tests/test_utils.py - master - dracat/chat_assembler

tests/test_utils.py
321 строка8 KB
Марченков Сергей Анатольевич
Иэрархический вывод сообщений в md
31 янв 2026, 04:36
31 янв 2026, 04:369c48888
import sys
import os
from pathlib import Path
sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))

import pytest
from chat_assembler.utils import merge_lines_prefer_new, string_to_lines as s2l, human_readable_file_size_in_bytes, sanitize_markdown_text

def test_no_changes():
    lines1 = s2l("""
a
b
c
""")
    lines2 = s2l("""
a
b
c
""")
    assert merge_lines_prefer_new(lines1, lines2) == lines2

def test_addition_in_new():
    lines1 = s2l("""
a
c
""")
    lines2 = s2l("""
a
b
c
""")
    assert merge_lines_prefer_new(lines1, lines2) == lines2

def test_deletion_in_new():
    lines1 = s2l("""
a
b
c
""")
    lines2 = s2l("""
a
c
""")
    assert merge_lines_prefer_new(lines1, lines2) == lines1

def test_preserve_user_additions_simple():
    lines1 = s2l("""
a
user_comment
b
""")
    lines2 = s2l("""
a
b
""")
    expected = s2l("""
a
user_comment
b
""")
    assert merge_lines_prefer_new(lines1, lines2) == expected

def test_preserve_user_additions_multiple_lines():
    lines1 = s2l("""
a
comment1
comment2
b
""")
    lines2 = s2l("""
a
b
c
""")
    expected = s2l("""
a
comment1
comment2
b
c
""")
    assert merge_lines_prefer_new(lines1, lines2) == expected

def test_modification():
    lines1 = s2l("""
a
b_old
c
""")
    lines2 = s2l("""
a
b_new
c
""")
    assert merge_lines_prefer_new(lines1, lines2) == lines2

def test_mixed_changes():
    lines1 = s2l("""
line1
comment1
line11
line2_old
line22
comment2
line3
""")
    lines2 = s2l("""
line1
line11
line2_new
line22
line3
line4_new
""")
    expected = s2l("""
line1
comment1
line11
line2_new
line22
comment2
line3
line4_new
""")
    assert merge_lines_prefer_new(lines1, lines2) == expected

# @pytest.mark.skip(reason="Test requires local files not present in the repository")
# def test_wtf1():
#     lines1 = s2l(Path('/mnt/work6/telegram/telegram_data/channel/ИСП_РАН___Env/General/2021-01-chat.md').read_text())
#     lines2 = s2l(Path('/mnt/work6/telegram/telegram_data/tmp-new.md').read_text())
#     reslines = merge_lines_prefer_new(lines1, lines2)

#     assert reslines == lines1



def test_empty_files_both_empty():
    lines1 = s2l("")
    lines2 = s2l("")
    assert merge_lines_prefer_new(lines1, lines2) == lines2

def test_empty_files_old_empty():
    lines1 = s2l("")
    lines2 = s2l("b")
    assert merge_lines_prefer_new(lines1, lines2) == lines2

def test_empty_files_new_empty():
    lines1 = s2l("a")
    lines2 = s2l("")
    assert merge_lines_prefer_new(lines1, lines2) == lines1

def test_user_content_at_ends():
    lines1 = s2l("""
comment_start
a
b
comment_end
""")
    lines2 = s2l("""
a
b
""")
    expected = s2l("""
comment_start
a
b
comment_end
""")
    assert merge_lines_prefer_new(lines1, lines2) == expected

@pytest.mark.parametrize("input_str, expected_bytes", [
    ("1K", 1024),
    ("5M", 5 * 1024 * 1024),
    ("7G", 7 * 1024 * 1024 * 1024),
    ("2.5k", int(2.5 * 1024)),
    ("1024", 1024),
    (" 1G ", 1024**3),
    ("1GB", 1024**3),
    ("", 0),
    ("0", 0),
])
def test_human_readable_file_size_in_bytes_valid(input_str, expected_bytes):
    assert human_readable_file_size_in_bytes(input_str) == expected_bytes

@pytest.mark.parametrize("invalid_input", [
    "1X",
    "abc",
    "1MK",
])
def test_human_readable_file_size_in_bytes_invalid(invalid_input):
    with pytest.raises(ValueError):
        human_readable_file_size_in_bytes(invalid_input)


@pytest.mark.parametrize("char_code", ['\u200e', '\u200f'])
def test_sanitize_markdown_text_removes_bidi_marks(char_code):
    text_with_mark = f"Hello{char_code} World"
    expected_text = "Hello World"
    assert sanitize_markdown_text(text_with_mark) == expected_text


