"""Black-box acceptance checks for the documented reporting CLI contract.""" import csv import io import os from pathlib import Path import re import subprocess import sys import pytest PROJECT = Path(__file__).resolve().parents[1] HEADER = "date,project,category,minutes\n" VALID_ROW = "2026-10-19,Orion,Development,7\n" def run_cli(path): # Deliberately do not inherit API credentials or Python startup settings. env = { "PATH": str(Path(sys.executable).parent), "PYTHONIOENCODING": "utf-8", "PYTHONDONTWRITEBYTECODE": "1", "PYTHONHASHSEED": "0", } for name in ("SystemRoot", "WINDIR"): if name in os.environ: env[name] = os.environ[name] return subprocess.run( [sys.executable, "-B", str(PROJECT / "report.py"), str(path)], cwd=PROJECT, env=env, capture_output=True, timeout=3, check=False, ) def input_file(tmp_path, text): path = tmp_path / "activity.csv" path.write_bytes(text.encode("utf-8")) return path def assert_success(result, expected): assert result.returncode == 0, result.stderr.decode("utf-8", errors="replace") assert result.stdout == expected.encode("utf-8") assert result.stderr == b"" def assert_input_error(result, *, field=None, row=None, kind=None): assert result.returncode == 2 assert result.stdout == b"", "Invalid input must not emit a partial report" message = result.stderr.decode("utf-8") assert message.startswith("Input error:"), message assert message.removeprefix("Input error:").strip(), message assert "Traceback" not in message if field is not None: assert re.search(rf"\b{re.escape(field)}\b", message, re.IGNORECASE), message if row is not None: assert re.search(rf"\b(?:row|record)\s*(?:#|:)?\s*{row}\b", message, re.IGNORECASE), message if kind is not None: assert re.search(kind, message, re.IGNORECASE), message def csv_record(values): output = io.StringIO(newline="") csv.writer(output, lineterminator="\n").writerow(values) return output.getvalue() def test_sample_matches_exact_golden_report(): expected = (PROJECT / "expected" / "activity-report.txt").read_bytes() result = run_cli(PROJECT / "data" / "activity.csv") assert result.returncode == 0, result.stderr.decode("utf-8", errors="replace") assert result.stdout == expected assert result.stderr == b"" def test_columns_can_be_reordered_and_blank_lines_ignored(tmp_path): source = ( "minutes,category,date,project\r\n" "\r\n" "9,Testing,2026-10-19,Lumen\r\n" "\r\n" "4,Testing,2026-10-20,Lumen\r\n" ) assert_success(run_cli(input_file(tmp_path, source)), "Project activity report\nEntries: 2\nTotal minutes: 13\n\n" "By project\n- Lumen: 13\n\nBy category\n- Testing: 13\n") @pytest.mark.parametrize("source,expected", [ (HEADER, "Project activity report\nEntries: 0\nTotal minutes: 0\n\n" "By project\n(none)\n\nBy category\n(none)\n"), (HEADER + "2026-10-19,Lumen,Testing,0\n", "Project activity report\nEntries: 1\nTotal minutes: 0\n\n" "By project\n- Lumen: 0\n\nBy category\n- Testing: 0\n"), ], ids=["header-only", "zero-minutes"]) def test_zero_reports(tmp_path, source, expected): assert_success(run_cli(input_file(tmp_path, source)), expected) def test_group_names_use_python_string_order(tmp_path): source = HEADER + ( "2026-10-19,álpha,Ωmega,5\n" "2026-10-19,alpha,zed,3\n" "2026-10-19,Alpha,Zed,2\n" "2026-10-19,_work,zed,0\n" ) assert_success(run_cli(input_file(tmp_path, source)), "Project activity report\nEntries: 4\nTotal minutes: 10\n\n" "By project\n- Alpha: 2\n- _work: 0\n- alpha: 3\n- álpha: 