gitcode-insight 0.1.0__py3-none-any.whl

gitcode_insight/pr.py

@@ -0,0 +1,843 @@
+# -*- coding: utf-8 -*-
+"""
+GitCode PR 洞察模块
+分析指定仓库近 N 天的 PR 情况
+"""
+
+import json
+import time
+import os
+from datetime import datetime, timedelta, timezone
+from typing import List, Dict, Optional
+import requests
+
+from .utils import request_with_retry
+
+
+class GitCodePRInsight:
+    """GitCode PR 洞察分析器"""
+
+    def __init__(self, repo: str, token: str, owner: str = None, days: int = 30, output_dir: str = None):
+        """
+        初始化
+
+        Args:
+            repo: 仓库名称（path）
+            token: API 访问令牌
+            owner: 组织名
+            days: 统计天数
+            output_dir: 输出目录
+        """
+        self.repo = repo
+        self.token = token
+        self.owner = owner or self._get_default_owner()
+        self.days = days
+        self.base_url = "https://api.gitcode.com/api/v5"
+
+        # 设置输出目录
+        if output_dir is None:
+            output_dir = os.path.join(os.getcwd(), "output")
+        self.output_dir = output_dir
+
+        self.session = requests.Session()
+        self.session.headers.update({"Content-Type": "application/json"})
+
+        # 计算时间范围
+        self.since_date = (datetime.now(timezone.utc) - timedelta(days=days)).isoformat()
+
+    def _get_default_owner(self) -> str:
+        """从配置文件获取默认 owner"""
+        config_file = os.path.join(os.getcwd(), "config", "gitcode.json")
+        if os.path.exists(config_file):
+            with open(config_file, 'r', encoding='utf-8') as f:
+                config = json.load(f)
+                return config.get("owner", "")
+        return ""
+
+    def get_prs(self) -> List[Dict]:
+        """获取 PR 列表"""
+        print(f"获取 {self.owner}/{self.repo} 近 {self.days} 天的 PR 列表...")
+
+        url = f"{self.base_url}/repos/{self.owner}/{self.repo}/pulls"
+        all_prs = []
+        page = 1
+        max_pages = 50
+
+        while page <= max_pages:
+            params = {
+                "access_token": self.token,
+                "state": "all",
+                "per_page": 100,
+                "page": page,
+                "sort": "created",
+                "direction": "desc"
+            }
+
+            data = request_with_retry(self.session, url, params)
+            if data is None:
+                break
+
+            if not isinstance(data, list) or len(data) == 0:
+                break
+
+            # 过滤在时间范围内的 PR
+            filtered = [
+                pr for pr in data
+                if self._is_within_range(pr.get("created_at", ""))
+                or self._is_within_range(pr.get("updated_at", ""))
+            ]
+
+            all_prs.extend(filtered)
+            print(f"  第 {page} 页获取到 {len(filtered)} 条 PR")
+
+            if len(data) < 100:
+                break
+
+            page += 1
+            time.sleep(0.6)  # API 限流控制
+
+        print(f"共获取到 {len(all_prs)} 条 PR")
+        return all_prs
+
+    def _is_within_range(self, date_str: str) -> bool:
+        """检查日期是否在统计范围内"""
+        if not date_str:
+            return False
+        try:
+            pr_date = datetime.fromisoformat(date_str.replace("Z", "+00:00"))
+            since = datetime.fromisoformat(self.since_date.replace("Z", "+00:00"))
+            return pr_date >= since
+        except:
+            return False
+
+    def get_pr_comments(self, pr_number: int) -> List[Dict]:
+        """获取 PR 评论列表"""
+        url = f"{self.base_url}/repos/{self.owner}/{self.repo}/pulls/{pr_number}/comments"
+        params = {"access_token": self.token, "per_page": 100}
+        data = request_with_retry(self.session, url, params)
+        return data if isinstance(data, list) else []
+
+    def analyze_pr(self, pr: Dict) -> Dict:
+        """分析单个 PR，计算各项指标"""
+        pr_number = pr.get("number", "")
+        created_at_str = pr.get("created_at", "")
+        merged_at = pr.get("merged_at")
+        closed_at = pr.get("closed_at")
+
+        # 获取评论计算首次评审时间
+        comments = self.get_pr_comments(pr_number)
+
+        first_review_time = None
+        if comments:
+            creator_id = pr.get("user", {}).get("id")
+            for comment in comments:
