#!/usr/bin/env python3 # SPDX-License-Identifier: Apache-2.0 """Recheck Year 8 English W03–04 content, source rows, aids and receipt.""" from __future__ import annotations import argparse import hashlib import json import re import subprocess import sys import tempfile import xml.etree.ElementTree as ET from pathlib import Path ROOT = Path(__file__).resolve().parent STUDIO = ROOT.parents[4] FRAMEWORK = STUDIO / "data/frameworks/acara-v9.json" WORKBOOK = STUDIO / "research/sources/acara-australian-curriculum-v9-download-2026-09-29.xlsx" SOURCE_SHA = "db446882d2c00cf7c085a03e250e2442fda6c44011fc114680c46c1dc7a822c3" MANIFEST = ROOT / "MANIFEST.sha256" ROWS = { "AC9E8LA02": 934, "AC9E8LA03": 937, "AC9E8LA04": 941, "AC9E8LA05": 944, "AC9E8LA06": 947, "AC9E8LA08": 952, "AC9E8LA09": 955, "AC9E8LE02": 964, "AC9E8LE03": 967, "AC9E8LE05": 973, "AC9E8LE06": 978, "AC9E8LY02": 989, "AC9E8LY03": 994, "AC9E8LY04": 997, "AC9E8LY05": 1002, "AC9E8LY06": 1007, "AC9E8LY07": 1012, } ASSETS = ("equivalence-evidence", "hybrid-source-map", "line-meaning") TEACHING = ("README.md", "TEXTS.md", "LESSONS.md", "LEARNER-COPY.md", "OPTIONAL-PRACTICE-AND-HOME.md") def need(condition: bool, message: str) -> None: if not condition: raise AssertionError(message) def sha(path: Path) -> str: return hashlib.sha256(path.read_bytes()).hexdigest() def official() -> dict[str, dict]: data = json.loads(FRAMEWORK.read_text(encoding="utf-8")) need(data["framework"] == "Australian Curriculum Version 9.0", "framework label drift") need(data["source_sha256"] == SOURCE_SHA == sha(WORKBOOK), "source workbook hash mismatch") selected = [r for r in data["records"] if r.get("code") in ROWS] need(len(selected) == len(ROWS) == 17, "seventeen unique official content-description rows required") by_code = {r["code"]: r for r in selected} for code, expected_row in ROWS.items(): row = by_code[code] attr = row["attributes"] need(row["record_type"] == "content_description" and row["source_row"] == expected_row, f"{code}: source row/type mismatch") need(attr["learning_area"] == "English" and attr["level"] == "Year 8", f"{code}: area/level mismatch") need(" ".join(row["plain_text"].split()) == " ".join(attr["content_description"].split()), f"{code}: official wording drift") return by_code def lesson_sections() -> dict[int, tuple[str, str]]: content = (ROOT / "LESSONS.md").read_text(encoding="utf-8") marks = list(re.finditer(r"^## Week (\d+) Day (\d+) (.+)$", content, re.MULTILINE)) need([int(m.group(2)) for m in marks] == list(range(11, 21)), "lesson headers must be Days 11–20 once") sections = {} used = set() for i, mark in enumerate(marks): day = int(mark.group(2)) need(int(mark.group(1)) == (3 if day <= 15 else 4), f"Day {day}: week mismatch") body = content[mark.end(): marks[i + 1].start() if i + 1 < len(marks) else len(content)] steps = [int(x) for x in re.findall(r"^\d+\. \*\*[^\n]*? · (\d+) min\.\*\*", body, re.MULTILINE)] need(len(steps) == 6 and sum(steps) == 25, f"Day {day}: six timed steps must total 25, got {steps}") need("**Goal:**" in body and "**Prepare:**" in body and ("**Exit" in body or (day in (15, 20) and "**Collect" in body)), f"Day {day}: goal/preparation/exit absent") codes = set(re.findall(r"\bAC9E8[A-Z0-9]+\b", body)) need(codes and codes <= ROWS.keys(), f"Day {day}: invalid/no code {codes - ROWS.keys()}") used |= codes sections[day] = (mark.group(3), body) need(used == ROWS.keys(), f"lesson codes differ from crosswalk: {used ^ ROWS.keys()}") need("listening-supported comprehension" in content and "not independent print decoding" in content, "decoding/access boundary missing") return sections def crosswalk(by_code: dict[str, dict], sections: dict[int, tuple[str, str]]) -> str: lines = [ "# ACARA v9 crosswalk · Year 8 English Weeks 3–4", "", "**Pinned source:** [Australian Curriculum Version 9.0 official workbook](https://www.australiancurriculum.edu.au/downloads), retrieved 29 September 2026, original XLSX SHA-256 `" + SOURCE_SHA + "`; exact source-row import in local `products/curriculum-studio/data/frameworks/acara-v9.json`. Wording below is the official content description after plain-text whitespace normalisation. All entries are **Year 8 English**. These are selected lesson opportunities, not whole-description completion, a national achievement decision or local syllabus approval.", "", "| Code | Original workbook row | Exact official Year 8 content description | Day(s) with a planned action |", "| --- | ---: | --- | --- |", ] for code, source_row in