# ============================================================
# 闲言APP — 句子卡片循环重复根因深度验证
# 创建时间: 2026-05-01
# 作用: 精确模拟客户端 refreshDailySentences 的完整数据流
# ============================================================

import requests
import json
import time
import hashlib

BASE = "https://tools.wktyl.com"

def fetch_mix(mode="random", limit=5, channels=None, sort=None):
    params = {"mode": mode, "limit": limit}
    if channels:
        params["channels"] = ",".join(channels)
    if sort:
        params["sort"] = sort
    try:
        r = requests.get(f"{BASE}/api/feed/mix", params=params, timeout=10)
        data = r.json()
        if data.get("code") == 1:
            items = data.get("data", {}).get("list", [])
            return items
    except Exception as e:
        print(f"  ❌ 请求失败: {e}")
    return []

def make_id(item):
    return f"{item.get('feed_type','?')}_{item.get('id')}"

def simulate_client_flow(rounds=5, limit=5):
    """
    精确模拟客户端 refreshDailySentences 逻辑:
    1. existingIds = state.dailySentences.map((s) => s.id).toSet()
    2. fetchMix(config)
    3. dailyList = result.list.map(HomeSentence.fromFeedItem).toList()
    4. unique = dailyList.where((s) => !existingIds.contains(s.id)).toList()
    5. if unique.isEmpty && existingIds.isNotEmpty: retry with 3x limit
    6. finalList = unique.isNotEmpty ? unique : dailyList
    7. state = state.copyWith(dailySentences: finalList)
    """
    print("=" * 70)
    print("模拟客户端 refreshDailySentences 完整数据流")
    print("=" * 70)

    existing_ids = set()
    all_seen_ids = set()
    daily_sentences_ids = []

    for i in range(rounds):
        print(f"\n--- 第 {i+1} 轮 ---")
        print(f"  existingIds = {existing_ids}")

        # Step 1: fetchMix
        items = fetch_mix(mode="random", limit=limit)
        if not items:
            print("  ❌ API 返回空数据!")
            continue

        # Step 2: 构建 dailyList
        daily_list = [make_id(item) for item in items]
        print(f"  API 返回: {daily_list}")

        # Step 3: 计算 unique
        unique = [sid for sid in daily_list if sid not in existing_ids]
        print(f"  unique (不在existingIds中): {unique}")

        # Step 4: 如果 unique 为空，重试
        if not unique and existing_ids:
            print(f"  ⚠️ unique为空! 重试 (limit={limit*3})")
            retry_items = fetch_mix(mode="random", limit=limit*3)
            if retry_items:
                retry_list = [make_id(item) for item in retry_items]
                print(f"  重试API返回: {retry_list}")
                unique = [sid for sid in retry_list if sid not in existing_ids]
                print(f"  重试unique: {unique}")
                if unique:
                    daily_list = unique

        # Step 5: finalList
        final_list = unique if unique else daily_list
        print(f"  finalList: {final_list}")

        # Step 6: 更新 state
        daily_sentences_ids = final_list
        existing_ids = set(final_list)
        all_seen_ids.update(final_list)

        print(f"  更新后 dailySentences = {daily_sentences_ids}")
        print(f"  更新后 existingIds = {existing_ids}")

    print(f"\n{'='*70}")
    print(f"总结: {rounds}轮后共看到 {len(all_seen_ids)} 个不同句子")
    print(f"所有ID: {all_seen_ids}")

def simulate_with_dedup_issue(rounds=5, limit=5):
    """
    模拟潜在问题: existingIds 只包含当前 dailySentences 的ID
    而不是所有历史看过的ID，导致旧句子可能重新出现
    """
    print("\n" + "=" * 70)
    print("模拟问题场景: existingIds 只包含当前5条的ID")
    print("=" * 70)

    existing_ids = set()
    history_ids = set()

    for i in range(rounds):
        items = fetch_mix(mode="random", limit=limit)
        if not items:
            continue

