| """DuckDB helpers for the migration prototype.""" | |
| from __future__ import annotations | |
| from pathlib import Path | |
| from typing import Dict, List | |
| import duckdb | |
| DB_PATH = Path(__file__).resolve().parent.parent / "off_quality.db" | |
| TABLE_NAME = "nutrition_table" | |
| SCHEMA_SQL = f""" | |
| CREATE TABLE IF NOT EXISTS {TABLE_NAME} ( | |
| product_id TEXT, | |
| energy_kj DOUBLE, | |
| energy_kj_computed DOUBLE, | |
| energy_kcal DOUBLE, | |
| fat DOUBLE, | |
| saturated_fat DOUBLE, | |
| carbohydrates DOUBLE, | |
| sugars DOUBLE, | |
| starch DOUBLE, | |
| sodium DOUBLE, | |
| ingredients_text TEXT, | |
| ingredients_text_present INTEGER, | |
| contains_statement_present INTEGER, | |
| allergen_evidence_present INTEGER, | |
| fop_threshold_exceeded INTEGER, | |
| fop_symbol_present INTEGER, | |
| fop_exempt_proxy INTEGER, | |
| product_is_prepackaged_proxy INTEGER, | |
| lc TEXT, | |
| lang TEXT, | |
| language_code TEXT | |
| ) | |
| """ | |
| def connect(db_path: Path = DB_PATH) -> duckdb.DuckDBPyConnection: | |
| return duckdb.connect(db_path.as_posix()) | |
| def recreate_nutrition_table(db_path: Path = DB_PATH) -> None: | |
| """Drop and recreate the main table for deterministic runs.""" | |
| with connect(db_path) as con: | |
| con.execute(f"DROP TABLE IF EXISTS {TABLE_NAME}") | |
| con.execute(SCHEMA_SQL) | |
| def load_jsonl_to_duckdb(jsonl_path: Path, db_path: Path = DB_PATH) -> None: | |
| """Load JSONL records into DuckDB.""" | |
| with connect(db_path) as con: | |
| con.execute( | |
| f""" | |
| INSERT INTO {TABLE_NAME} | |
| SELECT | |
| product_id, | |
| energy_kj, | |
| energy_kj_computed, | |
| energy_kcal, | |
| fat, | |
| saturated_fat, | |
| carbohydrates, | |
| sugars, | |
| starch, | |
| sodium, | |
| ingredients_text, | |
| ingredients_text_present, | |
| contains_statement_present, | |
| allergen_evidence_present, | |
| fop_threshold_exceeded, | |
| fop_symbol_present, | |
| fop_exempt_proxy, | |
| product_is_prepackaged_proxy, | |
| lc, | |
| lang, | |
| language_code | |
| FROM read_json_auto(?) | |
| """, | |
| [jsonl_path.as_posix()], | |
| ) | |
| def fetch_products(db_path: Path = DB_PATH) -> List[Dict[str, object]]: | |
| """Read all products from DuckDB as dictionaries.""" | |
| with connect(db_path) as con: | |
| rows = con.execute(f"SELECT * FROM {TABLE_NAME}").fetchdf() | |
| return rows.to_dict("records") | |
| def count_rows(db_path: Path = DB_PATH) -> int: | |
| with connect(db_path) as con: | |
| value = con.execute(f"SELECT COUNT(*) FROM {TABLE_NAME}").fetchone() | |
| return int(value[0] if value else 0) | |
| def count_violations(condition_sql: str, db_path: Path = DB_PATH) -> int: | |
| """Count records matching a rule condition.""" | |
| query = f"SELECT COUNT(*) FROM {TABLE_NAME} WHERE {condition_sql}" | |
| with connect(db_path) as con: | |
| value = con.execute(query).fetchone() | |
| return int(value[0] if value else 0) | |
| def sample_violations(condition_sql: str, limit: int = 5, db_path: Path = DB_PATH) -> List[Dict[str, object]]: | |
| """Return sample violating records for dashboard drilldown.""" | |
| query = f""" | |
| SELECT * | |
| FROM {TABLE_NAME} | |
| WHERE {condition_sql} | |
| ORDER BY product_id | |
| LIMIT {int(limit)} | |
| """ | |
| with connect(db_path) as con: | |
| rows = con.execute(query).fetchdf() | |
| return rows.to_dict("records") | |