""" Unit tests for the ascii (n-column) reader """ import os.path import warnings import math import unittest from sasdata.dataloader.loader import Loader from sasdata.dataloader.data_info import Data2D warnings.simplefilter("ignore") def find(filename): return os.path.join(os.path.dirname(__file__), 'data', filename) class ABSReaderTests(unittest.TestCase): def setUp(self): self.loader = Loader() self.f1_list = self.loader.load(find("ascii_test_1.txt")) self.f1 = self.f1_list[0] self.f2_list = self.loader.load(find("ascii_test_2.txt")) self.f2 = self.f2_list[0] self.f3_list = self.loader.load(find("ascii_test_3.txt")) self.f3 = self.f3_list[0] self.f4_list = self.loader.load(find("ascii_test_4.abs")) self.f4 = self.f4_list[0] self.f5_list = self.loader.load(find("ascii_test_5.txt")) self.f5 = self.f5_list[0] def test_checkdata(self): """ Test .ABS file loaded as ascii """ # The length of the data is 10 self.assertEqual(len(self.f1_list), 1) self.assertEqual(len(self.f2_list), 1) self.assertEqual(len(self.f3_list), 1) self.assertEqual(len(self.f4_list), 1) self.assertEqual(len(self.f5_list), 1) self.assertEqual(len(self.f1.x), 10) self.assertEqual(self.f1.x[0],0.002618) self.assertEqual(self.f1.x[9],0.0497) self.assertTrue(self.f1.x_unit == 'A^{-1}') self.assertTrue(self.f1.y_unit == 'cm^{-1}') self.assertEqual(self.f1.meta_data['loader'],"ASCII") def test_truncated_1(self): """ Test an ascii file with header and a comment line in the middle of the data section. The business rule says that we should stop reading at the first comment once the data section has started (and treat the comment as though it were the start of a footer). """ # The length of the data is 5 self.assertEqual(len(self.f2.x), 5) self.assertEqual(self.f2.x[0],0.002618) self.assertEqual(self.f2.x[4],0.02356) def test_truncated_2(self): """ Test a 6-col ascii file with header and a line with only 2 columns in the middle of the data section. The business rule says that we should stop reading at the first inconsitent line. """ # The length of the data is 5 self.assertEqual(len(self.f3.x), 5) self.assertEqual(self.f3.x[0],0.002618) self.assertEqual(self.f3.x[4],0.02356) def test_truncated_3(self): """ Test a 6-col ascii file with complex header and many lines with 2 or 2 columns in the middle of the data section. The business rule says that we should stop reading at the last line of header. """ # The length of the data is 5 self.assertEqual(len(self.f4.x), 5) self.assertEqual(self.f4.x[0],0.012654) self.assertEqual(self.f4.x[4],0.02654) def test_truncated_4(self): """ Test mix of 6-col and 2-col. Only the last 5 2-col lines should be read. """ # The length of the data is 5 self.assertEqual(len(self.f5.x), 5) self.assertEqual(self.f5.x[0],0.02879) self.assertEqual(self.f5.x[4],0.0497) def test_truncated_5(self): """ Test a 6-col ascii file with complex header where one of them has a letter and many lines with 2 or 2 columns in the middle of the data section. Will be rejected because fewer than 5 lines. """ # Test .ABS file loaded as ascii f = None try: f = self.loader.load(find("ascii_test_6.txt")) # The length of the data is 5 except: self.assertEqual(f, None) def test_nan_values(self): """ Test loading an ascii data file with nan values saved in x, y, or dy. """ f_1d = self.loader.load(find("nans_in_1d_data.dat"))[0] f_2d = self.loader.load(find("nans_in_2d_data.DAT"))[0] for i in range(0, len(f_1d.x) - 1): self.assertFalse(math.isnan(f_1d.x[i])) self.assertFalse(math.isnan(f_1d.y[i])) self.assertFalse(math.isnan(f_1d.dy[i])) self.assertTrue(isinstance(f_2d, Data2D)) f_2d.data = f_2d.data.flatten() f_2d.qx_data = f_2d.qx_data.flatten() f_2d.qy_data = f_2d.qy_data.flatten() for i in range(0, len(f_2d.data) - 1): self.assertFalse(math.isnan(f_2d.data[i])) self.assertFalse(math.isnan(f_2d.qx_data[i])) self.assertFalse(math.isnan(f_2d.qy_data[i])) def test_encoding(self): # Compare loading of utf-8 versus ansi/windows-1252 ansi = self.loader.load(find("encoding_ANSI.txt")) utf8 = self.loader.load(find("encoding_UTF_8.txt")) self._check_cl_data(ansi) self._check_cl_data(utf8) def _check_cl_data(self, data): self.assertEqual(len(data), 1) data_set = data[0] self.assertEqual(len(data_set.x), 969) self.assertEqual(len(data_set.x), len(data_set.y)) self.assertEqual(data_set.x[0], 0.000201634) self.assertEqual(data_set.y[0], 208487062.3) self.assertEqual(data_set.x_unit, 'A^{-1}') self.assertEqual(data_set.y_unit, 'cm^{-1}') def test_save_ascii(self): f_name = "test_f1_output.txt" f_name_csv = "test_f1_output.csv" # Save and load text file self.loader.save(f_name, self.f1, None) self.assertTrue(os.path.isfile(f_name)) reload = self.loader.load(f_name) f1_reload = reload[0] # Compare data from saved text file to previously loaded data self.assertEqual(len(self.f1.x), len(f1_reload.x)) self.assertEqual(len(self.f1.y), len(f1_reload.y)) self.assertEqual(self.f1.y[0], f1_reload.y[0]) # Save and load csv file self.loader.save(f_name_csv, self.f1, None) self.assertTrue(os.path.isfile(f_name_csv)) reload_csv = self.loader.load(f_name_csv) f1_reload_csv = reload_csv[0] # Compare data from saved csv file to previously loaded data self.assertEqual(len(self.f1.x), len(f1_reload_csv.x)) self.assertEqual(len(self.f1.y), len(f1_reload_csv.y)) self.assertEqual(self.f1.y[0], f1_reload_csv.y[0]) # Destroy generated files os.remove(f_name) self.assertFalse(os.path.isfile(f_name)) os.remove(f_name_csv) self.assertFalse(os.path.isfile(f_name_csv)) if __name__ == '__main__': unittest.main()