"""Focused block-structure probes. No network; all writes stay beside this script.""" import hashlib import importlib import importlib.metadata import inspect import json import platform import subprocess from collections import Counter from datetime import datetime, timezone from html.parser import HTMLParser from pathlib import Path import markdown from markdown_it import MarkdownIt ROOT = Path(__file__).resolve().parent SAMPLES = { 'types-inline': 'Use `alpha\nbeta` in prose.\n', 'types-backticks': '```python\nprint("alpha")\nprint("beta")\n```\n', 'types-tildes': '~~~python\nprint("alpha")\nprint("beta")\n~~~\n', 'types-indented': ' print("alpha")\n print("beta")\n', 'types-indented-attached': 'Run this:\n print("alpha")\n print("beta")\n', 'types-indented-separated': 'Run this:\n\n print("alpha")\n print("beta")\n', 'types-fenced-attached': 'Run this:\n```python\nprint("alpha")\nprint("beta")\n```\n', 'types-overindented': ' ```python\n print("alpha")\n print("beta")\n ```\n', 'types-list-fenced': '- Run:\n\n ```python\n print("alpha")\n print("beta")\n ```\n', 'types-list-indented': '- Run:\n\n print("alpha")\n print("beta")\n', 'types-nested-example': '````markdown\n```python\nprint("alpha")\nprint("beta")\n```\n````\n', } class Structure(HTMLParser): def __init__(self): super().__init__() self.tags = Counter() self.stack = [] self.pre_in_li = 0 self.block_texts = [] self.inline_texts = [] self.current_block = None self.current_inline = None def handle_starttag(self, tag, attrs): self.tags[tag] += 1 if tag == 'pre': self.pre_in_li += int('li' in self.stack) self.current_block = '' if tag == 'code' and 'pre' not in self.stack: self.current_inline = '' self.stack.append(tag) def handle_endtag(self, tag): if tag == 'pre' and self.current_block is not None: self.block_texts.append(self.current_block) self.current_block = None if tag == 'code' and self.current_inline is not None: self.inline_texts.append(self.current_inline) self.current_inline = None if tag in self.stack: self.stack = self.stack[:len(self.stack) - 1 - self.stack[::-1].index(tag)] def handle_data(self, data): if self.current_block is not None: self.current_block += data if self.current_inline is not None: self.current_inline += data def sha(text): return hashlib.sha256(text.encode()).hexdigest() def main(): sources = [] for name in ('markdown_it.rules_block.code', 'markdown_it.rules_block.paragraph', 'markdown_it.rules_block.fence', 'markdown_it.rules_inline.backticks', 'markdown_it.renderer', 'markdown.extensions.fenced_code'): module = importlib.import_module(name) source = inspect.getsource(module) sources.append(f'### {name}; sha256={sha(source)}\n{source}\n') (ROOT / 'types-source-audit.txt').write_text('\n'.join(sources)) rows = [] for sample, source in SAMPLES.items(): source_path = ROOT / f'{sample}.md' source_path.write_text(source) outputs = { 'markdown-it-commonmark': MarkdownIt('commonmark').render(source), 'python-fenced': markdown.markdown(source, extensions=['tables', 'fenced_code']), 'pandoc-gfm': subprocess.run(['pandoc', '-f', 'gfm', '-t', 'html'], input=source, text=True, capture_output=True, check=True).stdout, } for renderer, output in outputs.items(): output_path = ROOT / f'{sample}--{renderer}.txt' output_path.write_text(output) parser = Structure() parser.feed(output) rows.append({'sample': sample, 'renderer': renderer, 'source': f'lab/renderers/{source_path.name}', 'output': f'lab/renderers/{output_path.name}', 'source_sha256': sha(source), 'output_sha256': sha(output), 'tags': dict(parser.tags), 'pre_in_li': parser.pre_in_li, 'block_texts': parser.block_texts, 'inline_texts': parser.inline_texts}) result = { 'kind': 'markdown-rendering', 'date': datetime.now(timezone.utc).isoformat(), 'python': platform.python_version(), 'versions': {'markdown-it-py': importlib.metadata.version('markdown-it-py'), 'Python-Markdown': importlib.metadata.version('Markdown'), 'pandoc': subprocess.check_output(['pandoc', '--version'], text=True).splitlines()[0]}, 'command': '.venv/bin/python runs/2026-11-07-markovo-markdown-code-block-types/lab/renderers/types_probe.py', 'options': {'markdown-it-commonmark': "MarkdownIt('commonmark'); no highlight callback", 'python-fenced': "markdown.markdown(source, extensions=['tables', 'fenced_code']); no CodeHilite", 'pandoc-gfm': 'pandoc -f gfm -t html; default highlighting'}, 'scope': 'Local implementations only; no GitHub.com or VS Code UI, CSS, browser layout, code execution, or PDF conversion tests.', 'metrics': {'tags': 'HTMLParser start-tag counts; code includes inline and block code', 'pre_in_li': 'pre start tags encountered while a li ancestor is open', 'block_texts': 'decoded text captured inside each pre element', 'inline_texts': 'decoded text inside code without a pre ancestor'}, 'results': rows, } (ROOT / 'types-results.json').write_text(json.dumps(result, indent=2) + '\n') print(json.dumps({k: result[k] for k in ('kind', 'date', 'python', 'versions', 'scope')}, indent=2)) for sample in SAMPLES: print(sample, [(r['renderer'], r['tags'].get('pre', 0), r['pre_in_li'], r['tags'].get('code', 0)) for r in rows if r['sample'] == sample]) if __name__ == '__main__': main()