# Find all test case directories in the examples directory
EXAMPLE_DIR = Path(__file__).parent / "sanitize_examples"
TEST_DIRS = sorted([d for d in EXAMPLE_DIR.iterdir() if d.is_dir()])
if not TEST_DIRS:
    raise RuntimeError(f"No test case directories found in {EXAMPLE_DIR}")


@pytest.mark.parametrize("test_dir", TEST_DIRS, ids=[d.name for d in TEST_DIRS])
def test_sanitize_markdown_text(test_dir):
    """
Перебираем все каталоги в `test_dir` отсортированные алфавитно
- Если есть в каталоге `data.json` и `data.md`
    - считываем `data.json`, берем message из поля "message"
    - сравниваем `sanitize_markdown(message)` с содержимым файла `data.md` 
        - без учета конечного вайтспейса и переводов строк
    """
    md_path = test_dir / "data.md"
    json_path = test_dir / "data.json"

    if not md_path.exists() or not json_path.exists():
        pytest.skip(f"Missing data.md or data.json in {test_dir}")

    import json
    json_data = json.loads(json_path.read_text())
    input_message = json_data.get("message")

    expected_markdown = md_path.read_text()
    actual_markdown = sanitize_markdown_text(input_message)

    assert actual_markdown.strip() == expected_markdown.strip()


def test_sanitize_markdown_text_adds_newline_before_setext_separators():
    # Test for '---'
    text_with_separator = "Some text\n---\nMore text"
    expected_text = "Some text\n\n---\nMore text"
    assert sanitize_markdown_text(text_with_separator) == expected_text

    text_with_separator_and_space = "Some text\n\n---\nMore text"
    assert sanitize_markdown_text(text_with_separator_and_space) == text_with_separator_and_space

    text_with_long_separator = "Some text\n-----\nMore text"
    expected_long_separator = "Some text\n\n-----\nMore text"
    assert sanitize_markdown_text(text_with_long_separator) == expected_long_separator

    # Test for '==='
    text_with_equals = "Some text\n===\nMore text"
    expected_equals = "Some text\n\n===\nMore text"
    assert sanitize_markdown_text(text_with_equals) == expected_equals

    text_with_equals_and_space = "Some text\n\n===\nMore text"
    assert sanitize_markdown_text(text_with_equals_and_space) == text_with_equals_and_space

    text_with_short_equals = "Some text\n==\nMore text"
    expected_short_equals = "Some text\n\n==\nMore text"
    assert sanitize_markdown_text(text_with_short_equals) == expected_short_equals


# Reply formatting tests
from chat_assembler.utils import format_reply_message


def test_format_reply_message():
    """Test reply message formatting"""
    # Mock messages
    all_messages = {
        137: {
            "message": "Кстати, вы всем офисом уезжаете?",
            "sender": {"username": "Sakrat91", "id": 169164243}
        }
    }
    
    # Reply message
    reply_msg = {
        "id": 140,
        "message": "да не, леста-то никуда не денется",
        "sender": {"username": "lanvin_s", "id": 124114202},
        "reply_to": {"reply_to_msg_id": 137}
    }
    
    result = format_reply_message(reply_msg, all_messages)
    expected = "да не, леста-то никуда не денется"
    assert result == expected


def test_format_reply_message_no_original():
    """Test reply when original message doesn't exist"""
    all_messages = {}
    
    reply_msg = {
        "id": 140,
        "message": "да не, леста-то никуда не денется",
        "sender": {"username": "lanvin_s", "id": 124114202},
        "reply_to": {"reply_to_msg_id": 999}
    }
    
    result = format_reply_message(reply_msg, all_messages)
    expected = "да не, леста-то никуда не денется"
    assert result == expected


def test_format_reply_message_multiline():
    """Test reply with multiline original message"""
    all_messages = {
        37: {
            "message": "Думаю, после первых встреч станет понятнее, куда эта история может привести. \n\nГлавное набросать небольшой пул правил, чтобы было проще.",
            "sender": {"username": "VErmakova", "id": 519704541}
        }
    }
    
    reply_msg = {
        "id": 39,
        "message": "Согласен на все 100%",
        "sender": {"username": "nvandreev", "id": 49236267},
        "reply_to": {"reply_to_msg_id": 37}
    }
    
    result = format_reply_message(reply_msg, all_messages)
    expected = "Согласен на все 100%"
    assert result == expected