Preserve strikethrough from <strike> and CSS line-through (#2342)

* Keep strikethrough for <strike> and CSS line-through.

<s> and <del> already became ~~text~~. The obsolete strike element and inline text-decoration: line-through were flattened to plain text, so deleted wording looked current.

* Don't handle css style tags yet.

---------

Co-authored-by: afourney <adamfo@microsoft.com>

packages/markitdown/src/markitdown/converters/_markdownify.py

@@ -122,5 +122,9 @@ class _CustomMarkdownify(markdownify.MarkdownConverter):
             return "[x] " if el.has_attr("checked") else "[ ] "
         return ""
 
+    def convert_strike(self, el: Any, text: str, *args, **kwargs) -> str:
+        """Obsolete <strike> is still in the wild; treat it like <s>/<del>."""
+        return self.convert_s(el, text, *args, **kwargs)  # type: ignore
+
     def convert_soup(self, soup: Any) -> str:
         return super().convert_soup(soup)  # type: ignore

packages/markitdown/tests/test_module_misc.py

@@ -264,6 +264,44 @@ def test_docx_comments() -> None:
     validate_strings(result, DOCX_COMMENT_TEST_STRINGS)
 
 
+def test_html_strikethrough_variants(tmp_path) -> None:
+    html = """<!doctype html>
+<html><body>
+<p>Plain <s>s element</s> after.</p>
+<p>Plain <del>del element</del> after.</p>
+<p>Plain <strike>strike element</strike> after.</p>
+<p>Spaces A<strike> B </strike>C.</p>
+<p>Runs D<strike>  E  </strike>F.</p>
+<p>Empty G<strike></strike>H.</p>
+<p>Newline I<strike>J
+K</strike>L.</p>
+<p>Break M<strike>N<br>O</strike>P.</p>
+</body></html>
+"""
+    path = tmp_path / "strike.html"
+    path.write_text(html, encoding="utf-8")
+    markdown = MarkItDown().convert(str(path)).markdown
+
+    assert markdown == "\n\n".join(
+        [
+            # <s>, <del> and the obsolete <strike> all mean strikethrough
+            "Plain ~~s element~~ after.",
+            "Plain ~~del element~~ after.",
+            "Plain ~~strike element~~ after.",
+            # Surrounding whitespace stays outside of the markup ...
+            "Spaces A ~~B~~ C.",
+            # ... and runs of it collapse to a single space
+            "Runs D ~~E~~ F.",
+            # An empty element contributes nothing
+            "Empty GH.",
+            # A line break inside the element is kept, and the markup
+            # survives it because strikethrough may span a single newline
+            "Newline I~~J\nK~~L.",
+            "Break M~~N\nO~~P.",
+        ]
+    )
+
+
 def test_docx_equations() -> None:
     markitdown = MarkItDown()
     docx_file = os.path.join(TEST_FILES_DIR, "equations.docx")