#!/usr/bin/env python3 # SPDX-License-Identifier: Apache-2.0 """Audit the original Year 4 English continuation and its pinned curriculum source.""" from __future__ import annotations import argparse import hashlib import json import re import subprocess import sys import xml.etree.ElementTree as ET from pathlib import Path from urllib.parse import unquote ROOT = Path(__file__).resolve().parent STUDIO = ROOT.parents[4] IMPORT = STUDIO / "data/frameworks/acara-v9.json" SOURCE_SHA = "db446882d2c00cf7c085a03e250e2442fda6c44011fc114680c46c1dc7a822c3" CODES = ["AC9E4LA02", "AC9E4LA03", "AC9E4LA04", "AC9E4LA05", "AC9E4LA06", "AC9E4LA10", "AC9E4LA11", "AC9E4LY02", "AC9E4LY03", "AC9E4LY04", "AC9E4LY05", "AC9E4LY06", "AC9E4LY07", "AC9E4LY09", "AC9E4LY10"] EVIDENCE = { "AC9E4LA02": "Days 13, 17 and 20 sort source-supported report from a marked preference; no whole-year opinion-language claim.", "AC9E4LA03": "Days 11 and 13 compare report, procedure and story stages; Check A transfers structure to a new page.", "AC9E4LA04": "Days 14 and 18 use temporal, conditional and causal links in short source-bound sentences.", "AC9E4LA05": "Day 12 examines a text-only online mockup's headline, breadcrumb, menu, link and headings; Check A is a new mockup, not a tested site.", "AC9E4LA06": "Days 14 and 18 join a complete clause to a dependent because/if clause; a brief application only.", "AC9E4LA10": "Day 16 contrasts top and side framing; Check B asks what a side view foregrounds, not full moving-image study.", "AC9E4LA11": "Day 17 tests whether a synonym or antonym preserves a source claim; limited word set.", "AC9E4LY02": "Days 13 and 19 ask learners to acknowledge a listener's interpretation and respond to a question, in flexible modes.", "AC9E4LY03": "Days 11 and 13 identify purposes and features of several text types; Check A transfers to a new page.", "AC9E4LY04": "Days 11, 12 and 16 include separate matched independent reading; adult-read packet comprehension is not fluency evidence.", "AC9E4LY05": "Across both weeks pupils locate, question, summarise and evaluate model limits; fresh checks sample transfer.", "AC9E4LY06": "Days 14 and 18–20 plan, compose, revise and check short informative text; other genres need later work.", "AC9E4LY07": "Day 19 rehearses one brief audience-specific oral/AAC/multimodal explanation and revises after a question.", "AC9E4LY09": "Days 14 and 17 practise known morphemes and word parts; two fresh checks sample new forms after taught-prerequisite review.", "AC9E4LY10": "Days 14 and 17 examine word-family spelling; fresh words invite explanation without asserting full spelling mastery.", } HELD_OUT = ("relabelled", "unmarked", "repositioned", "unmeasured") def require(test: bool, message: str) -> None: if not test: raise AssertionError(message) def norm(value: str) -> str: return " ".join(value.split()) def source() -> tuple[dict, dict[str, dict]]: data = json.loads(IMPORT.read_text(encoding="utf-8")) require(data["source_sha256"] == SOURCE_SHA, "Pinned ACARA source SHA changed") original = STUDIO / "research/sources" / data["source_file"] require(original.is_file(), "Official workbook bytes missing") require(hashlib.sha256(original.read_bytes()).hexdigest() == SOURCE_SHA, "Official workbook bytes changed") require(data["framework"] == "Australian Curriculum Version 9.0", "Framework version changed") require(data["source_url"].startswith("https://www.australiancurriculum.edu.au/"), "Nonofficial source URL") records = {r["code"]: r for r in data["records"] if r.get("record_type") == "content_description" and r.get("code")} for code in CODES: require(code in records, f"Missing source code {code}") attrs = records[code]["attributes"] require(attrs["level"] == "Year 4" and attrs["learning_area"] == "English" and attrs["subject"] == "English", f"Wrong level or