+                commenter_id = comment.get("user", {}).get("id")
+                if commenter_id != creator_id:
+                    try:
+                        comment_time = datetime.fromisoformat(comment.get("created_at", "").replace("Z", "+00:00"))
+                        created_time = datetime.fromisoformat(created_at_str.replace("Z", "+00:00"))
+                        first_review_time = (comment_time - created_time).total_seconds() / 60
+                    except:
+                        pass
+                    break
+
+        # 计算合并耗时
+        merge_duration = None
+        if merged_at and created_at_str:
+            try:
+                merged_time = datetime.fromisoformat(merged_at.replace("Z", "+00:00"))
+                created_time = datetime.fromisoformat(created_at_str.replace("Z", "+00:00"))
+                merge_duration = (merged_time - created_time).total_seconds() / 60
+            except:
+                pass
+
+        # 计算关闭耗时（未合并的 PR）
+        close_duration = None
+        if closed_at and not merged_at and created_at_str:
+            try:
+                closed_time = datetime.fromisoformat(closed_at.replace("Z", "+00:00"))
+                created_time = datetime.fromisoformat(created_at_str.replace("Z", "+00:00"))
+                close_duration = (closed_time - created_time).total_seconds() / 60
+            except:
+                pass
+
+        # 计算打开天数
+        open_days = None
+        if pr.get("state") == "open":
+            try:
+                created_time = datetime.fromisoformat(created_at_str.replace("Z", "+00:00"))
+                open_days = (datetime.now(timezone.utc) - created_time).total_seconds() / 86400
+            except:
+                pass
+
+        # 代码变更行数
+        added_lines = pr.get("added_lines", 0) or 0
+        removed_lines = pr.get("removed_lines", 0) or 0
+        total_changes = added_lines + removed_lines
+
+        # 提取评审者
+        assignees = [a.get("login", "") for a in pr.get("assignees", [])]
+        testers = [t.get("login", "") for t in pr.get("testers", [])]
+
+        # 提取标签
+        labels = [l.get("name", "") for l in pr.get("labels", [])]
+
+        # 合并者
+        merged_by = pr.get("merged_by", {})
+        merged_by_login = merged_by.get("login", "") if merged_by else ""
+
+        return {
+            "pr_number": pr_number,
+            "title": pr.get("title", ""),
+            "state": pr.get("state", ""),
+            "draft": pr.get("draft", False),
+            "locked": pr.get("locked", False),
+            "created_at": created_at_str,
+            "updated_at": pr.get("updated_at", ""),
+            "merged_at": merged_at or "",
+            "closed_at": closed_at or "",
+            "creator": pr.get("user", {}).get("login", ""),
+            "source_branch": pr.get("source_branch", ""),
+            "target_branch": pr.get("target_branch", ""),
+            "added_lines": added_lines,
+            "removed_lines": removed_lines,
+            "total_changes": total_changes,
+            "notes_count": pr.get("notes", 0) or 0,
+            "labels": ",".join(labels),
+            "assignees": ",".join(assignees),
+            "testers": ",".join(testers),
+            "merged_by": merged_by_login,
+            "mergeable": pr.get("mergeable"),
+            "pipeline_status": pr.get("pipeline_status", ""),
+            "html_url": pr.get("html_url", ""),
+            "first_review_time": round(first_review_time, 2) if first_review_time else None,
+            "merge_duration": round(merge_duration, 2) if merge_duration else None,
+            "close_duration": round(close_duration, 2) if close_duration else None,
+            "open_days": round(open_days, 2) if open_days else None
+        }
+
+    def calculate_insights(self, prs_data: List[Dict]) -> tuple:
+        """计算洞察指标"""
+        total = len(prs_data)
+
+        # 状态分布
+        opened = [p for p in prs_data if p["state"] == "open"]
+        merged = [p for p in prs_data if p["merged_at"]]
+        closed_not_merged = [p for p in prs_data if p["state"] == "closed" and not p["merged_at"]]
+
+        # 冲突 PR
+        conflicts = [p for p in prs_data if p["mergeable"] is False]
