Fix whitespace inside tags

VikParuchuri · Jan 15, 2025 · 82b97ca · 82b97ca
1 parent 81fd95e
commit 82b97ca
Showing 1 changed file with 5 additions and 4 deletions.
diff --git a/marker/renderers/__init__.py b/marker/renderers/__init__.py
@@ -47,7 +47,11 @@ def merge_consecutive_tags(html, tag):
             return html
 
         def replace_whitespace(match):
-            return match.group(1)
+            whitespace = match.group(1)
+            if len(whitespace) == 0:
+                return ""
+            else:
+                return " "
 
         pattern = fr'</{tag}>(\s*)<{tag}>'
 
@@ -57,9 +61,6 @@ def replace_whitespace(match):
                 break
             html = new_merged
 
-        # Replace consecutive whitespace
-        html = re.sub(r'\s+', ' ', html)
-
         return html
 
     def generate_page_stats(self, document: Document, document_output):