fix: support extended markdown

disnake/utils.py

@@ -10,6 +10,7 @@
 import os
 import pkgutil
 import re
+import string
 import sys
 import unicodedata
 import warnings
@@ -776,15 +777,16 @@ def resolve_template(code: Union[Template, str]) -> str:
     r"\{0}(?=([\s\S]*((?<!\{0})\{0})))".format(c) for c in ("*", "`", "_", "~", "|")
 )
 
-_MARKDOWN_ESCAPE_COMMON = r"^>(?:>>)?\s|\[.+\]\(.+\)"
+# todo: fix issue with lists including the preceeding spaces more sustainably
+_MARKDOWN_ESCAPE_COMMON = r"^>(?:>>)?\s|^#{1,3}\s|^(?P<keep>\s*?)?(?P<char>[-*])\s|\[.+\]\(.+\)"
 
 _MARKDOWN_ESCAPE_REGEX = re.compile(
     rf"(?P<markdown>{_MARKDOWN_ESCAPE_SUBREGEX}|{_MARKDOWN_ESCAPE_COMMON})", re.MULTILINE
 )
 
 _URL_REGEX = r"(?P<url><[^: >]+:\/[^ >]+>|(?:https?|steam):\/\/[^\s<]+[^<.,:;\"\'\]\s])"
 
-_MARKDOWN_STOCK_REGEX = rf"(?P<markdown>[_\\~|\*`#\-]|{_MARKDOWN_ESCAPE_COMMON})"
+_MARKDOWN_STOCK_REGEX = rf"(?P<markdown>[_\\~|\*`]|{_MARKDOWN_ESCAPE_COMMON})"
 
 
 def remove_markdown(text: str, *, ignore_links: bool = True) -> str:
@@ -813,7 +815,7 @@ def remove_markdown(text: str, *, ignore_links: bool = True) -> str:
 
     def replacement(match):
         groupdict = match.groupdict()
-        return groupdict.get("url", "")
+        return groupdict.get("url", "") or groupdict.get("keep", "")
 
     regex = _MARKDOWN_STOCK_REGEX
     if ignore_links:
@@ -847,12 +849,12 @@ def escape_markdown(text: str, *, as_needed: bool = False, ignore_links: bool =
     """
     if not as_needed:
 
-        def replacement(match):
+        def replacement(match: re.Match[str]) -> str:
             groupdict = match.groupdict()
             is_url = groupdict.get("url")
             if is_url:
                 return is_url
-            return "\\" + groupdict["markdown"]
+            return "".join(c if c in string.whitespace else "\\" + c for c in groupdict["markdown"])
 
         regex = _MARKDOWN_STOCK_REGEX
         if ignore_links:

tests/test_utils.py

@@ -515,6 +515,64 @@ def test_markdown_links(text: str, expected, expected_ignore) -> None:
     assert utils.remove_markdown(text, ignore_links=True) == expected_ignore
 
 
+@pytest.mark.parametrize(
+    ("text", "expected"),
+    [
+        (
+            r"# disnake",
+            r"disnake",
+        ),
+        (
+            r"## woah this should be removed\n  ### but not this one",
+            r"woah this should be removed\n  ### but not this one",
+        ),
+        (
+            r"""Inside is a long list of why markdown is an amazing tool
+- markdown supports lists
+ - honestly its a great tool that markdown supports said lists
+   - this is wrong but uh we'll get to that
+""",
+            r"""Inside is a long list of why markdown is an amazing tool
+markdown supports lists
+ honestly its a great tool that markdown supports said lists
+   this is wrong but uh we'll get to that
+""",
+        ),
+    ],
+)
+def test_remove_markdown(text: str, expected: str) -> None:
+    assert utils.remove_markdown(text) == expected
+
+
+@pytest.mark.parametrize(
+    ("text", "expected"),
+    [
+        (
+            r"# disnake",
+            r"\# disnake",
+        ),
+        (
+            "## escape these quotes\n  ### but not these",
+            "\\#\\# escape these quotes\n  ### but not these",
+        ),
+        (
+            r"""Inside is a long list of why markdown is an amazing tool
+- markdown supports lists
+ - honestly its a great tool that markdown supports said lists
+   - this is wrong but uh we'll get to that
+""",
+            r"""Inside is a long list of why markdown is an amazing tool
+\- markdown supports lists
+ \- honestly its a great tool that markdown supports said lists
+   \- this is wrong but uh we'll get to that
+""",
+        ),
+    ],
+)
+def test_escape_markdown(text: str, expected: str) -> None:
+    assert utils.escape_markdown(text) == expected
+
+
 @pytest.mark.parametrize(
     ("text", "expected"),
     [