From 5b9eafdf6a92c3e52b8d461304da660266bdf955 Mon Sep 17 00:00:00 2001
From: Maurizio Dipierro <maurizio.dipierro@bvtech.com>
Date: Wed, 19 Mar 2025 16:34:02 +0100
Subject: [PATCH] fix end of string

---
 app.py | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)

diff --git a/app.py b/app.py
index 10bdf28..be9ee73 100644
--- a/app.py
+++ b/app.py
@@ -116,11 +116,24 @@ def model_inference(
         buffer += html.escape(new_text)
         yield buffer
 
+    # After finishing the streamer loop:
     cleaned_output = full_output.replace("<end_of_utterance>", "").strip()
+    final_output = cleaned_output
 
-    if cleaned_output:
+    # If markdown conversion is needed, combine it with cleaned_output
+    if any(tag in cleaned_output for tag in ["<doctag>", "<otsl>", "<code>", "<chart>", "<formula>"]):
         doctag_output = cleaned_output
-        yield cleaned_output
+        if "<chart>" in doctag_output:
+            doctag_output = doctag_output.replace("<chart>", "<otsl>").replace("</chart>", "</otsl>")
+            doctag_output = re.sub(r'(<loc_500>)(?!.*<loc_500>)<[^>]+>', r'\1', doctag_output)
+        doc = DoclingDocument(name="Document")
+        doctags_doc = DocTagsDocument.from_doctags_and_image_pairs([doctag_output], images)
+        doc.load_from_doctags(doctags_doc)
+        final_output += "\n\n**MD Output:**\n\n" + doc.export_to_markdown()
+
+    # Yield the final combined output only once
+    yield final_output
+
 
     if any(tag in doctag_output for tag in ["<doctag>", "<otsl>", "<code>", "<chart>", "<formula>"]):
         doc = DoclingDocument(name="Document")