subject for {code}") require(isinstance(records[code]["source_row"], int), f"Missing source row {code}") return data, records def make_crosswalk(data: dict, records: dict[str, dict]) -> str: lines = [ "# Australian Curriculum v9 · Year 4 English Weeks 3–4 crosswalk", "", "These are **partial links to ten 25-minute sessions**, not proof of full content-description coverage, Year 4 achievement, a mandated order or state/territory adoption. Adult-read comprehension is not independent reading evidence.", "", f"Official source: [ACARA curriculum workbook]({data['source_url']}), retrieved {data['retrieved_at']}; original XLSX SHA-256 `{data['source_sha256']}`. Exact description wording follows the source with whitespace normalised; source rows permit an audit.", "", "| Official code | Workbook row | Level | Exact official content description | Taught opportunity and limit |", "|---|---:|---|---|---|", ] for code in CODES: rec = records[code] cells = (code, str(rec["source_row"]), rec["attributes"]["level"], norm(rec["plain_text"]), EVIDENCE[code]) lines.append("| " + " | ".join(c.replace("|", "\\|") for c in cells) + " |") lines += [ "", "© Australian Curriculum, Assessment and Reporting Authority (ACARA) 2010 to present, unless otherwise indicated. Curriculum wording was downloaded from the [Australian Curriculum website](https://www.australiancurriculum.edu.au/downloads) (accessed 29 September 2026), with whitespace/plain-text normalisation, 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 evidence notes and table arrangement © NeuroForgeIO Pty Ltd 2026, SubjectNest, CC BY 4.0. See the [source ledger](SOURCES-AND-REVIEW.md). Recheck current official and local sources before reissue.", "", ] return "\n".join(lines) def slug(heading: str) -> str: return "".join(c for c in heading.lower() if c.isalnum() or c in " -_").replace(" ", "-") def links_rights() -> int: count = 0 for path in ROOT.rglob("*.md"): body = path.read_text(encoding="utf-8") require("CC BY 4.0" in body, f"Rights missing: {path.relative_to(ROOT)}") for target in re.findall(r"(? {target}") if anchor and dest.suffix == ".md": other = dest.read_text(encoding="utf-8") heads = re.findall(r"^#{1,6}\s+(.+?)\s*$", other, re.MULTILINE) ids = set(re.findall(r'', other)) require(anchor in {slug(h) for h in heads} | ids, f"Broken anchor {path.relative_to(ROOT)} -> {target}") count += 1 return count def content() -> None: scripts = (ROOT / "LESSONS.md").read_text(encoding="utf-8") learner = (ROOT / "LEARNER.md").read_text(encoding="utf-8") extras = (ROOT / "DAILY-EXTRAS.md").read_text(encoding="utf-8") checks = (ROOT / "STUDENT-CHECKS.md").read_text(encoding="utf-8") key = (ROOT / "teacher/ANSWER-AND-NEXT.md").read_text(encoding="utf-8") for name, body in (("LESSONS", scripts), ("LEARNER", learner), ("DAILY-EXTRAS", extras)): days = [int(d) for d in re.findall(r"^## Day (\d+)\b", body, re.MULTILINE)] require(days == list(range(11, 21)), f"{name}: expected Days 11–20 once, got {days}") lesson_blocks = re.split(r"^## Day \d+\b", scripts, flags=re.MULTILINE)[1:] for day, block in enumerate(lesson_blocks, 11): minutes = [int(x) for x in re.findall(r"^\d+\. \*\*[^*]+ · (\d+) min\.\*\*", block, re.MULTILINE)] expected = [2, 3, 3, 12, 3, 2] if day in (15, 20) else [2, 5, 6, 7, 3, 2] require(minutes == expected, f"Day {day} phases drift: {minutes}") require("AC9E4" in block, f"Day {day} missing code") require(f"Day {day}" in learner and f"Day {day}" in extras, f"Day {day} choices missing") require(set(re.findall(r"\bAC9E4[A-Z0-9]+\b", scripts)) == set(CODES), "Script code set differs from exact crosswalk") require(re.findall(r"^- \*\*([ABC]) · ", learner, re.MULTILINE) == list("ABC") * 10, "Exactly 30 core choice cards required") require(re.findall(r"^- \*\*([AB]) · ", extras, re.MULTILINE) == list("AB") * 10, "Exactly 