        daily_list = [make_id(item) for item in items]
        unique = [sid for sid in daily_list if sid not in existing_ids]

        final_list = unique if unique else daily_list

        # 关键: existingIds 只包含当前 dailySentences
        # 之前看过的句子如果不在当前 dailySentences 中，会被视为"新"的
        old_ids_reappeared = [sid for sid in final_list if sid in history_ids and sid not in existing_ids]

        existing_ids = set(final_list)
        history_ids.update(final_list)

        print(f"  第{i+1}轮: finalList={final_list}")
        if old_ids_reappeared:
            print(f"  ⚠️ 旧句子重新出现: {old_ids_reappeared}")

def test_specific_mode_with_hottest():
    """
    测试 mode=specific + sort=hottest 是否导致数据固定
    """
    print("\n" + "=" * 70)
    print("测试: mode=specific + sort=hottest 数据是否固定")
    print("=" * 70)

    all_ids = []
    for i in range(3):
        items = fetch_mix(mode="specific", limit=5,
                         channels=["poetry", "wisdom", "story"],
                         sort="hottest")
        ids = [make_id(item) for item in items]
        all_ids.extend(ids)
        print(f"  第{i+1}次: {ids}")
        time.sleep(0.3)

    unique = len(set(all_ids))
    total = len(all_ids)
    print(f"  📊 总ID={total}, 去重={unique}, 重复率={1-unique/total:.1%}")

def test_specific_mode_with_newest():
    """
    测试 mode=specific + sort=newest 数据是否变化
    """
    print("\n" + "=" * 70)
    print("测试: mode=specific + sort=newest 数据是否变化")
    print("=" * 70)

    all_ids = []
    for i in range(3):
        items = fetch_mix(mode="specific", limit=5,
                         channels=["poetry", "wisdom", "story"],
                         sort="newest")
        ids = [make_id(item) for item in items]
        all_ids.extend(ids)
        print(f"  第{i+1}次: {ids}")
        time.sleep(0.3)

    unique = len(set(all_ids))
    total = len(all_ids)
    print(f"  📊 总ID={total}, 去重={unique}, 重复率={1-unique/total:.1%}")

def test_default_config_no_sort():
    """
    测试: 默认配置 (mode=random, limit=5, 无sort参数)
    这正是客户端当前发送的请求
    """
    print("\n" + "=" * 70)
    print("测试: 客户端当前请求参数 (mode=random, limit=5, 无sort)")
    print("=" * 70)

    all_ids = []
    for i in range(5):
        items = fetch_mix(mode="random", limit=5)
        ids = [make_id(item) for item in items]
        all_ids.extend(ids)
        print(f"  第{i+1}次: {ids}")
        time.sleep(0.3)

    unique = len(set(all_ids))
    total = len(all_ids)
    print(f"  📊 总ID={total}, 去重={unique}, 重复率={1-unique/total:.1%}")

if __name__ == "__main__":
    simulate_client_flow(rounds=5, limit=5)
    simulate_with_dedup_issue(rounds=5, limit=5)
    test_specific_mode_with_hottest()
    test_specific_mode_with_newest()
    test_default_config_no_sort()

    print("\n" + "=" * 70)
    print("🔍 根因分析总结")
    print("=" * 70)
    print()
    print("问题: 句子卡片5个句子循环重复，永远不会出现新的")
    print()
    print("可能根因:")
    print("  1. API层面: mode=specific+sort=hottest 返回固定热门数据")
    print("  2. 客户端层面: existingIds去重逻辑导致数据子集化")
    print("  3. 状态层面: didUpdateWidget未正确检测新数据")
    print("  4. 并发层面: 多次refreshDailySentences竞态条件")
    print()
    print("修复方案:")
    print("  A. FeedMixConfig添加sort字段，默认'newest'")
    print("  B. refreshDailySentences添加防重入锁")
    print("  C. 清空existingIds后再请求，确保获取全新数据")
    print("  D. 添加详细日志追踪数据流")