _log.info(f"writing HTML output to {fname}")
+                conv_res.document.save_as_html(
+                    filename=fname, image_mode=image_export_mode, split_page_view=True
                 )
 
             # Export Text format:
@@ -471,6 +480,7 @@ def convert(
 
         export_json = OutputFormat.JSON in to_formats
         export_html = OutputFormat.HTML in to_formats
+        export_html_split_page = OutputFormat.HTML_SPLIT_PAGE in to_formats
         export_md = OutputFormat.MARKDOWN in to_formats
         export_txt = OutputFormat.TEXT in to_formats
         export_doctags = OutputFormat.DOCTAGS in to_formats
@@ -578,6 +588,7 @@ def convert(
             output_dir=output,
             export_json=export_json,
             export_html=export_html,
+            export_html_split_page=export_html_split_page,
             export_md=export_md,
             export_txt=export_txt,
             export_doctags=export_doctags,
diff --git a/docling/datamodel/base_models.py b/docling/datamodel/base_models.py
index 76827a1b..b3f6c2aa 100644
--- a/docling/datamodel/base_models.py
+++ b/docling/datamodel/base_models.py
@@ -50,6 +50,7 @@ class OutputFormat(str, Enum):
     MARKDOWN = "md"
     JSON = "json"
     HTML = "html"
+    HTML_SPLIT_PAGE = "html_split_page"
     TEXT = "text"
     DOCTAGS = "doctags"