\", \"\", \"\", \"{inner}\")\n output.append(f\"<{tag}{start}>\\n\" + \"\\n\".join(rendered_items) + f\"\\n\")\n else: output.append(f\"{_inline(block.text, document.link_references)}
\")\n return \"\\n\".join(output) + (\"\\n\" if output else \"\")\n\"\"\"Deterministic CommonMark leaf-block parsing.\"\"\"\nfrom dataclasses import dataclass\nimport re\nfrom html.entities import html5\n\n@dataclass(frozen=True)\nclass Block:\n kind: str\n text: str = \"\"\n level: int = 0\n info: str = \"\"\n items: tuple[tuple[str, bool], ...] = ()\n tight: bool = True\n\n@dataclass(frozen=True)\nclass Document:\n blocks: tuple[Block, ...]\n link_references: dict[str, tuple[str, str | None]]\n\n_FENCE = re.compile(r\"^( {0,3})(`{3,}|~{3,})([^\\r\\n]*)$\")\n_ATX = re.compile(r\"^( {0,3})(#{1,6})(?:[ \\t]+(.*?)\\s*|[ \\t]*)$\")\n_REF = re.compile(r\"^ {0,3}\\[((?:\\\\.|[^\\]])+)\\]:[ \\t]*(?:<([^>\\r\\n]*)>|([^<\\s][^\\s]*))(?:[ \\t]+(?:\\\"((?:\\\\.|[^\\\"\\\\])*)\\\"|'((?:\\\\.|[^'\\\\])*)'|\\(((?:\\\\.|[^)\\\\])*)\\)))?[ \\t]*$\")\n\ndef _thematic(line: str) -> bool:\n s = line.lstrip(\" \")\n if len(line) - len(s) > 3 or not s: return False\n s = s.rstrip(\" \\t\")\n chars = s.replace(\" \", \"\").replace(\"\\t\", \"\")\n return bool(chars) and chars[0] in \"-_ *\" and chars[0] != \" \" and len(chars) >= 3 and all(c == chars[0] for c in chars)\n\ndef _label(label: str) -> str:\n return \" \".join(label.casefold().split())\n\ndef _unescape(value: str) -> str:\n return re.sub(r\"\\\\([!\\\"#$%&'()*+,\\-./:;<=>?@\\[\\\\\\]^_`{|}~])\", r\"\\1\", value)\n\ndef _decode_info(value: str) -> str:\n def replace(match: re.Match[str]) -> str:\n token = match.group(0)\n if token.startswith(\"&#x\") or token.startswith(\"&#X\"):\n return chr(int(token[3:-1], 16))\n if token.startswith(\"&#\"):\n return chr(int(token[2:-1], 10))\n return html5.get(token[1:], token)\n return re.sub(r\"&(?:#[0-9]{1,7}|#x[0-9A-Fa-f]{1,6}|#X[0-9A-Fa-f]{1,6}|[A-Za-z][A-Za-z0-9]{1,31});\", replace, value)\n\n\ndef _quote_lazy_continuation(line: str, inner: list[str]) -> bool:\n \"\"\"Return whether an unmarked line can lazily continue a quote paragraph.\"\"\"\n if not inner or not inner[-1].strip():\n return False\n if re.match(r\"^ {0,3}(?:`{3,}|~{3,})\", inner[-1]):\n return False\n if re.match(r\"^ {0,3}>\", line):\n return False\n if _thematic(line):\n return False\n if re.match(r\"^ {0,3}(?:`{3,}|~{3,})[^`\\n]*$\", line):\n return False\n if re.match(r\"^ {0,3}#{1,6}(?:[ \\t]|$)\", line):\n return False\n if re.match(r\"^ {0,3}(?:[-+*](?:[ \\t]|$)|\\d{1,9}[.)](?:[ \\t]|$))\", line):\n return False\n if re.match(r\"^ {0,3}|/>)\", line, re.I):\n return False\n # Four spaces can only be lazy text when they could not begin an\n # indented code block in the quote's current paragraph. A list marker\n # at that indentation is the CommonMark example of this case.\n if len(line) - len(line.lstrip(\" \")) >= 4:\n return bool(re.match(r\"^ {4}(?:[-+*](?:[ \\t]|$)|\\d{1,9}[.)](?:[ \\t]|$))\", line))\n return True\n\ndef parse_blocks(markdown: str) -> Document:\n lines = markdown.replace(\"\\r\\n\", \"\\n\").replace(\"\\r\", \"\\n\").split(\"\\n\")\n if lines and lines[-1] == \"\": lines.pop()\n blocks: list[Block] = []; refs: dict[str, tuple[str, str | None]] = {}; paragraph: list[str] = []; i = 0\n def flush() -> None:\n nonlocal paragraph\n if paragraph:\n blocks.append(Block(\"paragraph\", \"\\n\".join(x.lstrip(\" \\t\") if n == 0 else x for n, x in enumerate(paragraph)).rstrip(\" \\t\"))); paragraph = []\n while i < len(lines):\n line = lines[i]\n if not line.strip(): flush(); i += 1; continue\n if re.match(r\"^ {0,3}>\", line):\n flush(); inner = []\n raw_quote = []\n while i < len(lines):\n if re.match(r\"^ {0,3}>\", lines[i]):\n raw_quote.append(lines[i]); inner.append(re.sub(r\"^ {0,3}>[ \\t]?