feat: Preserve HTML data attributes (from spans to anchors)

Issue-#41: #41

Author: pawamoy

src/mkdocs_autorefs/references.py

@@ -16,8 +16,8 @@
     from markdown import Markdown
 
 AUTO_REF_RE = re.compile(
-    r"<span data-(?P<kind>autorefs-identifier|autorefs-optional|autorefs-optional-hover)="
-    r'("?)(?P<identifier>[^"<>]*)\2>(?P<title>.*?)</span>',
+    r"<span data-(?P<kind>autorefs-(?:identifier|optional|optional-hover))="
+    r'("?)(?P<identifier>[^"<>]+)\2(?P<attrs> [^<>]+)?>(?P<title>.*?)</span>',
     flags=re.DOTALL,
 )
 """A regular expression to match mkdocs-autorefs' special reference markers
@@ -157,6 +157,7 @@ def inner(match: Match) -> str:
         identifier = match["identifier"]
         title = match["title"]
         kind = match["kind"]
+        attrs = match["attrs"] or ""
 
         try:
             url = url_mapper(unescape(identifier))
@@ -175,8 +176,8 @@ def inner(match: Match) -> str:
         classes = ["autorefs", "autorefs-external" if external else "autorefs-internal"]
         class_attr = " ".join(classes)
         if kind == "autorefs-optional-hover":
-            return f'<a class="{class_attr}" title="{identifier}" href="{escape(url)}">{title}</a>'
-        return f'<a class="{class_attr}" href="{escape(url)}">{title}</a>'
+            return f'<a class="{class_attr}" title="{identifier}" href="{escape(url)}"{attrs}>{title}</a>'
+        return f'<a class="{class_attr}" href="{escape(url)}"{attrs}>{title}</a>'
 
     return inner
 

tests/test_references.py

@@ -224,3 +224,11 @@ def test_external_references() -> None:
     output, unmapped = fix_refs(source, url_map.__getitem__)
     assert output == '<a class="autorefs autorefs-external" href="https://example.com">example</a>'
     assert unmapped == []
+
+
+def test_keep_data_attributes() -> None:
+    """Keep HTML data attributes from autorefs spans."""
+    url_map = {"example": "https://e.com"}
+    source = '<span data-autorefs-optional="example" data-foo data-bar="0">e</span>'
+    output, _ = fix_refs(source, url_map.__getitem__)
+    assert output == '<a class="autorefs autorefs-external" href="https://e.com" data-foo data-bar="0">e</a>'