Fix whitespace inside tags

VikParuchuri · Jan 15, 2025 · 7678463 · 7678463
1 parent 81fd95e
commit 7678463
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 7 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):

diff --git a/marker/renderers/html.py b/marker/renderers/html.py
@@ -78,12 +78,10 @@ def extract_html(self, document, document_output, level=0):
                 images.update(sub_images)
                 ref.replace_with(BeautifulSoup(f"{content}", 'html.parser'))
 
+        output = str(soup)
         if level == 0:
-            output = soup.prettify()
             output = self.merge_consecutive_tags(output, 'b')
             output = self.merge_consecutive_tags(output, 'i')
-        else:
-            output = str(soup)
 
         return output, images