import 'dart:convert';
import 'dart:io';

import 'package:csv/csv.dart';
import 'package:jadb/_data_ingestion/tanos-jlpt/objects.dart';

Future<List<JLPTRankedWord>> parseJLPTRankedWords(
  Map<String, File> files,
) async {
  final List<JLPTRankedWord> result = [];

  for (final entry in files.entries) {
    final jlptLevel = entry.key;
    final file = entry.value;

    if (!file.existsSync()) {
      throw Exception('File $jlptLevel does not exist');
    }

    final rows = await file
        .openRead()
        .transform(utf8.decoder)
        .transform(CsvToListConverter())
        .toList();

    for (final row in rows) {
      if (row.length != 3) {
        throw Exception('Invalid line in $jlptLevel: $row');
      }

      final kanji = (row[0] as String).isEmpty
          ? null
          : (row[0] as String)
                .replaceFirst(RegExp('^お・'), '')
                .replaceAll(RegExp(r'（.*）'), '');

      final readings = (row[1] as String)
          .split(RegExp('[・/、(:?s+)]'))
          .map((e) => e.trim())
          .toList();

      final meanings = (row[2] as String)
          .split(',')
          .expand(cleanMeaning)
          .toList();

      result.add(
        JLPTRankedWord(
          readings: readings,
          kanji: kanji,
          jlptLevel: jlptLevel,
          meanings: meanings,
        ),
      );
    }
  }

  return result;
}

List<String> cleanMeaning(String meaning) {
  final initialTrim = meaning.trim().replaceAll(RegExp(r'^\d.\s+'), '');
  final woParens = initialTrim.replaceAll(RegExp(r'\s*\(.*?\)\s*'), '');
  return {initialTrim, woParens}.toList();
}