20 optional choice cards required") for word in HELD_OUT: for name in ("LESSONS.md", "LEARNER.md", "DAILY-EXTRAS.md", "MATERIALS.md", "print/TEXT-ALTERNATIVES.md"): require(re.search(rf"\b{word}\b", (ROOT / name).read_text(encoding="utf-8"), re.IGNORECASE) is None, f"Held-out word {word} leaked into {name}") require(re.search(rf"\b{word}\b", checks, re.IGNORECASE) is not None and re.search(rf"\b{word}\b", key, re.IGNORECASE) is not None, f"Held-out word {word} absent from check/key") require("adult-read texts, not controlled decodables" in (ROOT / "README.md").read_text(encoding="utf-8").lower(), "Independent-reading boundary missing") require("not yet observed" in key.lower(), "Missing evidence handling absent") for term in ("Check A", "Check B", "Day 15", "Day 20"): require(term in checks and term in key, f"Fresh check/key incomplete: {term}") for term in ("two trays", "one handle", "RETURN", "three panels", "two square picture cards", "no dimensions"): require(term.lower() in checks.lower(), f"Fresh model missing detail {term}") def visuals() -> None: ns = "{http://www.w3.org/2000/svg}" for stem, phrases in { "source-purpose-sorter": ("One plan, four text jobs", "REPORT", "PROCEDURE", "VIEWPOINT", "STORY", "One thing the source cannot prove"), "model-evidence-planner": ("Model", "WHAT THE MODEL SHOWS", "WHAT SOMEONE THINKS", "WHAT IS STILL UNKNOWN", "My short summary"), }.items(): svg = ROOT / "print" / f"{stem}.svg" pdf = svg.with_suffix(".pdf") tree = ET.parse(svg).getroot() require(tree.get("viewBox") == "0 0 210 297" and tree.get("role") == "img", f"{stem}: A4/access metadata drift") require(tree.find(ns + "title") is not None and tree.find(ns + "desc") is not None, f"{stem}: title/desc missing") require("CC BY 4.0" in svg.read_text(encoding="utf-8"), f"{stem}: rights missing") info = subprocess.run(["pdfinfo", str(pdf)], capture_output=True, text=True, check=True).stdout require("(A4)" in info and "Pages: 1" in info, f"{stem}: PDF not one A4 page") pdf_text = subprocess.run(["pdftotext", str(pdf), "-"], capture_output=True, text=True, check=True).stdout for phrase in phrases: require(phrase in pdf_text, f"{stem}: PDF text missing {phrase}") fonts = subprocess.run(["pdffonts", str(pdf)], capture_output=True, text=True, check=True).stdout require("DejaVu" in fonts, f"{stem}: embedded font changed") alt = (ROOT / "print/TEXT-ALTERNATIVES.md").read_text(encoding="utf-8") for phrase in ("REPORT", "PROCEDURE", "VIEWPOINT", "STORY", "WHAT THE MODEL SHOWS", "WHAT IS STILL UNKNOWN", "Tactile build"): require(phrase in alt, f"Text/tactile alternative missing {phrase}") def hashes() -> str: paths = sorted(p for p in ROOT.rglob("*") if p.is_file() and p.name != "MANIFEST.sha256" and "__pycache__" not in p.parts) return "".join(f"{hashlib.sha256(p.read_bytes()).hexdigest()} {p.relative_to(ROOT)}\n" for p in paths) 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() data, records = source() expected = make_crosswalk(data, records) path = ROOT / "CURRICULUM-CROSSWALK.md" if args.write_crosswalk: path.write_text(expected, encoding="utf-8") require(path.read_text(encoding="utf-8") == expected, "Crosswalk differs from pinned workbook") content() visuals() count = links_rights() digest = hashes() manifest = ROOT / "MANIFEST.sha256" if args.write_manifest: manifest.write_text(digest, encoding="utf-8") require(manifest.read_text(encoding="utf-8") == digest, "Hash manifest drift") print(f"PASS: 10×25-minute days; 30 core and 20 optional routes; 2 held-out checks; {len(CODES)} exact Year 4 English codes; 2 original A4 SVG/PDF/text aids; {count} local links; {len(digest.splitlines())} hashes") if __name__ == "__main__": try: main() except (AssertionError, FileNotFoundError, KeyError, subprocess.CalledProcessError) as exc: print(f"FAIL: {exc}", file=sys.stderr) raise SystemExit(1)