+
+        # 代码变更统计
+        change_sizes = [p["total_changes"] for p in prs_data if p["total_changes"] > 0]
+        avg_changes = sum(change_sizes) / len(change_sizes) if change_sizes else 0
+        max_changes = max(change_sizes) if change_sizes else 0
+        large_prs = [p for p in prs_data if p["total_changes"] > 500]
+
+        # 评论密度
+        total_notes = sum(p["notes_count"] for p in prs_data)
+        total_lines = sum(p["total_changes"] for p in prs_data)
+        comment_density = total_notes / total_lines if total_lines > 0 else 0
+
+        # 评审时间统计
+        review_times = [p["first_review_time"] for p in prs_data if p["first_review_time"]]
+        avg_review_time = sum(review_times) / len(review_times) if review_times else 0
+
+        # 合并耗时统计（分钟）
+        merge_durations = [p["merge_duration"] for p in prs_data if p["merge_duration"]]
+        avg_merge_duration = sum(merge_durations) / len(merge_durations) if merge_durations else 0
+        min_merge_duration = min(merge_durations) if merge_durations else 0
+        max_merge_duration = max(merge_durations) if merge_durations else 0
+
+        # 创建者分布
+        creator_dist = {}
+        for pr in prs_data:
+            creator = pr["creator"]
+            if creator:
+                creator_dist[creator] = creator_dist.get(creator, 0) + 1
+
+        # 目标分支分布
+        branch_dist = {}
+        for pr in prs_data:
+            branch = pr["target_branch"]
+            if branch:
+                branch_dist[branch] = branch_dist.get(branch, 0) + 1
+
+        # 标签分布
+        label_dist = {}
+        for pr in prs_data:
+            for label in pr["labels"].split(","):
+                if label:
+                    label_dist[label] = label_dist.get(label, 0) + 1
+
+        # 评审者分布
+        reviewer_dist = {}
+        for pr in prs_data:
+            for assignee in pr["assignees"].split(","):
+                if assignee:
+                    reviewer_dist[assignee] = reviewer_dist.get(assignee, 0) + 1
+
+        # 每日趋势
+        daily_trend = {}
+        for pr in prs_data:
+            try:
+                created = datetime.fromisoformat(pr["created_at"].replace("Z", "+00:00"))
+                date_str = created.strftime("%Y-%m-%d")
+                if date_str not in daily_trend:
+                    daily_trend[date_str] = {"created": 0, "merged": 0, "closed": 0}
+                daily_trend[date_str]["created"] += 1
+            except:
+                pass
+
+        for pr in merged:
+            try:
+                merged_date = datetime.fromisoformat(pr["merged_at"].replace("Z", "+00:00"))
+                date_str = merged_date.strftime("%Y-%m-%d")
+                if date_str not in daily_trend:
+                    daily_trend[date_str] = {"created": 0, "merged": 0, "closed": 0}
+                daily_trend[date_str]["merged"] += 1
+            except:
+                pass
+
+        for pr in closed_not_merged:
+            try:
+                closed_date = datetime.fromisoformat(pr["closed_at"].replace("Z", "+00:00"))
+                date_str = closed_date.strftime("%Y-%m-%d")
+                if date_str not in daily_trend:
+                    daily_trend[date_str] = {"created": 0, "merged": 0, "closed": 0}
+                daily_trend[date_str]["closed"] += 1
+            except:
+                pass
+
+        # 24小时内评审率
+        timely_reviews = len([t for t in review_times if t <= 1440])
+        timely_review_rate = timely_reviews / len(review_times) * 100 if review_times else 0
+
+        # CI 成功率：根据标签计算
+        ci_success_count = label_dist.get("ci_successful", 0)
+        ci_failed_count = label_dist.get("ci_failed", 0)
+        ci_total = ci_success_count + ci_failed_count
+        ci_success_rate = round(ci_success_count / ci_total * 100, 2) if ci_total > 0 else 0
+
+        return {
+            "repo": f"{self.owner}/{self.repo}",
+            "analysis_period": f"近 {self.days} 天",
+            "analysis_time": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+            "summary": {
+                "total_prs": total,
+                "opened_prs": len(opened),
