scripts/validate_dataset.py

from __future__ import annotations

import argparse
import csv
from collections import Counter
from pathlib import Path


EXPECTED_COLUMNS = [
    "id",
    "input_text",
    "normalized_text",
    "task_type",
    "style",
    "domain",
    "source",
    "license",
    "split",
]

EXPECTED_SPLITS = {
    "data/train.csv": "train",
    "data/test.csv": "test",
}


def read_rows(path: Path) -> tuple[list[str], list[dict[str, str]]]:
    with path.open("r", encoding="utf-8", newline="") as handle:
        reader = csv.DictReader(handle)
        return list(reader.fieldnames or []), list(reader)


def validate_file(path: Path, expected_split: str | None) -> list[str]:
    errors: list[str] = []
    fieldnames, rows = read_rows(path)

    if fieldnames != EXPECTED_COLUMNS:
        errors.append(f"{path}: unexpected columns: {fieldnames}")

    if not rows:
        errors.append(f"{path}: file has no rows")

    seen_ids: set[str] = set()
    for index, row in enumerate(rows, start=2):
        row_id = row.get("id", "")
        if not row_id:
            errors.append(f"{path}:{index}: missing id")
        elif row_id in seen_ids:
            errors.append(f"{path}:{index}: duplicate id {row_id}")
        seen_ids.add(row_id)

        for column in EXPECTED_COLUMNS:
            if not row.get(column):
                errors.append(f"{path}:{index}: missing value in {column}")

        if expected_split and row.get("split") != expected_split:
            errors.append(f"{path}:{index}: expected split {expected_split}, got {row.get('split')}")

        if row.get("input_text", "").strip() == row.get("normalized_text", "").strip():
            errors.append(f"{path}:{index}: input_text equals normalized_text")

    return errors


def print_summary(path: Path) -> None:
    _, rows = read_rows(path)
    print(f"{path}: {len(rows):,} rows")
    for column in ("task_type", "style", "domain", "source"):
        counts = Counter(row[column] for row in rows)
        formatted = ", ".join(f"{key}={value}" for key, value in sorted(counts.items()))
        print(f"  {column}: {formatted}")


def main() -> int:
    parser = argparse.ArgumentParser(description="Validate Turkish Chat Normalization Mini CSV files.")
    parser.add_argument("paths", nargs="*", type=Path, default=[Path("data/train.csv"), Path("data/test.csv")])
    args = parser.parse_args()

    all_errors: list[str] = []
    for path in args.paths:
        expected_split = EXPECTED_SPLITS.get(path.as_posix().replace("\\", "/"))
        all_errors.extend(validate_file(path, expected_split))
        print_summary(path)

    if all_errors:
        print("\nValidation failed:")
        for error in all_errors[:50]:
            print(f"  {error}")
        if len(all_errors) > 50:
            print(f"  ... {len(all_errors) - 50} more")
        return 1

    print("\nValidation passed.")
    return 0


if __name__ == "__main__":
    raise SystemExit(main())