{
  "reviews": [
    {
      "id": "R1EXAMPLE123",
      "title": "三个月实测：精准定位两个产品缺陷",
      "body": "我在这个品类测评了超过200款产品...",
      "rating": 3,
      "date": "2026-06-10",
      "helpful_votes": 147,
      "vine": false,
      "verified_purchase": true,
      "reviewer": {
        "name": "TechReviewer_Alex",
        "profile_url": "https://www.amazon.com/gp/profile/amzn1.account.xxx",
        "reviewer_rank": 89,
        "badges": ["top_contributor", "hall_of_fame"],
        "badge_categories": ["Electronics", "Computers"],
        "total_reviews": 1847,
        "helpful_votes_total": 23456
      }
    }
  ]
}

import requests
import json
from typing import List, Dict

API_KEY = "YOUR_PANGOLINFO_API_KEY"
REVIEW_API_URL = "https://api.pangolinfo.com/v1/amazon/product/reviews"

# 评论者权重系数（基于 badge 类型）
REVIEWER_WEIGHT = {
    "hall_of_fame": 10.0,
    "top_contributor": 5.0,
    "vine_voice": 4.0,
    "verified_purchase": 1.0,
    "unverified": 0.5
}

def get_reviews_with_reviewer_data(asin: str, zip_code: str = "10001") -> List[Dict]:
    """获取包含 reviewer 权威字段的完整评论数据"""
    headers = {"Authorization": f"Bearer {API_KEY}"}
    all_reviews = []

    for page in range(1, 6):
        resp = requests.get(
            REVIEW_API_URL,
            params={"asin": asin, "country": "us", "zip_code": zip_code, "page": page},
            headers=headers,
            timeout=20
        )
        if resp.status_code != 200:
            break
        reviews = resp.json().get("reviews", [])
        if not reviews:
            break
        all_reviews.extend(reviews)

    return all_reviews

def calculate_reviewer_weight(reviewer: Dict) -> float:
    """根据 badge 计算评论者权重系数"""
    badges = reviewer.get("badges", [])
    if "hall_of_fame" in badges:
        return REVIEWER_WEIGHT["hall_of_fame"]
    elif "top_contributor" in badges:
        return REVIEWER_WEIGHT["top_contributor"]
    elif "vine_voice" in badges:
        return REVIEWER_WEIGHT["vine_voice"]
    elif reviewer.get("verified_purchase"):
        return REVIEWER_WEIGHT["verified_purchase"]
    return REVIEWER_WEIGHT["unverified"]

def analyze_weighted_sentiment(asin: str) -> Dict:
    """
    基于评论者权威度的加权情感分析
    高权威评论者的意见权重更高，输出更准确的竞品口碑画像
    """
    reviews = get_reviews_with_reviewer_data(asin)

    if not reviews:
        return {"asin": asin, "error": "no reviews found"}

    weighted_rating_sum = 0.0
    weight_total = 0.0
    top_reviewer_reviews = []
    weighted_negative = []

    for review in reviews:
        reviewer = review.get("reviewer", {})
        weight = calculate_reviewer_weight(reviewer)

        weighted_rating_sum += review["rating"] * weight
        weight_total += weight

        # 提取 Top Reviewer（rank 前1000 或持有高级 badge）
        rank = reviewer.get("reviewer_rank")
        badges = reviewer.get("badges", [])
        is_top_reviewer = (
            (rank and rank <= 1000) or
            "hall_of_fame" in badges or
            "top_contributor" in badges
        )

        if is_top_reviewer:
            top_reviewer_reviews.append({
                "reviewer_name": reviewer.get("name"),
                "reviewer_rank": rank,
                "badges": badges,
                "rating": review["rating"],
                "title": review["title"],
                "helpful_votes": review.get("helpful_votes", 0)
            })

        if review["rating"] <= 3 and weight >= 4.0:
            weighted_negative.append({
                "title": review["title"],
                "body": review["body"][:300],
                "weight": weight,
                "badges": badges
            })

    weighted_avg = weighted_rating_sum / weight_total if weight_total > 0 else 0

    return {
        "asin": asin,
        "raw_avg_rating": sum(r["rating"] for r in reviews) / len(reviews),
        "weighted_avg_rating": round(weighted_avg, 2),
        "total_reviews": len(reviews),
        "top_reviewer_count": len(top_reviewer_reviews),
        "top_reviewer_reviews": sorted(
            top_reviewer_reviews,
            key=lambda x: x.get("reviewer_rank") or 9999
        )[:10],
        "high_weight_negatives": weighted_negative[:5]
    }


# 使用示例：批量分析竞品
competitor_asins = ["B08N5WRWNW", "B09XK3MJVT", "B0CK9FZBT3"]
results = {}

for asin in competitor_asins:
    print(f"[*] 分析 ASIN: {asin}")
    result = analyze_weighted_sentiment(asin)
    results[asin] = result
    print(f"    原始平均分: {result.get('raw_avg_rating', 'N/A'):.2f}")
    print(f"    加权平均分: {result.get('weighted_avg_rating', 'N/A'):.2f}")
    print(f"    Top Reviewer 数量: {result.get('top_reviewer_count', 0)}")

with open("weighted_competitor_analysis.json", "w", encoding="utf-8") as f:
    json.dump(results, f, ensure_ascii=False, indent=2)
print("\n[+] 加权竞品分析结果已保存")