+                "merged_prs": len(merged),
+                "closed_prs": len(closed_not_merged),
+                "merge_rate": round(len(merged) / total * 100, 2) if total > 0 else 0,
+                "conflict_rate": round(len(conflicts) / total * 100, 2) if total > 0 else 0
+            },
+            "efficiency": {
+                "avg_first_review_time_minutes": round(avg_review_time, 2),
+                "avg_merge_duration_hours": round(avg_merge_duration / 60, 2),
+                "min_merge_duration_hours": round(min_merge_duration / 60, 2),
+                "max_merge_duration_hours": round(max_merge_duration / 60, 2),
+                "timely_review_rate": round(timely_review_rate, 2),
+                "review_time_samples": len(review_times),
+                "merge_duration_samples": len(merge_durations)
+            },
+            "quality": {
+                "avg_change_lines": round(avg_changes, 2),
+                "max_change_lines": max_changes,
+                "large_pr_count": len(large_prs),
+                "large_pr_rate": round(len(large_prs) / total * 100, 2) if total > 0 else 0,
+                "comment_density": round(comment_density, 4),
+                "ci_success_rate": ci_success_rate
+            },
+            "distribution": {
+                "by_creator": dict(sorted(creator_dist.items(), key=lambda x: x[1], reverse=True)[:10]),
+                "by_target_branch": dict(sorted(branch_dist.items(), key=lambda x: x[1], reverse=True)[:10]),
+                "by_label": dict(sorted(label_dist.items(), key=lambda x: x[1], reverse=True)[:10]),
+                "by_reviewer": dict(sorted(reviewer_dist.items(), key=lambda x: x[1], reverse=True)[:10])
+            },
+            "daily_trend": dict(sorted(daily_trend.items()))
+        }, prs_data
+
+    def generate_html_report(self, insights: Dict, prs_data: List[Dict], output_file: str):
+        """生成 HTML 报告（统计数据总结）"""
+        summary = insights["summary"]
+        efficiency = insights["efficiency"]
+        quality = insights["quality"]
+        distribution = insights["distribution"]
+        daily_trend = insights["daily_trend"]
+
+        # 准备图表数据
+        dates = list(daily_trend.keys())
+        created_counts = [daily_trend[d]["created"] for d in dates]
+        merged_counts = [daily_trend[d]["merged"] for d in dates]
+        closed_counts = [daily_trend[d]["closed"] for d in dates]
+
+        creator_names = list(distribution["by_creator"].keys())
+        creator_counts = list(distribution["by_creator"].values())
+
+        branch_names = list(distribution["by_target_branch"].keys())
+        branch_counts = list(distribution["by_target_branch"].values())
+
+        # 准备规模分布数据
+        size_ranges = {"0-50": 0, "51-200": 0, "201-500": 0, "501-1000": 0, ">1000": 0}
+        for pr in prs_data:
+            size = pr["total_changes"]
+            if size <= 50:
+                size_ranges["0-50"] += 1
+            elif size <= 200:
+                size_ranges["51-200"] += 1
+            elif size <= 500:
+                size_ranges["201-500"] += 1
+            elif size <= 1000:
+                size_ranges["501-1000"] += 1
+            else:
+                size_ranges[">1000"] += 1
+
+        html_content = f'''<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>PR 洞察报告 - {insights["repo"]}</title>
+    <script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
+    <style>
+        * {{ margin: 0; padding: 0; box-sizing: border-box; }}
+        body {{ font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif; background: #f5f7fa; color: #333; }}
+        .container {{ max-width: 1200px; margin: 0 auto; padding: 20px; }}
+        h1 {{ text-align: center; color: #1a365d; margin-bottom: 30px; padding-bottom: 15px; border-bottom: 2px solid #e2e8f0; }}
+        h2 {{ color: #1a365d; margin: 20px 0 15px 0; }}
+        .stats-grid {{ display: grid; grid-template-columns: repeat(auto-fit, minmax(140px, 1fr)); gap: 12px; margin-bottom: 25px; }}
+        .stat-card {{ background: white; border-radius: 8px; padding: 15px; box-shadow: 0 2px 10px rgba(0,0,0,0.1); text-align: center; }}
