re-xxm/tests/test_classify.py

# -*- coding: utf-8 -*-
"""候选分类纯函数测试: uv run python tests/test_classify.py"""

import os
import sys

sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
sys.stdout.reconfigure(encoding="utf-8")

from reaction_board.classify import (  # noqa: E402
    clean_title,
    contains_any,
    date_hits,
    matched_words,
    normalize_title,
    parse_duration,
    passes_required_title,
    score_item,
)

CLASSIFY = {
    "require_title": "咻咻满",
    "self_mid": 37754047,
    "reaction": ["re", "reaction", "反应"],
    "charge": ["加急", "特急", "充电", "专场"],
    "weak_positive": ["双人", "路人", "第一次", "首次"],
    "negative": ["歌切", "切片", "搬运", "完整版", "录播", "录像"],
    "date_negative": True,
    "duration_boost": {"gt_30min": 1, "gt_60min": 2},
}

NEUTRAL = 600          # 10min：不触发 <10min 惩罚，也不触发 >30min 加权
SHORT = 60             # 1min：<10min
MID = 31 * 60          # >30min
LONG = 61 * 60         # >60min


def score(title, up_mid=100, duration=NEUTRAL, classify=CLASSIFY):
    return score_item(
        title=title, up_mid=up_mid,
        duration_seconds=duration, classify=classify,
    )


def test_title_helpers():
    assert normalize_title("咻咻满 REACTION") == "咻咻满 reaction"
    assert clean_title('<em class="keyword">咻咻满</em> reaction') == "咻咻满 reaction"
    assert passes_required_title("【咻咻满】初听", "咻咻满")
    assert passes_required_title("咻咻满", "咻咻满")
    assert not passes_required_title("其他 UP 的 reaction", "咻咻满")
    assert contains_any("咻咻满 REACTION", ["reaction"])
    assert not contains_any("咻咻满", [])
    assert matched_words("咻咻满 REACTION 反应", ["reaction", "反应"]) == ["reaction", "反应"]
    assert matched_words("咻咻满", []) == []


def test_parse_duration():
    assert parse_duration("105:5") == 105 * 60 + 5
    assert parse_duration("1:02:03") == 3723
    assert parse_duration("15:30") == 930
    assert parse_duration(600) == 600
    assert parse_duration("bad") == 0
    assert parse_duration(None) == 0
    assert parse_duration(-5) == 0


def test_date_hits():
    assert date_hits("咻咻满-2026-10-28-素颜满") == ["2026-10-28"]
    assert date_hits("【咻咻满】10月25日直播片段") == ["10月25日"]
    assert date_hits("咻咻满 ☆ 2026.10.26") == ["2026.10.26"]
    assert date_hits("【咻咻满】2026/10/25直播回放") == ["2026/10/25"]
    assert date_hits("咻咻满：《发如雪》2024年10月5日直播歌切") == ["2024年10月5日"]
    assert date_hits("咻咻满 reaction") == []


def test_title_only_scores_zero():
    # 只过标题关（标题仅含 require_title、无任何正向词）不给分，时长也不计入
    assert score("咻咻满") == 0
    assert score("咻咻满", duration=LONG) == 0
    assert score("咻咻满 日常") == 0
    assert score("【咻咻满】交换余生（升2）") == 0
    # 但时长未知(0) vs <10min 的惩罚不同
    assert score("咻咻满", duration=0) == 0
    assert score("咻咻满", duration=SHORT) == -1


def test_hit_count_is_superlinear():
    # 正向命中次数 n：require_title（仅在有正向命中时计 1）+ reaction(0/1) + charge 词数 + weak 词数
    assert score("咻咻满 reaction") == 4
    assert score("咻咻满 反应") == 4
    assert score("咻咻满 reaction 充电") == 9
    assert score("咻咻满 reaction 加急充电专场") == 25
    assert score("咻咻满 充电") == 4
    assert score("咻咻满 加急专场") == 9
    assert score("咻咻满 双人") == 4
    assert score("咻咻满 reaction 双人") == 9


def test_reaction_typo_and_gate():
    # "re" 兼容 reaction/reactio/reation 等拼写
    assert score("咻咻满 reactio") == 4
    assert score("咻咻满 Reactuon 充电") == 9
    # 标题不含 require_title 直接 0 分
    assert score("其他 UP 的 reaction", duration=LONG) == 0
    assert score("普通视频 reactio", duration=LONG) == 0


def test_duration_boost_and_short_penalty():
    assert score("咻咻满 reaction", duration=LONG) == 6
    assert score("咻咻满 reaction", duration=MID) == 5
    assert score("咻咻满 reaction", duration=30 * 60) == 4
    assert score("咻咻满 reaction", duration=NEUTRAL) == 4
    # 时长 <10 分钟 -1
    assert score("咻咻满 reaction", duration=SHORT) == 3
    assert score("咻咻满 reaction 充电", duration=SHORT) == 8
    assert score("咻咻满 双人", duration=SHORT) == 3
    # 时长未知(0) 不扣
    assert score("咻咻满 reaction", duration=0) == 4


def test_negative_single_and_multi():
    # 单个负向词 -2（"咻咻满 切片" 无正向命中，基础为 0）
    assert score("咻咻满 切片") == -2
    assert score("咻咻满 reaction 切片") == 2
    assert score("咻咻满 reaction 直播录像") == 2
    # 2 个负向词：-2*2 -2 = -6
    assert score("咻咻满 歌切 切片") == -6
    # 3 个负向词：-2*3 -6 = -12
    assert score("咻咻满 reaction 歌切 切片 完整版") == -8


def test_self_penalty():
    assert score("咻咻满 reaction", up_mid=37754047) == 1


def test_date_negative():
    assert score("咻咻满 reaction 2026-10-28 完整录播") == -2
    assert score("咻咻满 reaction 2026-10-28") == 0
    assert score("咻咻满 充电 10月25日") == 0
    disabled = dict(CLASSIFY, date_negative=False)
    assert score("咻咻满 reaction 2026-10-28", classify=disabled) == 4


def test_removed_words():
    # 纯享不再算负向
    assert score("咻咻满 reaction 纯享受") == 4
    # 转场不再算正向（作为错别字忽略），且只过标题关不给分
    assert score("咻咻满 转场") == 0


def test_multi_hit_survives_single_negative():
    # 回归：命中 咻咻满+reaction+充电 的 204min reaction 标题，扣掉单个「切片」后仍应过线
    title = "【充电2.0 | 咻咻满reaction】翻唱和直播切片完全是在开盲盒呀！不知道今天会开到哪一款"
    assert score(title, duration=204 * 60) == 9
    assert score(title, duration=NEUTRAL) == 7


def main():
    test_title_helpers()
    test_parse_duration()
    test_date_hits()
    test_title_only_scores_zero()
    test_hit_count_is_superlinear()
    test_reaction_typo_and_gate()
    test_duration_boost_and_short_penalty()
    test_negative_single_and_multi()
    test_self_penalty()
    test_date_negative()
    test_removed_words()
    test_multi_hit_survives_single_negative()

    # 直播回放示例：命中 require+专场+路人+第一次 = 4 次 -> 16；39min > 30 -> +1 = 17
    title = "6【直播回放】咻咻满专场 路人第一次听满满唱歌 真的超级绝超级好听耶~"
    assert score(title, duration=39 * 60) == 17
    print("OK: 候选标题归一化、时长解析与分类打分测试全部通过")


if __name__ == "__main__":
    main()