Add opt-in minimum list indentation for Python-Markdown compatibility

README.rst

@@ -79,6 +79,14 @@ bullets
   lists are nested. Otherwise, the bullet will alternate based on nesting
   level. Defaults to ``'*+-'``.
 
+list_indent_width
+  Minimum width, in spaces, of a list item's marker and following whitespace.
+  Defaults to ``0``, which uses the marker's natural width. Set to ``4`` to
+  preserve nested lists when the output is parsed by Python-Markdown.
+  Extra spaces are added after the marker, and continuation lines use the
+  same indentation. Markers wider than this minimum are never truncated.
+  Available as ``--list-indent-width`` on the command line.
+
 strong_em_symbol
   In markdown, both ``*`` and ``_`` are used to encode **strong** or
   *emphasized* texts. Either of these symbols can be chosen by the options

markdownify/__init__.py

@@ -187,6 +187,7 @@ class DefaultOptions:
         escape_misc = False
         heading_style = UNDERLINED
         keep_inline_images_in = []
+        list_indent_width = 0
         newline_style = SPACES
         strip = None
         strip_document = STRIP
@@ -647,7 +648,8 @@ def convert_li(self, el, text, parent_tags):
                 el = el.parent
             bullets = self.options['bullets']
             bullet = bullets[depth % len(bullets)]
-        bullet = bullet + ' '
+        # Pad the marker and continuation lines to the requested minimum width.
+        bullet = (bullet + ' ').ljust(self.options['list_indent_width'])
         bullet_width = len(bullet)
         bullet_indent = ' ' * bullet_width
 

markdownify/__init__.pyi

@@ -29,6 +29,7 @@ def markdownify(
     escape_misc: bool = ...,
     heading_style: str = ...,
     keep_inline_images_in: list[str] = ...,
+    list_indent_width: int = ...,
     newline_style: str = ...,
     strip: Union[list[str], None] = ...,
     strip_document: Union[str, None] = ...,
@@ -57,6 +58,7 @@ class MarkdownConverter:
         escape_misc: bool = ...,
         heading_style: str = ...,
         keep_inline_images_in: list[str] = ...,
+        list_indent_width: int = ...,
         newline_style: str = ...,
         strip: Union[list[str], None] = ...,
         strip_document: Union[str, None] = ...,

markdownify/main.py

@@ -35,6 +35,9 @@ def main(argv=sys.argv[1:]):
     parser.add_argument('-b', '--bullets', default='*+-',
                         help="A string of bullet styles to use; the bullet will "
                         "alternate based on nesting level.")
+    parser.add_argument('--list-indent-width', type=int, default=0,
+                        help="Minimum list item indentation. Use 4 for compatibility "
+                        "with Python-Markdown; 0 uses the marker width.")
     parser.add_argument('--strong-em-symbol', default=ASTERISK,
                         choices=(ASTERISK, UNDERSCORE),
                         help="Use * or _ to convert strong and italics text"),

tests/test_lists.py

@@ -1,3 +1,8 @@
+import subprocess
+import sys
+
+import pytest
+
 from .utils import md
 
 
@@ -87,3 +92,39 @@ def test_bullets():
 
 def test_li_text():
     assert md('<ul><li>foo <a href="#">bar</a></li><li>foo bar  </li><li>foo <b>bar</b>   <i>space</i>.</ul>') == '\n\n* foo [bar](#)\n* foo bar\n* foo **bar** *space*.\n'
+
+
+@pytest.mark.parametrize('html, expected', [
+    ('<ul><li>a<ul><li>b<ul><li>c<ul><li>d</li></ul></li></ul></li></ul></li></ul>',
+     '\n\n*   a\n    +   b\n        -   c\n            *   d\n'),
+    ('<ol><li>a<ol><li>b<ol><li>c<ol><li>d</li></ol></li></ol></li></ol></li></ol>',
+     '\n\n1.  a\n    1.  b\n        1.  c\n            1.  d\n'),
+    ('<ul><li><p>first</p><p>second</p><blockquote>quote</blockquote></li></ul>',
+     '\n\n*   first\n\n    second\n\n    > quote\n'),
+    ('<ul><li>a<ol><li>b</li></ol></li><li>c</li></ul>',
+     '\n\n*   a\n    1.  b\n*   c\n'),
+    ('<ol start="1234"><li><p>first</p><p>second</p></li></ol>',
+     '\n\n1234. first\n\n      second\n'),
+])
+def test_list_indent_width(html, expected):
+    assert md(html, list_indent_width=4) == expected
+
+
+def test_list_indent_width_minimum():
+    html = '<ol start="10"><li>a<ul><li>b</li></ul></li></ol>'
+    assert md(html, list_indent_width=0) == md(html)
+    assert md(html, list_indent_width=1) == md(html)
+
+
+@pytest.mark.parametrize('use_file', [False, True])
+def test_list_indent_width_cli(tmp_path, use_file):
+    html = '<ul><li>a<ul><li>b</li></ul></li></ul>'
+    command = [sys.executable, '-m', 'markdownify.main', '--list-indent-width', '4']
+    if use_file:
+        path = tmp_path / 'nested.html'
+        path.write_text(html, encoding='utf-8')
+        command.append(str(path))
+    result = subprocess.run(command, input='' if use_file else html,
+                            text=True, capture_output=True)
+    assert result.returncode == 0, result.stderr
+    assert result.stdout == '*   a\n    +   b\n'

tests/types.py

@@ -22,6 +22,7 @@
     escape_misc=False,
     heading_style=UNDERLINED,
     keep_inline_images_in=[],
+    list_indent_width=0,
     newline_style=SPACES,
     strip=None,
     strip_document=STRIP,
@@ -47,6 +48,7 @@
     sup_symbol='^',
     sub_symbol='^',
     keep_inline_images_in=['h3'],
+    list_indent_width=4,
     wrap=True,
     wrap_width=80,
     strong_em_symbol=UNDERSCORE,
@@ -68,3 +70,4 @@ def callback(el: BeautifulSoup) -> Union[str, None]:
 
 markdownify('<pre class="python">test\n    foo\nbar</pre>', code_language_callback=callback)
 markdownify('<pre class="python">test\n    foo\nbar</pre>', code_language_callback=lambda el: None)
+markdownify('<ul><li>item</li></ul>', list_indent_width=4)