+        .stat-value {{ font-size: 26px; font-weight: bold; color: #1e40af; }}
+        .stat-label {{ color: #64748b; margin-top: 5px; font-size: 13px; }}
+        .charts-grid {{ display: grid; grid-template-columns: repeat(auto-fit, minmax(400px, 1fr)); gap: 20px; margin-bottom: 25px; }}
+        .chart-box {{ background: white; border-radius: 8px; padding: 20px; box-shadow: 0 2px 10px rgba(0,0,0,0.1); }}
+        .chart-title {{ font-size: 15px; font-weight: bold; color: #1a365d; margin-bottom: 12px; }}
+        .dist-section {{ background: white; border-radius: 8px; padding: 20px; box-shadow: 0 2px 10px rgba(0,0,0,0.1); margin-bottom: 20px; }}
+        .dist-item {{ display: flex; justify-content: space-between; padding: 8px 0; border-bottom: 1px solid #e2e8f0; }}
+        .dist-item:last-child {{ border-bottom: none; }}
+        .footer {{ text-align: center; color: #64748b; padding: 20px; font-size: 12px; }}
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>PR 洞察报告 - {insights["repo"]}</h1>
+
+        <h2>概览统计</h2>
+        <div class="stats-grid">
+            <div class="stat-card">
+                <div class="stat-value">{summary["total_prs"]}</div>
+                <div class="stat-label">总 PR 数</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{summary["opened_prs"]}</div>
+                <div class="stat-label">打开中</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{summary["merged_prs"]}</div>
+                <div class="stat-label">已合并</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{summary["closed_prs"]}</div>
+                <div class="stat-label">已关闭(未合并)</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{summary["merge_rate"]}%</div>
+                <div class="stat-label">合并率</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{summary["conflict_rate"]}%</div>
+                <div class="stat-label">冲突率</div>
+            </div>
+        </div>
+
+        <h2>效率指标</h2>
+        <div class="stats-grid">
+            <div class="stat-card">
+                <div class="stat-value">{efficiency["avg_first_review_time_minutes"]:.1f}</div>
+                <div class="stat-label">平均首次评审(分钟)</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{efficiency["avg_merge_duration_hours"]:.1f}</div>
+                <div class="stat-label">平均合并耗时(小时)</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{efficiency["min_merge_duration_hours"]:.1f}</div>
+                <div class="stat-label">最短合并耗时(小时)</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{efficiency["max_merge_duration_hours"]:.1f}</div>
+                <div class="stat-label">最长合并耗时(小时)</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{efficiency["timely_review_rate"]}%</div>
+                <div class="stat-label">24h评审率</div>
+            </div>
+        </div>
+
+        <h2>质量指标</h2>
+        <div class="stats-grid">
+            <div class="stat-card">
+                <div class="stat-value">{quality["avg_change_lines"]:.0f}</div>
+                <div class="stat-label">平均变更行数</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{quality["max_change_lines"]}</div>
+                <div class="stat-label">最大变更行数</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{quality["large_pr_count"]}</div>
+                <div class="stat-label">大PR数(>500行)</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{quality["large_pr_rate"]}%</div>
+                <div class="stat-label">大PR占比</div>
+            </div>
+            <div class="stat-card">
+                <div class="stat-value">{quality["comment_density"]:.4f}</div>
+                <div class="stat-label">评论密度</div>
+            </div>
+        </div>
+
+        <h2>趋势图表</h2>
+        <div class="charts-grid">
+            <div class="chart-box">
+                <div class="chart-title">每日 PR 趋势</div>
+                <canvas id="trendChart"></canvas>
+            </div>
+            <div class="chart-box">
+                <div class="chart-title">创建者分布 Top 10</div>
+                <canvas id="creatorChart"></canvas>
+            </div>
+        </div>
+
+        <div class="charts-grid">
+            <div class="chart-box">
+                <div class="chart-title">目标分支分布</div>