\", \"\", lines[i])); i += 1\n continue\n if not lines[i].strip() or not _quote_lazy_continuation(lines[i], inner):\n break\n raw_quote.append(lines[i]); inner.append(re.sub(r\"^ {0,3}>[ \\t]?\", \"\", lines[i])); i += 1\n for n, raw in enumerate(raw_quote):\n if not re.match(r\"^ {0,3}>\", raw) and re.match(r\"^ {0,3}(=+|-+)[ \\t]*$\", inner[n]):\n inner[n] = \"\\\\\" + inner[n].lstrip()\n nested = parse_blocks(\"\\n\".join(inner)); refs.update(nested.link_references); blocks.append(Block(\"blockquote\", \"\\n\".join(inner))); continue\n # A setext underline has precedence over a thematic break when it\n # follows paragraph text.\n if paragraph and re.match(r\"^ {0,3}(=+|-+)[ \\t]*$\", line):\n blocks.append(Block(\"heading\", \"\\n\".join(paragraph).strip(\" \\t\"), 1 if line.lstrip().startswith(\"=\") else 2)); paragraph = []; i += 1; continue\n if _thematic(line): flush(); blocks.append(Block(\"thematic_break\")); i += 1; continue\n list_match = re.match(r\"^( {0,3})([-+*])([ \\t]+|$)(.*)$\", line) or re.match(r\"^( {0,3})(\\d{1,9})([.)])([ \\t]+|$)(.*)$\", line)\n list_content = (list_match.group(5) if list_match.group(2).isdigit() else list_match.group(4)) if list_match else \"\"\n if list_match and (not paragraph or (list_content.strip() and (list_match.group(2).isdigit() and list_match.group(2) == \"1\" or not list_match.group(2).isdigit()))):\n if paragraph:\n flush()\n ordered = list_match.group(2).isdigit()\n delimiter = list_match.group(3) if ordered else list_match.group(2)\n first_number = int(list_match.group(2)) if ordered else 1\n items: list[tuple[str, bool]] = []\n while i < len(lines):\n current = (re.match(r\"^( {0,3})([-+*])([ \\t]+|$)(.*)$\", lines[i]) if not ordered\n else re.match(r\"^( {0,3})(\\d{1,9})\" + re.escape(delimiter) + r\"([ \\t]+|$)(.*)$\", lines[i]))\n if not current or (not ordered and current.group(2) != delimiter) or (not ordered and _thematic(lines[i]) and current.group(2) != \"-\"):\n break\n indent = len(current.group(1)); marker_width = len(current.group(2)) + (1 if ordered else 0)\n gap = current.group(3); gap_width = min(len(gap.expandtabs(4)), 4) if gap else 1\n body_indent = indent + marker_width + gap_width\n body = [\" \" * max(0, len(gap.expandtabs(4)) - 1) + current.group(4)]; i += 1; had_blank = False; separated = False; blank_outer = False\n while i < len(lines):\n next_item = re.match(r\"^( {0,3})([-+*]|\\d{1,9}[.)])([ \\t]+|$)(.*)$\", lines[i])\n if next_item and len(next_item.group(1)) < body_indent:\n separated = had_blank\n break\n raw = lines[i]\n if not raw.strip():\n if not body[0].strip():\n had_blank = True; i += 1; continue\n had_blank = True; body.append(\"\")\n elif not body[0].strip() and had_blank:\n break\n elif body[0].startswith(\" \") and had_blank:\n body.append(raw[3:])\n elif raw.expandtabs(4).startswith(\" \" * body_indent):\n body.append(raw[body_indent:])\n elif len(raw) - len(raw.lstrip(\" \")) >= 4 and re.match(r\"^ *(?:[-+*]|\\d{1,9}[.)])