Coverage for tests/test_main.py: 96%

1"""Tests for exporting the FJC's files and the command line.""" 

2 

3from __future__ import annotations 

4 

5import pathlib 

6import warnings 

7 

8import pandas as pd 

9from conftest import SERVICE, tables 

10 

11import fjc_biographies 

12from fjc_biographies import __main__ as cli 

13 

14# Where the tests save the roster of the invented judges in both formats, to 

15# be looked at. The folder is not in the repository (see ".gitignore"). 

16OUTPUT = pathlib.Path(__file__).parent / 'output' 

17 

18 

19def test_save_both_formats(tmp_path: pathlib.Path) -> None: 

20 files = tables() 

21 roster = fjc_biographies.build_roster( 

22 files['service'], files['demographics'], files['career']) 

23 formats = { 

24 'roster_by_appointment.csv': roster, 

25 'roster_by_judge.csv': fjc_biographies.roster_by_judge(roster)} 

26 saved = { 

27 name: pd.read_csv(fjc_biographies.export(table, tmp_path / name)) 

28 for name, table in formats.items()} 

29 assert len(saved['roster_by_appointment.csv']) == len(SERVICE) 

30 assert saved['roster_by_judge.csv']['nid'].is_unique 

31 # The career text is kept in both files, with its line breaks. 

32 for table in saved.values(): 

33 lynch = table.loc[table['nid'] == 101, 'career'].iloc[0] 

34 assert lynch.splitlines()[1].startswith('Assistant attorney general') 

35 # A file that is open in a spreadsheet (which locks it on Windows) is 

36 # not replaced, which is not a failure of the package. 

37 for name, table in formats.items(): 

38 try: 

39 fjc_biographies.export(table, OUTPUT / name) 

40 except PermissionError: 

41 warnings.warn( 

42 f'{OUTPUT / name} was not replaced because it is open in ' 

43 f'another program, such as Excel: close it and run the tests ' 

44 f'again to see the latest version', 

45 stacklevel = 1) 

46 

47 

48def test_version() -> None: 

49 assert fjc_biographies.__version__ == '0.1.0' 

50 

51 

52def test_run(requests: list[str], data_folder: pathlib.Path) -> None: 

53 paths = fjc_biographies.run() 

54 assert list(paths) == [ 

55 'demographics', 'service', 'other_service', 'education', 'career', 

56 'other_nominations', 'roster_by_appointment', 'roster_by_judge'] 

57 assert paths['roster_by_appointment'] == data_folder / 'roster_by_appointment.csv' 

58 roster = pd.read_csv(paths['roster_by_appointment']) 

59 assert len(roster) == len(SERVICE) 

60 by_judge = pd.read_csv(paths['roster_by_judge']) 

61 assert len(by_judge) == len({row[0] for row in SERVICE}) 

62 assert by_judge.loc[by_judge['nid'] == 107, 'court_2'].iloc[0].endswith('First Circuit') 

63 # Dates are written without times. 

64 assert roster.loc[0, 'start_date'] == '1995-03-17' 

65 service = pd.read_csv(paths['service']) 

66 assert service.loc[0, 'commission_date'] == '1995-03-17' 

67 assert len(requests) == 6 

68 fjc_biographies.run(update = True) 

69 assert len(requests) == 12 

70 

71 

72def test_cli(requests: list[str], tmp_path: pathlib.Path) -> None: 

73 cli.main(['--data-folder', str(tmp_path)]) 

74 assert (tmp_path / 'roster_by_appointment.csv').is_file() 

75 assert (tmp_path / 'roster_by_judge.csv').is_file() 

76 assert len(requests) == 6 

77 cli.main(['--data-folder', str(tmp_path), '--download-only', '--update']) 

78 assert len(requests) == 12 

79 

80 

81def test_cli_download_only(requests: list[str], data_folder: pathlib.Path) -> None: 

82 cli.main(['--download-only']) 

83 assert len(requests) == 6 

84 assert not (data_folder / 'roster_by_judge.csv').exists() 

85 assert (data_folder / 'raw' / 'demographics.csv').is_file()