+                <canvas id="branchChart"></canvas>
+            </div>
+            <div class="chart-box">
+                <div class="chart-title">代码变更规模分布</div>
+                <canvas id="sizeChart"></canvas>
+            </div>
+        </div>
+
+        <h2>分布统计</h2>
+        <div class="charts-grid">
+            <div class="dist-section">
+                <div class="chart-title">创建者分布 Top 10</div>
+'''
+        for creator, count in distribution["by_creator"].items():
+            html_content += f'                <div class="dist-item"><span>{creator}</span><span>{count}</span></div>\n'
+
+        html_content += f'''            </div>
+            <div class="dist-section">
+                <div class="chart-title">目标分支分布</div>
+'''
+        for branch, count in distribution["by_target_branch"].items():
+            html_content += f'                <div class="dist-item"><span>{branch}</span><span>{count}</span></div>\n'
+
+        html_content += f'''            </div>
+        </div>
+
+        <h2>PR 列表</h2>
+        <div class="dist-section">
+            <table style="width: 100%; border-collapse: collapse; font-size: 13px;">
+                <thead>
+                    <tr style="background: #f1f5f9;">
+                        <th style="padding: 10px; text-align: left; border-bottom: 2px solid #e2e8f0;">PR</th>
+                        <th style="padding: 10px; text-align: left; border-bottom: 2px solid #e2e8f0;">标题</th>
+                        <th style="padding: 10px; text-align: center; border-bottom: 2px solid #e2e8f0;">状态</th>
+                        <th style="padding: 10px; text-align: left; border-bottom: 2px solid #e2e8f0;">创建者</th>
+                        <th style="padding: 10px; text-align: left; border-bottom: 2px solid #e2e8f0;">目标分支</th>
+                        <th style="padding: 10px; text-align: right; border-bottom: 2px solid #e2e8f0;">变更行数</th>
+                        <th style="padding: 10px; text-align: center; border-bottom: 2px solid #e2e8f0;">创建时间</th>
+                    </tr>
+                </thead>
+                <tbody>
+'''
+        # PR 列表按创建时间倒序排列（API 返回顺序）
+        for pr in prs_data:
+            state_display = {"open": "打开", "merged": "已合并", "closed": "已关闭"}.get(pr["state"], pr["state"])
+            state_color = {"open": "#3b82f6", "merged": "#10b981", "closed": "#ef4444"}.get(pr["state"], "#64748b")
+            html_content += f'''                    <tr>
+                        <td style="padding: 8px; border-bottom: 1px solid #e2e8f0;"><a href="{pr["html_url"]}" target="_blank">#{pr["pr_number"]}</a></td>
+                        <td style="padding: 8px; border-bottom: 1px solid #e2e8f0; max-width: 300px; overflow: hidden; text-overflow: ellipsis; white-space: nowrap;">{pr["title"][:50]}{"..." if len(pr["title"]) > 50 else ""}</td>
+                        <td style="padding: 8px; border-bottom: 1px solid #e2e8f0; text-align: center;"><span style="color: {state_color}; font-weight: 500;">{state_display}</span></td>
+                        <td style="padding: 8px; border-bottom: 1px solid #e2e8f0;">{pr["creator"]}</td>
+                        <td style="padding: 8px; border-bottom: 1px solid #e2e8f0;">{pr["target_branch"]}</td>
+                        <td style="padding: 8px; border-bottom: 1px solid #e2e8f0; text-align: right;">{pr["total_changes"]}</td>
+                        <td style="padding: 8px; border-bottom: 1px solid #e2e8f0; text-align: center;">{pr["created_at"][:10] if pr["created_at"] else "-"}</td>
+                    </tr>
+'''
+
+        html_content += f'''                </tbody>
+            </table>
+        </div>
+
+        <div class="footer">
+            <p>分析时间: {insights["analysis_time"]} | 分析周期: {insights["analysis_period"]}</p>
+        </div>
+    </div>
+
+    <script>
+    // 趋势图
+    new Chart(document.getElementById('trendChart'), {{
+        type: 'line',
+        data: {{
+            labels: {json.dumps(dates)},
+            datasets: [
+                {{
+                    label: '创建',
+                    data: {json.dumps(created_counts)},
+                    borderColor: '#3b82f6',
