diff --git a/markdown_it/rules_block/html_block.py b/markdown_it/rules_block/html_block.py
index fe7e464e..8b195933 100644
--- a/markdown_it/rules_block/html_block.py
+++ b/markdown_it/rules_block/html_block.py
@@ -14,16 +14,21 @@
# last argument defines whether it can terminate a paragraph or not
HTML_SEQUENCES: list[tuple[re.Pattern[str], re.Pattern[str], bool]] = [
(
- re.compile(r"^<(script|pre|style|textarea)(?=(\s|>|$))", re.IGNORECASE),
- re.compile(r"<\/(script|pre|style|textarea)>", re.IGNORECASE),
+ re.compile(
+ r"^<(script|pre|style|textarea)(?=(\s|>|$))", re.IGNORECASE | re.ASCII
+ ),
+ re.compile(r"<\/(script|pre|style|textarea)>", re.IGNORECASE | re.ASCII),
True,
),
(re.compile(r"^"), True),
(re.compile(r"^<\?"), re.compile(r"\?>"), True),
- (re.compile(r"^"), True),
+ (re.compile(r"^"), True),
(re.compile(r"^"), True),
(
- re.compile("^|$))", re.IGNORECASE),
+ re.compile(
+ "^|$))",
+ re.IGNORECASE | re.ASCII,
+ ),
re.compile(r"^$"),
True,
),
diff --git a/tests/test_html_block_ascii.py b/tests/test_html_block_ascii.py
new file mode 100644
index 00000000..7293180d
--- /dev/null
+++ b/tests/test_html_block_ascii.py
@@ -0,0 +1,24 @@
+"""CommonMark HTML block starts use ASCII letters for tag and declarations."""
+
+from markdown_it import MarkdownIt
+
+
+def test_ascii_declaration_ends_paragraph() -> None:
+ md = MarkdownIt("commonmark")
+ for declaration in ("", ""):
+ assert md.render(f"para\n{declaration}\n") == f"para
\n{declaration}\n"
+
+
+def test_unicode_casefold_does_not_start_html_block() -> None:
+ md = MarkdownIt("commonmark")
+ for tag in ("\u017fcript", "\u017ftyle", "d\u0130v", "d\u0131v"):
+ assert md.render(f"para\n<{tag}>\ntext\n") == (
+ f"para\n<{tag}>\ntext
\n"
+ )
+
+
+def test_ascii_mixed_case_tag_still_starts_html_block() -> None:
+ md = MarkdownIt("commonmark")
+ assert md.render("para\n\n") == (
+ "para
\n\n"
+ )