#!/usr/bin/env python3 """Extract the reader's view of a rendered chapter. A reviewer needs to see what a reader sees, not the markup. This flattens one built HTML page into plain text that preserves the things a rendering or prose defect actually shows up in: heading structure, paragraph text, list items, table cells, captions, callout bodies, and math rendered back to its LaTeX source. Navigation chrome, scripts, and styles are dropped. Usage: render_reader_view.py [--section N] [--list] --list print the section outline with indices and line counts --section N print only section N (use --list first) """ from __future__ import annotations import argparse import html import re import sys from pathlib import Path DROP_BLOCK = re.compile( r"<(script|style|head|nav|footer)\b.*?", re.S | re.I ) # site chrome that is not chapter content DROP_ATTR = re.compile( r'<(?:div|aside|section)[^>]*\b(?:id|class)="[^"]*' r"(?:sidebar|navbar|toc|footer|announcement|quarto-secondary-nav|" r'margin-sidebar|nav-page|cookie)[^"]*"[^>]*>', re.I, ) def strip_chrome(doc: str) -> str: """Drop script, style, head, nav, and footer blocks, then keep only ``
`` when present.""" doc =DROP_BLOCK.sub(" ", doc) # keep it simple: remove obvious nav containers by cutting to
if present m = re.search(r"", doc, re.S | re.I) if m: doc = m.group(0) return doc def mathml_to_tex(doc: str) -> str: """Replace MathJax/MathML output with the LaTeX it came from.""" doc = re.sub( r'\s*(.*?)\s*', lambda m: " " + re.sub(r"<[^>]+>", "", m.group(1)) + " ", doc, flags=re.S, ) return doc BLOCK_END = re.compile( r"", re.I, ) CELL_SEP = re.compile(r"", re.I) HEADING = re.compile(r"<(h[1-6])\b[^>]*>(.*?)", re.S | re.I) def to_text(doc: str) -> str: """Flatten chapter HTML to plain text. Math spans are reduced to their source, headings become ``@@H@@`` marker lines, table cells are joined with `` | ``, block ends and ``
`` become newlines, remaining tags are removed, entities are unescaped, and runs of whitespace and blank lines are collapsed. """ doc =mathml_to_tex(doc) # mark headings so the outline survives flattening doc = HEADING.sub( lambda m: f"\n\n@@H{m.group(1)[1]}@@ " + re.sub(r"<[^>]+>", "", m.group(2)).strip() + "\n", doc, ) doc = CELL_SEP.sub(" | ", doc) doc = BLOCK_END.sub("\n", doc) doc = re.sub(r"", "\n", doc, flags=re.I) doc = re.sub(r"<[^>]+>", "", doc) doc = html.unescape(doc) doc = re.sub(r"[ \t]+", " ", doc) doc = re.sub(r"\n\s*\n\s*\n+", "\n\n", doc) return "\n".join(ln.strip() for ln in doc.splitlines()).strip() def split_sections(text: str) -> list[tuple[str, str]]: """Split flattened text into ``(title, body)`` sections at H1-H3 markers. Text before the first such heading is titled ``(front)``, and H4-H6 markers inside a body are rewritten as Markdown ``#`` headings. """ parts: list[tuple[str, str]] = [] cur_title, buf = "(front)", [] for line in text.splitlines(): m = re.match(r"@@H([1-6])@@ (.*)", line) if m and m.group(1) in "123": if buf: parts.append((cur_title, "\n".join(buf).strip())) cur_title, buf = f"H{m.group(1)}: {m.group(2)}", [] else: buf.append(re.sub(r"@@H([1-6])@@ ", lambda x: "#" * int(x.group(1)) + " ", line)) if buf: parts.append((cur_title, "\n".join(buf).strip())) return parts def main() -> int: """Print the reader view of one built HTML file. Prints every section by default, the outline with ``--list``, or a single section with ``--section``. Returns 1 when the section index is out of range, else 0. """ ap =argparse.ArgumentParser() ap.add_argument("path") ap.add_argument("--section", type=int) ap.add_argument("--list", action="store_true") a = ap.parse_args() raw = Path(a.path).read_text(errors="ignore") text = to_text(strip_chrome(raw)) secs = split_sections(text) if a.list: print(f"{len(secs)} sections in {a.path}") for i, (t, body) in enumerate(secs): print(f" [{i:2d}] {len(body.splitlines()):5d} lines {t[:90]}") return 0 if a.section is not None: if not 0 <= a.section < len(secs): print(f"section {a.section} out of range (0..{len(secs)-1})", file=sys.stderr) return 1 t, body = secs[a.section] print(f"===== [{a.section}] {t} =====\n{body}") return 0 for i, (t, body) in enumerate(secs): print(f"\n===== [{i}] {t} =====\n{body}") return 0 if __name__ == "__main__": raise SystemExit(main())