+                    backgroundColor: 'rgba(59, 130, 246, 0.1)',
+                    fill: true
+                }},
+                {{
+                    label: '合并',
+                    data: {json.dumps(merged_counts)},
+                    borderColor: '#10b981',
+                    backgroundColor: 'rgba(16, 185, 129, 0.1)',
+                    fill: true
+                }},
+                {{
+                    label: '关闭',
+                    data: {json.dumps(closed_counts)},
+                    borderColor: '#ef4444',
+                    backgroundColor: 'rgba(239, 68, 68, 0.1)',
+                    fill: true
+                }}
+            ]
+        }},
+        options: {{
+            responsive: true,
+            scales: {{ y: {{ beginAtZero: true }} }}
+        }}
+    }});
+
+    // 创建者分布图
+    new Chart(document.getElementById('creatorChart'), {{
+        type: 'bar',
+        data: {{
+            labels: {json.dumps(creator_names)},
+            datasets: [{{
+                label: 'PR 数量',
+                data: {json.dumps(creator_counts)},
+                backgroundColor: ['#3b82f6', '#10b981', '#f59e0b', '#ef4444', '#8b5cf6', '#ec4899', '#06b6d4', '#84cc16', '#f97316', '#6366f1']
+            }}]
+        }},
+        options: {{
+            responsive: true,
+            indexAxis: 'y',
+            scales: {{ x: {{ beginAtZero: true }} }}
+        }}
+    }});
+
+    // 目标分支分布图
+    new Chart(document.getElementById('branchChart'), {{
+        type: 'bar',
+        data: {{
+            labels: {json.dumps(branch_names)},
+            datasets: [{{
+                label: 'PR 数量',
+                data: {json.dumps(branch_counts)},
+                backgroundColor: '#8b5cf6'
+            }}]
+        }},
+        options: {{
+            responsive: true,
+            indexAxis: 'y',
+            scales: {{ x: {{ beginAtZero: true }} }}
+        }}
+    }});
+
+    // 规模分布图
+    new Chart(document.getElementById('sizeChart'), {{
+        type: 'bar',
+        data: {{
+            labels: {json.dumps(list(size_ranges.keys()))},
+            datasets: [{{
+                label: 'PR 数量',
+                data: {json.dumps(list(size_ranges.values()))},
+                backgroundColor: ['#10b981', '#3b82f6', '#f59e0b', '#ef4444', '#8b5cf6']
+            }}]
+        }},
+        options: {{
+            responsive: true,
+            scales: {{ y: {{ beginAtZero: true }} }}
+        }}
+    }});
+    </script>
+</body>
+</html>
+'''
+
+        with open(output_file, 'w', encoding='utf-8') as f:
+            f.write(html_content)
+
+        print(f"HTML 报告: {output_file}")
+
+    def generate_markdown_report(self, insights: Dict, prs_data: List[Dict], output_file: str):
+        """生成 Markdown 报告（统计数据总结）"""
+        summary = insights["summary"]
+        efficiency = insights["efficiency"]
+        quality = insights["quality"]
+        distribution = insights["distribution"]
+        daily_trend = insights["daily_trend"]
+
+        md_content = f'''# PR 洞察报告 - {insights["repo"]}
+
+> 分析时间: {insights["analysis_time"]} | 分析周期: {insights["analysis_period"]}
+
+## 概览统计
+
+| 指标 | 数值 |
+|------|------|
+| 总 PR 数 | {summary["total_prs"]} |
+| 打开中 | {summary["opened_prs"]} |
+| 已合并 | {summary["merged_prs"]} |
+| 已关闭(未合并) | {summary["closed_prs"]} |
+| 合并率 | {summary["merge_rate"]}% |
+| 冲突率 | {summary["conflict_rate"]}% |
+
+## 效率指标
+
+| 指标 | 数值 |
+|------|------|
+| 平均首次评审 | {efficiency["avg_first_review_time_minutes"]:.1f} 分钟 |
+| 平均合并耗时 | {efficiency["avg_merge_duration_hours"]:.1f} 小时 |
+| 最短合并耗时 | {efficiency["min_merge_duration_hours"]:.1f} 小时 |
+| 最长合并耗时 | {efficiency["max_merge_duration_hours"]:.1f} 小时 |
+| 24h评审率 | {efficiency["timely_review_rate"]}% |
+| 评审时间样本数 | {efficiency["review_time_samples"]} |
+| 合并耗时样本数 | {efficiency["merge_duration_samples"]} |
+
+## 质量指标
+
+| 指标 | 数值 |
+|------|------|
+| 平均变更行数 | {quality["avg_change_lines"]:.0f} |
+| 最大变更行数 | {quality["max_change_lines"]} |
+| 大PR数(>500行) | {quality["large_pr_count"]} |
+| 大PR占比 | {quality["large_pr_rate"]}% |
+| 评论密度 | {quality["comment_density"]:.4f} |
+
+## 每日趋势
+
+| 日期 | 创建 | 合并 | 关闭 |
