#!/usr/bin/env python3 """Validator for the H1 2026 report. Run after analyze.py; exits non-zero on any failure. Checks: 1. rendered report exists, no unresolved template tokens 2. key aggregates recomputed independently match figures.json 3. per-case figures quoted in prose match the dataset records 4. every tracker URL cited resolves to a built page in ../site 5. every lawlab.africa/analysis URL matches an S3 staged analysis 6. every chart referenced exists (svg + png) 7. editorial rules: no em dashes, no banned AI-tell phrases 8. prose word count inside the 4,000-7,000 commission """ import json import re import statistics import sys from collections import Counter from pathlib import Path HERE = Path(__file__).resolve().parent SITE = HERE.parent / "site" STAGING = (HERE.parent.parent.parent / "lawlab-organized" / "data_protection_digest" / "analysis-staging") REPORT = HERE / "state-of-enforcement-h1-2026.md" fails = [] def check(ok, msg): print((" ok " if ok else "FAIL ") + msg) if not ok: fails.append(msg) text = REPORT.read_text() figures = json.loads((HERE / "figures.json").read_text()) decisions = json.loads((HERE.parent / "data" / "decisions.json").read_text()) landscape = json.loads((HERE.parent / "data" / "landscape.json").read_text()) by_id = {x["id"]: x for x in decisions} # 1 — tokens check("{{" not in text, "no unresolved template tokens") # 2 — independent recompute of key aggregates check(figures["n_total"] == f"{len(decisions):,}", "n_total matches dataset") pens = sorted(x["penalty_usd"] for x in decisions if x.get("penalty_usd")) check(figures["penalty_median_usd"] == f"USD {round(statistics.median(pens)):,}", "penalty median matches") comps = sorted(x["compensation_amount"] for x in decisions if x.get("compensation_usd")) check(figures["comp_median_kes"] == f"KES {round(statistics.median(comps)):,}", "compensation median matches") check(figures["n_laws_inforce"] == str(sum( 1 for c in landscape if (c.get("dp_law") or {}).get("status") == "in_force")), "laws-in-force count matches landscape") check(figures["n_publishing"] == str(sum( 1 for c in landscape if c.get("decisions_published"))), "publishing-regulator count matches landscape") h1 = [x for x in decisions if "2026-01" <= str(x["date_decided"])[:7] <= "2026-06"] check(figures["n_h1_2026"] == f"{len(h1):,}", "H1 2026 count matches") check(all(x["confidence"] == "verified" for x in decisions), "every dataset record is verified") # 3 — per-case figures quoted in prose match the records quoted = { "ke-high-court-2026-0001": ("KES 250,000", lambda r: True), "ng-fccpc-2024-0001": ("USD 220,000,000", lambda r: r["penalty_amount"] == 220000000 and r["penalty_currency"] == "USD"), "ng-ndpc-2025-0001": ("USD 32,800,000", lambda r: r["penalty_amount"] == 32800000 and r["penalty_currency"] == "USD"), "ng-ndpc-2025-0004": ("NGN 766,242,500", lambda r: r["penalty_amount"] == 766242500 and r["penalty_currency"] == "NGN"), "ng-ndpc-2024-0002": ("NGN 555,800,000", lambda r: r["penalty_amount"] == 555800000 and r["penalty_currency"] == "NGN"), "ng-nitda-2021-0001": ("NGN 10,000,000", lambda r: r["penalty_amount"] == 10000000 and r["penalty_currency"] == "NGN"), "za-fsca-2026-0002": ("R358,750,000", lambda r: r["penalty_amount"] == 358750000 and r["penalty_currency"] == "ZAR"), "za-ir-2023-0003": ("R5,000,000", lambda r: r["penalty_amount"] == 5000000 and r["penalty_currency"] == "ZAR"), "za-ir-2024-0007": ("R5,000,000", lambda r: r["penalty_amount"] == 5000000 and r["penalty_currency"] == "ZAR"), "ug-pdpo-2025-0001": ("UGX 300,000", lambda r: r["penalty_amount"] == 300000 and r["penalty_currency"] == "UGX"), "ke-odpc-2023-0008": ("KES 5,000,000", lambda r: r["compensation_amount"] == 5000000), "ke-odpc-2025-0013": ("KES 10,000,000", lambda r: r["compensation_amount"] == 10000000), } for rid, (figure, pred) in quoted.items(): rec = by_id.get(rid) check(rec is not None and pred(rec) and figure in text, f"quoted figure {figure} matches record {rid}") # status claims check(by_id["ng-ndpc-2024-0002"]["status"] == "appealed", "Fidelity Bank status is 'appealed' as the prose claims") check(by_id["za-ir-2024-0007"]["status"] == "appealed", "DBE status is 'appealed' as the prose claims") check("12.7 million" in by_id["ng-ndpc-2026-0001"]["summary"] and "12.7 million" in text, "Temu user figure traces to record summary") check("1,369" in by_id["ng-ndpc-2025-0005"]["respondent"] and "1,369" in text, "1,369-organisation figure traces to record") # prosecution routes exist in KE (ODPC recommendations) and MU (s.20 police # referrals) - the prose names both; anything from a third jurisdiction means # the claim needs re-editing (wave-1 lesson: pattern-claims expire with data) pros = [x for x in decisions if "prosecution" in x["outcome"] or ("police" in x["outcome"] and x["jurisdiction"] == "MU")] check(all(x["jurisdiction"] in ("KE", "MU") for x in pros), "prosecution-route records are KE or MU only (prose names both)") # 4 — tracker URLs resolve against the built S2 site tracker_urls = sorted(set(re.findall( r"https://research\.lawlab\.africa(/tracker/[a-z0-9/-]+/)", text))) for u in tracker_urls: check((SITE / u.strip("/") / "index.html").exists(), f"site page {u}") # 5 — analysis URLs match S3 staged analyses analysis_urls = sorted(set(re.findall( r"https://lawlab\.africa/analysis/([a-z0-9-]+)/", text))) for slug in analysis_urls: check((STAGING / f"{slug}.html").exists(), f"analysis staged: {slug}") # 6 — charts exist for m in sorted(set(re.findall(r"\]\((charts/[a-z0-9.-]+)\)", text))): check((HERE / m).exists(), f"chart {m}") check((HERE / m.replace(".svg", ".png")).exists(), f"chart {m} png twin") # 7 — editorial rules check("—" not in text, "zero em dashes") banned = ["delve", "robust", "crucial", "seamless", "leverage as", "game-changer", "underscores", "testament to", "at the intersection of", "dive into", "unpack", "key takeaways", "navigate the complex", "rapidly evolving", "important to note", "worth noting", "significant implications", "furthermore", "moreover", "in conclusion", "not only"] low = text.lower() for b in banned: check(b not in low, f"banned phrase absent: {b}") # 8 — word count prose = re.sub(r"\[([^\]]*)\]\([^)]*\)", r"\1", text) prose = re.sub(r"https?://\S+", " ", prose) prose = re.sub(r"[|#*_`>-]", " ", prose) n = len(prose.split()) check(4000 <= n <= 7000, f"prose word count {n:,} within 4,000-7,000") print() if fails: print(f"{len(fails)} FAILURES") sys.exit(1) print("all checks passed")