5\n\n" "By category\n- Zed: 2\n- zed: 3\n- Ωmega: 5\n") def test_repeated_rows_are_counted(tmp_path): assert_success(run_cli(input_file(tmp_path, HEADER + VALID_ROW * 3)), "Project activity report\nEntries: 3\nTotal minutes: 21\n\n" "By project\n- Orion: 21\n\nBy category\n- Development: 21\n") def test_labels_preserve_spaces_case_commas_and_quotes(tmp_path): source = HEADER + csv_record(["2026-10-19", ' Orion, "UI" ', ' Docs, "API" ', "8"]) source += csv_record(["2026-10-19", "Orion", "docs", "2"]) assert_success(run_cli(input_file(tmp_path, source)), "Project activity report\nEntries: 2\nTotal minutes: 10\n\n" 'By project\n- Orion, "UI" : 8\n- Orion: 2\n\n' 'By category\n- Docs, "API" : 8\n- docs: 2\n') def test_calendar_and_ascii_minutes_boundaries(tmp_path): source = HEADER + ( "0001-01-01,Orion,Testing,0007\n" "2000-02-29,Orion,Testing,0\n" "2024-02-29,Orion,Testing,1000000000000\n" "9999-12-31,Orion,Testing,3\n" ) assert_success(run_cli(input_file(tmp_path, source)), "Project activity report\nEntries: 4\nTotal minutes: 1000000000010\n\n" "By project\n- Orion: 1000000000010\n\n" "By category\n- Testing: 1000000000010\n") @pytest.mark.parametrize("source", [ "", "project,category,minutes\nOrion,Testing,7\n", "date,category,minutes\n2026-10-19,Testing,7\n", "date,project,minutes\n2026-10-19,Orion,7\n", "date,project,category\n2026-10-19,Orion,Testing\n", "date,project,category,minutes,minutes\n2026-10-19,Orion,Testing,7,7\n", "date,project,category,minutes,note\n2026-10-19,Orion,Testing,7,x\n", "date,project,category,Minutes\n2026-10-19,Orion,Testing,7\n", ], ids=["empty", "missing-date", "missing-project", "missing-category", "missing-minutes", "duplicate", "extra", "case-mismatch"]) def test_invalid_headers(tmp_path, source): assert_input_error(run_cli(input_file(tmp_path, source)), kind=r"header|column|CSV") @pytest.mark.parametrize("bad_row", [ "2026-10-20,Orion,Testing\n", "2026-10-20,Orion,Testing,3,extra\n", ], ids=["short", "long"]) def test_invalid_row_width_after_valid_row(tmp_path, bad_row): result = run_cli(input_file(tmp_path, HEADER + VALID_ROW + bad_row)) assert_input_error(result, row=3, kind=r"column|cell|field|width|count|CSV|expected") @pytest.mark.parametrize("date", [ "2026-2-03", "26-02-03", "2026/02/03", "2026-02-30", "2025-02-29", "1900-02-29", "0000-01-01", "2026-13-01", " 2026-10-19", ], ids=["month-width", "year-width", "separator", "impossible-day", "non-leap", "century-non-leap", "year-zero", "month-range", "leading-space"]) def test_invalid_dates_after_valid_row(tmp_path, date): source = HEADER + VALID_ROW + csv_record([date, "Orion", "Testing", "3"]) assert_input_error(run_cli(input_file(tmp_path, source)), row=3, field="date") @pytest.mark.parametrize("minutes", [ "-1", "+1", "1.0", "1e2", " 1", "1 ", "", "NaN", "١", "１", "1_000", ], ids=["negative", "plus", "decimal", "exponent", "leading-space", "trailing-space", "empty", "nan", "arabic-digit", "fullwidth-digit", "underscore"]) def test_invalid_minutes_after_valid_row(tmp_path, minutes): source = HEADER + VALID_ROW + csv_record(["2026-10-20", "Orion", "Testing", minutes]) assert_input_error(run_cli(input_file(tmp_path, source)), row=3, field="minutes") @pytest.mark.parametrize("field,value", [ ("project", ""), ("project", " "), ("category", ""), ("category", "\t "), ], ids=["empty-project", "spaces-project", "empty-category", "whitespace-category"]) def