fix(markdown): fix handling of ordered lists (#175)

Signed-off-by: Panos Vagenas <[email protected]>

Author: vagenas

docling_core/types/doc/document.py

@@ -2489,11 +2489,16 @@ def _ingest_text(text: str, do_escape_html=True, do_escape_underscores=True):
                     )
                     # NOTE: assumes unordered (flag & marker currently in ListItem)
                     indent_str = list_level * indent * " "
+                    is_ol = item.label == GroupLabel.ORDERED_LIST
                     text = "\n".join(
                         [
                             # avoid additional marker on already evaled sublists
-                            cpt if cpt and cpt[0] == " " else f"{indent_str}- {cpt}"
-                            for cpt in comps
+                            (
+                                c
+                                if c and c[0] == " "
+                                else f"{indent_str}{f'{i + 1}.' if is_ol else '-'} {c}"
+                            )
+                            for i, c in enumerate(comps)
                         ]
                     )
                     _ingest_text(text=text)

test/data/doc/constructed_doc.embedded.md.gt

@@ -15,10 +15,10 @@ This paper introduces the biggest invention ever made. ...
 - list item 1
 - list item 2
 - list item 3
-    - list item 3.a
-    - list item 3.b
-    - list item 3.c
-        - list item 3.c.i
+    1. list item 3.a
+    2. list item 3.b
+    3. list item 3.c
+        1. list item 3.c.i
 - list item 4
 
 This is the caption of table 1.

test/data/doc/constructed_doc.placeholder.md.gt

@@ -15,10 +15,10 @@ This paper introduces the biggest invention ever made. ...
 - list item 1
 - list item 2
 - list item 3
-    - list item 3.a
-    - list item 3.b
-    - list item 3.c
-        - list item 3.c.i
+    1. list item 3.a
+    2. list item 3.b
+    3. list item 3.c
+        1. list item 3.c.i
 - list item 4
 
 This is the caption of table 1.

test/data/doc/constructed_doc.referenced.md.gt

@@ -15,10 +15,10 @@ This paper introduces the biggest invention ever made. ...
 - list item 1
 - list item 2
 - list item 3
-    - list item 3.a
-    - list item 3.b
-    - list item 3.c
-        - list item 3.c.i
+    1. list item 3.a
+    2. list item 3.b
+    3. list item 3.c
+        1. list item 3.c.i
 - list item 4
 
 This is the caption of table 1.

test/data/doc/constructed_document.yaml.md

@@ -15,10 +15,10 @@ This paper introduces the biggest invention ever made. ...
 - list item 1
 - list item 2
 - list item 3
-    - list item 3.a
-    - list item 3.b
-    - list item 3.c
-        - list item 3.c.i
+    1. list item 3.a
+    2. list item 3.b
+    3. list item 3.c
+        1. list item 3.c.i
 - list item 4
 
 This is the caption of table 1.