(?:[ \\t]|$)\", raw):\n if re.match(r\"^ *\\d{1,9}[.)](?:[ \\t]|$)\", raw):\n break\n body.append(\"\\\\\" + raw.lstrip(\" \"))\n elif had_blank and len(raw) - len(raw.lstrip(\" \\t\")) < body_indent:\n if body[0].startswith(\" \"):\n body.append(raw[3:])\n else:\n break\n elif raw.startswith((\" \", \"\\t\")):\n body.append(raw.lstrip(\" \\t\"))\n else:\n body.append(raw)\n i += 1\n while body and not body[-1].strip(): body.pop()\n text = \"\\n\".join(body)\n direct = parse_blocks(text).blocks if text else ()\n direct_loose = had_blank and len(direct) > 1 and (not any(b.kind in {\"bullet_list\", \"ordered_list\"} for b in direct) or direct[-1].kind == \"paragraph\")\n code_only = len(direct) == 1 and direct[0].kind in {\"fenced_code\", \"indented_code\"}\n has_list = any(b.kind in {\"bullet_list\", \"ordered_list\"} for b in direct)\n separated = separated and (not has_list or bool(re.match(r\"^(?:[-+*]|\\d{1,9}[.)])(?:[ \\t]|$)\", text.splitlines()[-1] if text else \"\")))\n items.append((text, direct_loose or (separated and not code_only)))\n blocks.append(Block(\"ordered_list\" if ordered else \"bullet_list\", level=first_number, items=tuple(items), tight=not any(x[1] for x in items))); continue\n fence = _FENCE.match(line)\n if fence:\n flush(); indent, marker, rest = fence.groups(); info = rest.strip()\n if marker[0] == \"`\" and \"`\" in info: paragraph.append(line); i += 1; continue\n content: list[str] = []; i += 1\n while i < len(lines):\n if re.match(r\"^ {0,3}\" + re.escape(marker[0]) + r\"{\" + str(len(marker)) + r\",}[ \\t]*$\", lines[i]): i += 1; break\n raw = lines[i]\n if indent:\n raw = raw[min(len(indent), len(raw) - len(raw.lstrip(\" \"))):]\n content.append(raw); i += 1\n blocks.append(Block(\"fenced_code\", \"\\n\".join(content) + (\"\\n\" if content else \"\"), info=_unescape(_decode_info(info.split()[0])) if info else \"\")); continue\n if paragraph and (re.match(r\"^ {4}\", line) or line.startswith(\"\\t\")):\n paragraph.append(line[4:] if line.startswith(\" \") else line[1:] if line.startswith(\"\\t\") else line); i += 1; continue\n if re.match(r\"^ {4}\", line) or line.startswith(\"\\t\"):\n flush(); content = []\n while i < len(lines) and (not lines[i].strip() or re.match(r\"^ {4}\", lines[i]) or lines[i].startswith(\"\\t\")):\n content.append(lines[i][4:] if lines[i].startswith(\" \") else lines[i][1:] if lines[i].startswith(\"\\t\") else \"\"); i += 1\n while content and not content[-1]: content.pop()\n blocks.append(Block(\"indented_code\", \"\\n\".join(content) + \"\\n\")); continue\n atx = _ATX.match(line)\n if atx:\n flush(); text = (atx.group(3) or \"\").strip(\" \\t\"); text = re.sub(r\"(?:^|[ \\t])+#+[ \\t]*$\", \"\", text).strip(\" \\t\")\n blocks.append(Block(\"heading\", text, len(atx.group(2)))); i += 1; continue\n if paragraph and re.match(r\"^ {0,3}(=+|-+)[ \\t]*$\", line):\n marker = line.strip(); blocks.append(Block(\"heading\", \"\\n\".join(paragraph).strip(\" \\t\"), 1 if marker[0] == \"=\" else 2)); paragraph = []; i += 1; continue\n ref = _REF.match(line)\n loose_ref = re.match(r\"^ {0,3}\\[([^\\]\\n]+)\\]:[ \\t]*$\", line)\n if not paragraph and line.startswith(\"[\") and \"]:\" not in line:\n j = i\n label_lines = []\n while j < len(lines) and \"]:\" not in lines[j] and j < i + 10:\n label_lines.append(lines[j].strip()); j += 1\n if j < len(lines) and \"]:\" in lines[j]:\n prefix, rest = lines[j].split(\"]:\", 1)\n candidate = \"[\" + \" \".join(label_lines) + prefix[1:] + \"]:\" + rest\n candidate_match = _REF.match(candidate)\n if candidate_match:\n destination = candidate_match.group(2) if candidate_match.group(2) is not None else candidate_match.group(3) or \"\"\n title = next((candidate_match.group(x) for x in (4, 5, 6) if candidate_match.group(x) is not None), None)\n refs.setdefault(_label(_unescape(candidate_match.group(1))), (_unescape(destination), _unescape(title) if title is not None else None)); i = j + 1; continue\n multiline_title = re.match(r\"^ {0,3}\\[((?:\\\\.