sorted(ROWS.items(), key=lambda entry: entry[1]): row = by_code[code] wording = " ".join(row["plain_text"].split()).replace("|", "\\|") days = ", ".join(f"{day} ({title})" for day, (title, body) in sections.items() if code in body) need(days != "", f"{code}: no lesson opportunity") lines.append(f"| `{code}` | {source_row} | {wording} | {days} |") lines += [ "", "## How to interpret these links", "", "Codes identify a **part** of a description addressed by a task. Source evaluation uses fictional paper display and timing models; no real site, event, wall or visitor has been studied. The literary work includes one original vignette and poem; the simile/metaphor, viewpoint and punctuation tasks sample only a small part of the language/literature descriptions. A school must plan a far wider range of Australian, First Nations Australian and world-authored literature with rights and cultural protocols; original SubjectNest texts cannot replace this. Creation codes receive draft, edit and private reader-test opportunities, not the full range of purposes, modes or publication. LY07 has actual private spoken delivery only for a learner who takes that route; an AAC or written response is recorded as that route rather than invented vocal evidence. A read-aloud response is listening-supported comprehension, not independently decoded reading.", "", "The [Day 15 and 20 checks](ASSESSMENT.md) sample transfer to new texts. The separate [staff key](TEACHER-KEY.md) gives criterion-level next moves; neither check alone decides the achievement standard. State/territory scope, school timetable, accessibility and text selection require local review.", "", "**ACARA attribution:** © Australian Curriculum, Assessment and Reporting Authority (ACARA) 2010 to present, unless otherwise indicated. This material was downloaded from the [Australian Curriculum website](https://www.australiancurriculum.edu.au/downloads) (accessed 29 September 2026) and modified only for plain-text whitespace and selected rows, under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) subject to [ACARA terms and exclusions](https://www.australiancurriculum.edu.au/copyright-and-terms-of-use). ACARA does not endorse SubjectNest. Original crosswalk commentary © NeuroForgeIO Pty Ltd 2026, CC BY 4.0.", "", ] return "\n".join(lines) def learner_and_checks() -> None: learner = (ROOT / "LEARNER-COPY.md").read_text(encoding="utf-8") marks = list(re.finditer(r"^## Day (\d+)\s*$", learner, re.MULTILINE)) need([int(m.group(1)) for m in marks] == list(range(11, 21)), "learner Days 11–20") for i, mark in enumerate(marks): day = int(mark.group(1)) body = learner[mark.end(): marks[i + 1].start() if i + 1 < len(marks) else len(learner)] need([x for x in re.findall(r"^- \*\*([ABC]) ·", body, re.MULTILINE)] == ["A", "B", "C"], f"Day {day}: three A/B/C choices") need("**Exit:**" in body, f"Day {day}: missing exit") need("TEACHER-KEY" not in learner and "ASSESSMENT.md" not in learner, "child-facing page links private check/key") practice = (ROOT / "OPTIONAL-PRACTICE-AND-HOME.md").read_text(encoding="utf-8") rows = re.findall(r"^\| (\d{2}) \| (.+) \| (.+) \| (.+) \|$", practice, re.MULTILINE) need([int(row[0]) for row in rows] == list(range(11, 21)) and all(all(cell.strip() for cell in row[1:]) for row in rows), "ten daily optional practice/home rows") need(practice.count("five minutes") >= 2 and "No paid account, device" in practice, "extras/home boundary missing") teaching_text = "\n".join((ROOT / name).read_text(encoding="utf-8") for name in TEACHING) for held_out in ("20 + 16c", "four-card result proves", "7 + 11b", "paper copy for everyone", "distribution will be arranged"): need(held_out.lower() not in teaching_text.lower(), f"held-out phrase leaked into teaching: {held_out}") checks = (ROOT / "ASSESSMENT.md").read_text(encoding="utf-8") key = (ROOT / "TEACHER-KEY.md").read_text(encoding="utf-8") need(all(x in checks for x in ("Check A Day 15", "Check B Day 20", "20 + 16c", "7 + 11b")), "fresh assessment source incomplete") for code in ("A1", "A2", "A3", "A4", "B1", "B2", "B3", "B4"): need(re.search(rf"^\| {code} ·", key, re.MULTILINE) is not None, f"rubric {code} absent") need("listening comprehension" in checks and "not independent print-reading evidence" in checks, "assessment access boundary missing") need("**A · Spoken presentation.**" in learner and "**B · AAC/text-to-speech presentation.**" in learner and "**C · Written alternative.