+|------|------|------|------|
+'''
+        for date, counts in sorted(daily_trend.items()):
+            md_content += f"| {date} | {counts['created']} | {counts['merged']} | {counts['closed']} |\n"
+
+        md_content += '''
+## 创建者分布 Top 10
+
+| 创建者 | 数量 |
+|------|------|
+'''
+        for creator, count in distribution["by_creator"].items():
+            md_content += f"| {creator} | {count} |\n"
+
+        md_content += '''
+## 目标分支分布
+
+| 分支 | 数量 |
+|------|------|
+'''
+        for branch, count in distribution["by_target_branch"].items():
+            md_content += f"| {branch} | {count} |\n"
+
+        md_content += '''
+## 标签分布 Top 10
+
+| 标签 | 数量 |
+|------|------|
+'''
+        for label, count in distribution["by_label"].items():
+            md_content += f"| {label} | {count} |\n"
+
+        md_content += '''
+## PR 列表
+
+| PR | 标题 | 状态 | 创建者 | 目标分支 | 变更行数 | 创建时间 |
+|------|------|------|------|------|------|------|
+'''
+        # PR 列表按创建时间倒序排列
+        for pr in prs_data:
+            state_display = {"open": "打开", "merged": "已合并", "closed": "已关闭"}.get(pr["state"], pr["state"])
+            title_short = pr["title"][:40] + "..." if len(pr["title"]) > 40 else pr["title"]
+            md_content += f"| [#{pr['pr_number']}]({pr['html_url']}) | {title_short} | {state_display} | {pr['creator']} | {pr['target_branch']} | {pr['total_changes']} | {pr['created_at'][:10] if pr['created_at'] else '-'} |\n"
+
+        with open(output_file, 'w', encoding='utf-8') as f:
+            f.write(md_content)
+
+        print(f"Markdown 报告: {output_file}")
+
+    def run(self) -> Dict:
+        """执行完整的分析流程"""
+        os.makedirs(self.output_dir, exist_ok=True)
+
+        print(f"\n{'='*60}")
+        print(f"PR 洞察分析: {self.owner}/{self.repo}")
+        print(f"分析周期: 近 {self.days} 天")
+        print(f"{'='*60}\n")
+
+        # 获取 PR 列表
+        prs = self.get_prs()
+
+        if not prs:
+            print("未获取到任何 PR 数据")
+            return {}
+
+        # 分析每个 PR
+        print(f"\n分析 PR 详情...")
+        prs_data = []
+        for i, pr in enumerate(prs, 1):
+            print(f"  处理 {i}/{len(prs)}: PR #{pr.get('number')}")
+            analyzed = self.analyze_pr(pr)
+            prs_data.append(analyzed)
+            time.sleep(0.6)  # API 限流控制
+
+        # 计算洞察指标
+        print(f"\n计算洞察指标...")
+        insights, raw_data = self.calculate_insights(prs_data)
+
+        # 保存 JSON（统计数据 + 原始数据）
+        json_file = os.path.join(self.output_dir, f"pr_insight_{self.repo}_{self.days}d.json")
+        full_data = {
+            "statistics": insights,
+            "raw_data": raw_data
+        }
+        with open(json_file, 'w', encoding='utf-8') as f:
+            json.dump(full_data, f, ensure_ascii=False, indent=2)
+
+        # 生成 HTML 报告
+        html_file = os.path.join(self.output_dir, f"pr_insight_{self.repo}_{self.days}d.html")
+        self.generate_html_report(insights, prs_data, html_file)
+
+        # 生成 Markdown 报告
+        md_file = os.path.join(self.output_dir, f"pr_insight_{self.repo}_{self.days}d.md")
+        self.generate_markdown_report(insights, prs_data, md_file)
+
+        # 打印摘要
+        print(f"\n{'='*60}")
+        print(f"分析完成!")
+        print(f"{'='*60}")
+        print(f"总 PR 数: {insights['summary']['total_prs']}")
+        print(f"打开中: {insights['summary']['opened_prs']}")
+        print(f"已合并: {insights['summary']['merged_prs']}")
+        print(f"合并率: {insights['summary']['merge_rate']}%")
+        print(f"平均首次评审: {insights['efficiency']['avg_first_review_time_minutes']:.1f} 分钟")
+        print(f"平均合并耗时: {insights['efficiency']['avg_merge_duration_hours']:.1f} 小时")
+        print(f"最短合并耗时: {insights['efficiency']['min_merge_duration_hours']:.1f} 小时")
+        print(f"最长合并耗时: {insights['efficiency']['max_merge_duration_hours']:.1f} 小时")
+        print(f"平均变更行数: {insights['quality']['avg_change_lines']:.0f}")
+        print(f"最大变更行数: {insights['quality']['max_change_lines']}")
+        print(f"\n输出文件:")
+        print(f"- JSON 数据: {json_file}")
+        print(f"- HTML 报告: {html_file}")
+        print(f"- Markdown 报告: {md_file}")
+        print(f"{'='*60}\n")
+
+        return full_data