|[^\\]])+)\\]:[ \\t]*(<[^>]*>|[^<\\s][^\\s]*)[ \\t]+([\\\"'])((?:.*))$\", line)\n if not paragraph and multiline_title and multiline_title.group(3) not in multiline_title.group(4):\n quote_char = multiline_title.group(3)\n title_parts = [multiline_title.group(4)]\n cursor = i + 1\n closed = False\n while cursor < len(lines):\n # A blank line terminates a reference definition; the\n # title cannot span it. Leave the lines for normal parsing.\n if not lines[cursor].strip():\n break\n title_parts.append(lines[cursor])\n if quote_char in lines[cursor]:\n closed = True\n cursor += 1\n break\n cursor += 1\n if closed:\n destination = multiline_title.group(2).strip(\"<>\")\n refs.setdefault(_label(_unescape(multiline_title.group(1))), (_unescape(destination), _unescape(\"\\n\".join(title_parts).rsplit(quote_char, 1)[0])))\n i = cursor\n continue\n nested_title = re.match(r\"^ {0,3}\\[((?:\\\\.|[^\\]])+)\\]:[ \\t]*(<[^>]*>|[^<\\s][^\\s]*)[ \\t]+(['\\\"])(.*)\\3[ \\t]*$\", line)\n if not paragraph and nested_title:\n destination = nested_title.group(2).strip(\"<>\")\n refs.setdefault(_label(_unescape(nested_title.group(1))), (_unescape(destination), _unescape(nested_title.group(4)))); i += 1; continue\n continued_title = re.match(r\"^ {0,3}\\[((?:\\\\.|[^\\]])+)\\]:[ \\t]*(<[^>]*>|[^<\\s][^\\s]*)[ \\t]*$\", line)\n if not paragraph and continued_title and i + 1 < len(lines):\n tm = re.match(r\"^[ \\t]*(['\\\"])(.*)\\1[ \\t]*$\", lines[i + 1])\n if tm:\n refs.setdefault(_label(_unescape(continued_title.group(1))), (_unescape(continued_title.group(2).strip(\"<>\")), _unescape(tm.group(2)))); i += 2; continue\n if loose_ref and not paragraph and i + 1 < len(lines):\n destination_line = lines[i + 1].strip()\n destination_match = re.match(r\"(?:<([^>]+)>|([^\\s]+))(?:[ \\t]+(?:\\\"([^\\\"]*)\\\"|'([^']*)'|\\(([^)]*)\\)))?$\", destination_line)\n if destination_match:\n destination = destination_match.group(1) if destination_match.group(1) is not None else destination_match.group(2) or \"\"\n title = next((destination_match.group(x) for x in (3, 4, 5) if destination_match.group(x) is not None), None)\n consumed = 2\n if title is None and i + 2 < len(lines):\n title_match = re.match(r\"^[ \\t]*(?:\\\"([^\\\"]*)\\\"|'([^']*)'|\\(([^)]*)\\))[ \\t]*$\", lines[i + 2])\n if title_match: title = next((title_match.group(x) for x in (1, 2, 3) if title_match.group(x) is not None), None); consumed = 3\n refs.setdefault(_label(loose_ref.group(1)), (destination, title)); i += consumed; continue\n if ref and not paragraph:\n destination = ref.group(2) if ref.group(2) is not None else ref.group(3) or \"\"\n title = next((ref.group(x) for x in (4, 5, 6) if ref.group(x) is not None), None)\n refs.setdefault(_label(_unescape(ref.group(1))), (_unescape(destination), _unescape(title) if title is not None else None)); i += 1; continue\n if i + 1 < len(lines) and re.match(r\"^ {0,3}(=+|-+)[ \\t]*$\", lines[i + 1]) and line.strip() and not paragraph and not line.lstrip().startswith((\">\", \"- \", \"* \", \"+ \")):\n blocks.append(Block(\"heading\", line.strip(\" \\t\"), 1 if \"=\" in lines[i + 1] else 2)); i += 2; continue\n stripped = line.lstrip()\n html_start = re.match(r\"<(?:pre|script|style|textarea)(?:\\s|>|$)|