"""
格式校验器单元测试
确保脏数据被拦截、干净数据通过。
"""
import pytest

from pl_japanese.cleaner.validator import clean_line, validate_line


class TestCleanLine:
    """行清理：去行号前缀、注释后缀"""

    def test_strip_line_number(self):
        assert clean_line("L123  中|国:ちゅう|ごく:zhong|guo") == "中|国:ちゅう|ごく:zhong|guo"

    def test_strip_comment(self):
        assert clean_line("中|国:ちゅう|ごく:zhong|guo    # 备注") == "中|国:ちゅう|ごく:zhong|guo"

    def test_strip_both(self):
        assert clean_line("L5  中|国:ちゅう|ごく:zhong|guo  # x") == "中|国:ちゅう|ごく:zhong|guo"

    def test_plain_line(self):
        assert clean_line("中|国:ちゅう|ごく:zhong|guo") == "中|国:ちゅう|ごく:zhong|guo"


class TestValidateValid:
    """合格数据应通过"""

    @pytest.mark.parametrize("s", [
        "中|国:ちゅう|ごく:zhong|guo",
        "食|べる:た|べる:shi|",          # 假名段拼音留空
        "U|ターン:ユー|ターン:u|",        # 字母+片假名
        "日|本|語:に|ほん|ご:ri|ben|yu",
    ])
    def test_valid_lines(self, s):
        ok, reason = validate_line(s)
        assert ok, f"应通过但被拒绝: {reason}"


class TestValidateReject:
    """脏数据应被拒绝"""

    def test_reject_question_mark(self):
        ok, reason = validate_line("青|島:チンタオビール:?")
        assert not ok
        assert "?" in reason

    def test_reject_comment_marker(self):
        ok, reason = validate_line("[原] これから:これから:")
        assert not ok

    def test_reject_wrong_colon_count(self):
        ok, reason = validate_line("CS公司:シーエス|コン|ス:|gong|si")
        assert not ok

    def test_reject_length_mismatch(self):
        # 3 个冒号段但分段数不一致
        ok, reason = validate_line("大|量:たい:da|liang")
        assert not ok
        assert "长度" in reason

    def test_reject_empty(self):
        ok, reason = validate_line("")
        assert not ok

    def test_reject_empty_kanji(self):
        ok, reason = validate_line(":ちゅう:zhong")
        assert not ok

    def test_reject_empty_kana(self):
        ok, reason = validate_line("中::zhong")
        assert not ok


if __name__ == '__main__':
    pytest.main([__file__, '-v'])