**" in learner and "not a spoken performance sample" in learner, "delivery evidence must follow actual Day 19 mode") need(12 + 18 * 2 + 12 == 60 and 24 + 18 * 4 == 96 and 24 + 18 * 5 == 114 and 5 + 15 * 3 == 50 and 5 + 15 * 4 == 65 and 10 + 16 * 4 + 10 == 84 and 20 + 16 * 5 == 100 and 7 + 11 * 3 == 40 and 7 + 11 * 4 == 51 and 3 + 2 * 4 + 3 == 6 + 2 * 4 and 4 + 10 * 2 + 4 == 8 + 10 * 2 and 5 + 10 * 4 == 45 and 5 + 10 * 5 > 45, "arithmetic invariant failed") def assets() -> None: alt = (ROOT / "print/TEXT-ALTERNATIVES.md").read_text(encoding="utf-8") ns = {"s": "http://www.w3.org/2000/svg"} heads = ("## Equivalence and evidence aid", "## Hybrid source map", "## Line and meaning aid") need(all(h in alt for h in heads), "three full text/tactile sections required") for stem in ASSETS: svg = ROOT / "print" / f"{stem}.svg" pdf = svg.with_suffix(".pdf") need(svg.exists() and pdf.exists(), f"{stem}: missing SVG/PDF") root = ET.parse(svg).getroot() need(root.attrib.get("width") == "210mm" and root.attrib.get("height") == "297mm" and root.attrib.get("viewBox") == "0 0 210 297" and root.attrib.get("role") == "img" and root.attrib.get("aria-labelledby") == "title desc", f"{stem}: A4/access metadata") need(root.find("s:title", ns) is not None and root.find("s:desc", ns) is not None, f"{stem}: title/desc") need("CC BY 4.0" in svg.read_text(encoding="utf-8"), f"{stem}: SVG rights") info = subprocess.check_output(["pdfinfo", str(pdf)], text=True) need("(A4)" in info and re.search(r"Pages:\s+1\b", info) is not None, f"{stem}: PDF geometry") extracted = subprocess.check_output(["pdftotext", str(pdf), "-"], text=True) need("SubjectNest" in extracted and "CC BY 4.0" in extracted and len(extracted) > 160, f"{stem}: PDF text extraction") fonts = subprocess.check_output(["pdffonts", str(pdf)], text=True) need("DejaVuSans" in fonts, f"{stem}: expected embedded DejaVu subset") with tempfile.TemporaryDirectory(prefix="subjectnest-y8-english-") as temporary: subprocess.run([sys.executable, str(ROOT / "print/generate_print.py"), "--output-dir", temporary], check=True, capture_output=True, text=True) for stem in ASSETS: for suffix in (".svg", ".pdf"): name = stem + suffix need(sha(Path(temporary) / name) == sha(ROOT / "print" / name), f"{name}: rendered bytes differ from original source") def slug(heading: str) -> str: heading = re.sub(r"<[^>]+>", "", heading) return re.sub(r"[^a-z0-9-]", "", heading.strip().lower().replace(" ", "-")) def links_and_rights() -> int: checked = 0 for path in ROOT.rglob("*.md"): source = path.read_text(encoding="utf-8") need("CC BY 4.0" in source, f"{path}: no original/curriculum licence note") for raw in re.findall(r"\[[^]]+\]\(([^)]+)\)", source): if raw.startswith(("https://", "http://")): continue relative, _, anchor = raw.partition("#") target = (path.parent / relative).resolve() if relative else path need(target.exists(), f"{path}: broken link {raw}") if anchor and target.suffix.lower() == ".md": headings = [slug(m.group(1)) for m in re.finditer(r"^#{1,6} (.+)$", target.read_text(encoding="utf-8"), re.MULTILINE)] need(anchor in headings, f"{path}: broken heading #{anchor} in {target}") checked += 1 return checked def payload_files() -> list[Path]: return sorted((p for p in ROOT.rglob("*") if p.is_file() and p != MANIFEST and "__pycache__" not in p.parts and ".ruff_cache" not in p.parts), key=lambda p: p.relative_to(ROOT).as_posix()) def manifest() -> str: return "".join(f"{sha(p)} {p.relative_to(ROOT).as_posix()}\n" for p in payload_files()) def main() -> None: parser = argparse.ArgumentParser() parser.add_argument("--write-crosswalk", action="store_true") parser.add_argument("--write-manifest", action="store_true") args = parser.parse_args() by_code = official() sections = lesson_sections() expected = crosswalk(by_code, sections) path = ROOT / "CURRICULUM-CROSSWALK.md" if args.write_crosswalk: path.write_text(expected, encoding="utf-8") else: need(path.read_text(encoding="utf-8") == expected, "crosswalk official wording/day drift") learner_and_checks() assets() link_count = links_and_rights() expected_manifest = manifest() if args.write_manifest: MANIFEST.write_text(expected_manifest, encoding="utf-8") else: need(MANIFEST.read_text(encoding="utf-8") == expected_manifest, "SHA manifest drift") print(f"PASS: 10 × 25-minute lessons; 30 core routes; 20 optional extras + 10 home routes; " f"2 fresh checks/8 rubric criteria; 17 exact Year 8 English codes/rows; " f"3 A4 SVG/PDF/text aids; {link_count} local links; {len(payload_files())} hashed files") if __name__ == "__main__": main()