test_blank_labels_after_valid_row(tmp_path, field, value): values = {"date": "2026-10-20", "project": "Orion", "category": "Testing", "minutes": "3"} values[field] = value source = HEADER + VALID_ROW + csv_record(list(values.values())) assert_input_error(run_cli(input_file(tmp_path, source)), row=3, field=field) def test_error_rows_count_csv_records_not_physical_lines(tmp_path): source = HEADER + csv_record(["2026-10-19", "Orion\nteam", "Testing", "7"]) source += "2026-10-20,Orion,Testing,wrong\n" assert_input_error(run_cli(input_file(tmp_path, source)), row=3, field="minutes") @pytest.mark.parametrize("bad_row", [ '2026-10-20,"Orion,Testing,3\n', '2026-10-20,"Orion"junk,Testing,3\n', ], ids=["unterminated-quote", "text-after-quote"]) def test_malformed_csv_after_valid_row(tmp_path, bad_row): source = HEADER + VALID_ROW + bad_row assert_input_error(run_cli(input_file(tmp_path, source)), kind=r"CSV") @pytest.mark.parametrize("kind", ["missing", "directory"], ids=["missing-file", "directory"]) def test_unreadable_input(tmp_path, kind): path = tmp_path / "missing.csv" if kind == "missing" else tmp_path assert_input_error(run_cli(path), kind=r"read|file|UTF-8") def test_non_utf8_input(tmp_path): path = tmp_path / "invalid-encoding.csv" path.write_bytes(HEADER.encode("ascii") + b"2026-10-19,Orion,Testing,\xff\n") assert_input_error(run_cli(path), kind=r"UTF-8|encoding|read") def test_help_is_usable(): result = run_cli("--help") assert result.returncode == 0 assert result.stderr == b"" help_text = result.stdout.decode("utf-8") assert "usage:" in help_text.lower() assert "csv" in help_text.lower() def test_large_minutes_above_decimal_conversion_limit(tmp_path): minutes = "1" + "0" * 4299 + "7" source = HEADER + f"2026-10-19,Orion,Testing,{minutes}\n" expected = ( f"Project activity report\nEntries: 1\nTotal minutes: {minutes}\n\n" f"By project\n- Orion: {minutes}\n\n" f"By category\n- Testing: {minutes}\n" ) assert_success(run_cli(input_file(tmp_path, source)), expected) def test_large_minutes_sum_crosses_decimal_conversion_limit(tmp_path): minutes = "9" * 4300 total = "1" + "0" * 4300 source = HEADER + ( f"2026-10-19,Orion,Testing,{minutes}\n" "2026-10-20,Orion,Testing,1\n" ) expected = ( f"Project activity report\nEntries: 2\nTotal minutes: {total}\n\n" f"By project\n- Orion: {total}\n\n" f"By category\n- Testing: {total}\n" ) assert_success(run_cli(input_file(tmp_path, source)), expected) @pytest.mark.parametrize("minutes,total", [ ("0" * 5000 + "7", "7"), ("0" * 5001, "0"), ], ids=["nonzero", "zero"]) def test_large_minutes_leading_zeros_render_canonically(tmp_path, minutes, total): source = HEADER + f"2026-10-19,Orion,Testing,{minutes}\n" expected = ( f"Project activity report\nEntries: 1\nTotal minutes: {total}\n\n" f"By project\n- Orion: {total}\n\n" f"By category\n- Testing: {total}\n" ) assert_success(run_cli(input_file(tmp_path, source)), expected) @pytest.mark.parametrize("minutes", [ "999999999", "1000000000", "999999999999999999", "1000000000000000001", ], ids=["nine-digits", "ten-digits", "eighteen-digits", "zero-middle-block"]) def test_large_minutes_decimal_block_boundaries(tmp_path, minutes): source = HEADER + f"2026-10-19,Orion,Testing,{minutes}\n" expected = ( f"Project activity report\nEntries: 1\nTotal minutes: {minutes}\n\n" f"By project\n- Orion: {minutes}\n\n" f"By category\n- Testing: {minutes}\n" ) assert_success(run_cli(input_file(tmp_path, source)), expected)