fix: only strip the leading bullet from a ListItem first line

marker/schema/blocks/listitem.py

@@ -13,7 +13,7 @@ def replace_bullets(child_blocks):
 
     if first_block is not None and first_block.id.block_type == BlockTypes.Line:
         bullet_pattern = r"(^|[\n ]|<[^>]*>)[•●○ഠ ം◦■▪▫–—-]( )"
-        first_block.html = re.sub(bullet_pattern, r"\1\2", first_block.html)
+        first_block.html = re.sub(bullet_pattern, r"\1\2", first_block.html, count=1)
 
 
 class ListItem(Block):

tests/schema/blocks/test_listitem.py

@@ -0,0 +1,18 @@
+from types import SimpleNamespace
+
+from marker.schema import BlockTypes
+from marker.schema.blocks.listitem import replace_bullets
+
+
+def test_replace_bullets_keeps_inner_dashes():
+    # Regression for #1024: removing the bullet must not drop a dash inside the text.
+    line = SimpleNamespace(
+        children=[],
+        id=SimpleNamespace(block_type=BlockTypes.Line),
+        html="• He paused — then left.",
+    )
+
+    replace_bullets([line])
+
+    assert "•" not in line.html
+    assert line.html.strip() == "He paused — then left."