🛠️ fix(parser): preserve link titles and wrap list parsers

Author: philippb

sdiff/parser.py

@@ -123,13 +123,17 @@ def flush_buffer():
             elif token_type == 'link':
                 flush_buffer()
                 text = self._flatten_inline_text(token.get('children', []))
-                url = token.get('attrs', {}).get('url', '')
-                nodes.append(Link(f"[{text}]({url})"))
+                attrs = token.get('attrs', {})
+                url = attrs.get('url', '')
+                title = attrs.get('title')
+                nodes.append(Link(_format_link_markup(text, url, title)))
             elif token_type == 'image':
                 flush_buffer()
                 alt = token.get('attrs', {}).get('alt') or self._flatten_inline_text(token.get('children', []))
-                url = token.get('attrs', {}).get('url', '')
-                nodes.append(Image(f"![{alt}]({url})"))
+                attrs = token.get('attrs', {})
+                url = attrs.get('url', '')
+                title = attrs.get('title')
+                nodes.append(Image(_format_image_markup(alt, url, title)))
             else:
                 flush_buffer()
                 children = token.get('children', [])
@@ -290,6 +294,21 @@ def _append_text(nodes, text):
         nodes.append(Text(text))
 
 
+def _format_title(title: str) -> str:
+    if title is None:
+        return ''
+    escaped = title.replace('"', '\\"')
+    return f' "{escaped}"'
+
+
+def _format_link_markup(text: str, url: str, title: str | None) -> str:
+    return f'[{text}]({url}{_format_title(title)})'
+
+
+def _format_image_markup(alt: str, url: str, title: str | None) -> str:
+    return f'![{alt}]({url}{_format_title(title)})'
+
+
 def _is_block_html(raw: str) -> bool:
     stripped = raw.lstrip()
     if stripped.startswith('<!--'):
@@ -363,4 +382,7 @@ def parse(text, parser_cls: type[MdParser] = MdParser):
     parser = parser_cls()
     if hasattr(parser, '_set_reference_definitions'):
         parser._set_reference_definitions(reference_definitions)
-    return parser.parse(text)
+    result = parser.parse(text)
+    if isinstance(result, list):
+        return Root(result)
+    return result

tests/test_parser.py

@@ -1,6 +1,6 @@
 from unittest import TestCase
 from sdiff import parser, MdParser, ZendeskHelpMdParser
-from sdiff.model import ZendeskHelpSteps
+from sdiff.model import Paragraph, Root, Text, ZendeskHelpSteps
 
 
 class ParserTestCase(TestCase):
@@ -76,6 +76,14 @@ def test_link_label_with_codespan(self):
         actual = self._parse('[use `foo`](url)')
         self.assertEqual('[use `foo`](url)', actual.nodes[0].nodes[0].text)
 
+    def test_link_title_preserved(self):
+        actual = self._parse('[label](https://example.com "Title Here")')
+        self.assertEqual('[label](https://example.com "Title Here")', actual.nodes[0].nodes[0].text)
+
+    def test_image_title_preserved(self):
+        actual = self._parse('![alt](https://img "Img Title")')
+        self.assertEqual('![alt](https://img "Img Title")', actual.nodes[0].nodes[0].text)
+
     def test_reference_definition_preserved(self):
         data = 'See [API][id].\n\n[id]: https://example.com'
         tree = self._parse(data)
@@ -215,3 +223,15 @@ def test_remove_ltr_rtl_marks(self):
         text = 'a\u200eb\u200f'
         actual = parser._remove_ltr_rtl_marks(text)
         self.assertEqual('ab', actual)
+
+
+class DummyParser:
+    def parse(self, text):
+        return [Paragraph([Text(text)])]
+
+
+class TestParseWrapper(TestCase):
+    def test_wraps_list_parser_output(self):
+        tree = parser.parse('hello', parser_cls=DummyParser)
+        self.assertIsInstance(tree, Root)
+        self.assertEqual('pt', tree.print_all())