#!/usr/bin/env python3 """ create_dataset.py — Provenance script for the crossword-construction-logic dataset (v1). The dataset teaches the CONSTRUCTION LOGIC of crossword puzzles: numbering, crossing contracts, candidate arithmetic (MRV + constraint propagation), shadow-entry verification, duplicate rules, declared conventions, grid diagnosis, and verification discipline — in five languages (en, ru, zh, hi, ja), on machine-verified mini grids, with every machine-checkable claim re-derived at import time by the generator modules. Reads raw rows from raw.jsonl, builds sharded parquet, and writes stats.json. Re-runnable: produces identical output given the same raw.jsonl and seed. USAGE ----- python create_dataset.py Inputs: ./raw.jsonl (420 rows; composition below) ./features.json (schema: 12 fields, all string) Outputs: ./data/train-*.parquet, ./data/test-*.parquet, ./stats.json DATASET COMPOSITION ------------------- 420 rows total in 5 languages, 9 task types: - 365 positive examples (quality: good) - 33 flawed examples (quality: flawed — recoverable construction defects, presented for diagnosis) - 22 bad examples (quality: bad — hallucinated grids, phantom verification reports, script/layer confusions; for critic training) Languages: - en (English): 228 rows (the teaching backbone incl. cryptic clues) - ru (Russian): 48 rows (native Cyrillic narration) - zh (Chinese): 48 rows (native simplified-Chinese narration) - hi (Hindi): 48 rows (native Devanagari narration) - ja (Japanese): 48 rows (native Japanese narration) Task types (all languages share the same machinery): - design_grid (54), fill_grid (56), fill_trace (40), diagnose_failure (50), write_clues (60), cryptic_clue (28, en only), validate_puzzle (40), methodology_qa (85), historical_note (7) Verification discipline (the dataset's core property): - every catalog grid re-validates on import (grids.check_fill: pool membership, crossing agreement, duplicates, minimum length, no unchecked cells); - every narrated solver number (candidate counts, node ledgers, forced cascades, shadow completions) comes from the actual solver and is re-asserted at import; - a dataset-level battery (see project worklog; generate the same via the source modules) additionally re-checks claims, script purity, field consistency, and quotas against raw.jsonl itself. """ import json import sys from pathlib import Path # Allow importing from the skill's scripts/ directory SKILL_SCRIPTS = "/home/z/my-project/skills/dataset-producer/scripts" sys.path.insert(0, SKILL_SCRIPTS) try: from produce_dataset import produce_dataset, load_features except ImportError: print("ERROR: produce_dataset.py not found at " + SKILL_SCRIPTS, file=sys.stderr) sys.exit(1) # ─── Constants ───────────────────────────────────────────────────────────── DATASET_NAME = "crossword-construction-logic" RAW_PATH = Path(__file__).parent / "raw.jsonl" FEATURES_PATH = Path(__file__).parent / "features.json" SEED = 42 SPLITS = {"train": 0.85, "test": 0.15} SHARD_SIZE = "500KB" # ─── Feature schema (12 fields) ──────────────────────────────────────────── FEATURES = { "instruction": {"_type": "Value", "dtype": "string"}, "input": {"_type": "Value", "dtype": "string"}, "output": {"_type": "Value", "dtype": "string"}, "task_type": {"_type": "Value", "dtype": "string"}, "language": {"_type": "Value", "dtype": "string"}, "quality": {"_type": "Value", "dtype": "string"}, "puzzle_style": {"_type": "Value", "dtype": "string"}, "grid_size": {"_type": "Value", "dtype": "string"}, "difficulty": {"_type": "Value", "dtype": "string"}, "source_format": {"_type": "Value", "dtype": "string"}, "verified": {"_type": "Value", "dtype": "string"}, "id": {"_type": "Value", "dtype": "string"}, } # ─── Main ───────────────────────────────────────────────────────────────── def main(): print(f"=== Building dataset: {DATASET_NAME} (v1) ===\n") print(f"Step 1: Loading raw rows from {RAW_PATH}") if not RAW_PATH.exists(): print(f" ERROR: {RAW_PATH} not found.", file=sys.stderr) sys.exit(1) with open(RAW_PATH, "r", encoding="utf-8") as f: rows = [json.loads(line) for line in f if line.strip()] print(f" Loaded {len(rows)} rows\n") for field in ("task_type", "language", "quality", "puzzle_style", "grid_size", "difficulty", "source_format", "verified"): missing = [r["id"] for r in rows if field not in r] if missing: print(f" ERROR: rows missing {field} field: {missing}", file=sys.stderr) sys.exit(1) out_dir = Path(__file__).parent / "data" out_dir.mkdir(parents=True, exist_ok=True) print("Step 2: Building parquet shards and computing stats...") features = load_features(FEATURES) stats = produce_dataset( rows=rows, features=features, splits=SPLITS, out_dir=out_dir, shard_size=SHARD_SIZE, seed=SEED, ) print("\n=== Production complete ===") print(f"Total examples: {stats['total_examples']:,}") print(f"Download size: {stats['total_download_size']:,} bytes") print(f"Dataset size: {stats['total_dataset_size']:,} bytes") print(f"Size bucket: {stats['size_categories']}") print(f"\nSplits:") for name, s in stats["splits"].items(): print(f" {name}: {s['num_examples']:,} rows, {len(s['shards'])} shard(s)") from collections import Counter quality_counts = Counter(r["quality"] for r in rows) language_counts = Counter(r["language"] for r in rows) task_counts = Counter(r["task_type"] for r in rows) print(f"\nQuality breakdown:") for q in ["good", "flawed", "bad"]: print(f" {q}: {quality_counts.get(q, 0)}") print(f"\nLanguage breakdown:") for lang in ["en", "ru", "zh", "hi", "ja"]: print(f" {lang}: {language_counts.get(lang, 0)}") print(f"\nTask-type breakdown:") for t, n in task_counts.most_common(): print(f" {t}: {n}") print(f"\nProvenance:") print(f" - Methodology: construction logic ported from the Crosshare " f"app's grid/autofill semantics (AGPL-3.0 code studied, logic " f"re-implemented independently in this project's generator)") print(f" - Rows written by GLM (Super Z) under the project lead's " f"editorial direction; every machine-checkable claim generated by " f"the project solver/grid library and re-asserted at import") print(f" - Generation date: 2026-10-20") print(f" - Random seed: {SEED}") print(f" - Split fractions: {SPLITS}") print(f" - Shard size: {SHARD_SIZE}") if __name__ == "__main__": main()