spa_test.dart

@Timeout(Duration(seconds: 1000))

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

import 'package:spa/src/spa.dart';
import 'package:test/test.dart';

bool doubleEq(double a, double b) => (a - b).abs() < 0.000001;

Future<bool> testCsv(String fileName) async {
  final file = File(fileName);
  if (!file.existsSync()) return false;

  test(fileName, () async {
    bool first = true;

    int lineNum = 0;

    await for (final line
        in const LineSplitter().bind(utf8.decoder.bind(file.openRead()))) {
      lineNum++;

      if (first) {
        // Skip column names
        first = false;
        continue;
      }

      final row = line.split(',');
      final params = SPAParams.list(row.map((e) => num.parse(e)).toList());

      final itm = SPAIntermediate();
      final output = spaCalculate(params, intermediate: itm);

      void check(double res, int idx) {
        if (!doubleEq(res, double.parse(row[idx]))) {
          throw TestFailure('Test failed at line $lineNum\n'
              'Row $idx: expected ${row[idx]} but got $res');
        }
      }

      check(output.zenith, 17);
      check(output.azimuthAstro, 18);
      check(output.azimuth, 19);
      check(output.incidence!, 20);
      check(output.sunTransit!, 21);
      check(output.sunrise!, 22);
      check(output.sunset!, 23);
    }
  });

  return true;
}

Future<void> main() async {
  // The full dataset is 1,000,000 rows containing randomly generated input
  // sets along with calculation results from the NREL C Implementation of SPA.
  //
  // It is not included in this repository due to it's size (280 MB)
  // but you download it here: https://i.tst.sh/spa/dataset.zip
  //
  // If dataset.csv is not found, fall back to the smaller version
  // which contains the first 1,000 rows of the full dataset.
  if (!await testCsv('test/dataset.csv')) {
    await testCsv('test/dataset_small.csv');
  }
}