#!/usr/bin/env python3 """ 二期报告生成脚本框架 数据 → PCA赋分(SPSS对齐) → 统计 → [LLM并行生成] → HTML报告 ⚠️ 默认 --no-llm 模式，不会调用 LLM（避免产生费用） 若需启用 LLM 生成，手动传 --enable-llm 参数 """ import sys import time import argparse import json import logging from pathlib import Path import numpy as np sys.path.insert(0, str(Path(__file__).parent)) from data_engine_era2 import DataEngineEra2 from engines.pca_scoring_engine_era2 import PcaScoringEngineEra2 from engines.stats_engine_era2 import StatsEngineEra2 as StatsEngine from config_era2 import DIMENSION_FRAMEWORK logging.basicConfig( level=logging.INFO, format="%(asctime)s [%(levelname)s] %(message)s", datefmt="%H:%M:%S", ) logger = logging.getLogger(__name__) def clean_for_json(obj): if isinstance(obj, dict): return {k: clean_for_json(v) for k, v in obj.items()} elif isinstance(obj, list): return [clean_for_json(v) for v in obj] elif isinstance(obj, (np.integer,)): return int(obj) elif isinstance(obj, (np.floating,)): return round(float(obj), 4) elif isinstance(obj, np.ndarray): return obj.tolist() elif isinstance(obj, float): return round(obj, 4) return obj def progress_callback(completed, total, segment_id): pct = completed / total * 100 bar = "█" * int(pct / 5) + "░" * (20 - int(pct / 5)) print(f"\r [{bar}] {pct:.0f}% ({completed}/{total}) {segment_id:<40}", end="", flush=True) def main(): parser = argparse.ArgumentParser(description="二期报告生成") parser.add_argument("--school", type=str, required=True, help="学校名称（简称或全称均可）") parser.add_argument("--district", type=str, default="长宁区", help="区域筛选，'all'表示全部，默认'长宁区'") parser.add_argument("--enable-llm", action="store_true", help="启用LLM生成（会产生API调用费用！）") parser.add_argument("--no-cache", action="store_true", help="不使用LLM缓存") parser.add_argument("--no-scoring-cache", action="store_true", help="不使用赋分缓存（强制重新计算全市赋分）") parser.add_argument("--list-schools", action="store_true", help="列出所有可用学校") parser.add_argument("--enable-agent", action="store_true", help="在报告中嵌入AI对话助手（右下角浮动按钮）") parser.add_argument("--lang", type=str, default="zh", choices=["zh", "en", "both"], help="报告语言：zh（中文，默认）/ en（英文）/ both（同时生成两份）") args = parser.parse_args() start_time = time.time() print("=" * 70) print("📊 二期课程实施监测报告生成系统") if not args.enable_llm: print("⚠️ LLM 已禁用（--no-llm 模式），仅生成数据+图表") if args.enable_agent: print("🤖 AI 对话助手已启用") print("=" * 70) # ===== Step 1: 加载数据 ===== district = None if args.district == "all" else args.district logger.info(f"[1/5] 加载二期数据... (区域: {args.district}, 全市基准)") # 加载全市数据，按区筛选报告范围 data_engine = DataEngineEra2(district_filter=district) data_engine.load_all() district_schools = data_engine.schools # 本区学校列表 # 全市数据引擎（用于赋分全市学校） if data_engine.use_city_data and district: city_engine = DataEngineEra2(district_filter=None) # 不筛选区 city_engine.load_all() all_schools_for_scoring = city_engine logger.info(f" 全市基准: {len(city_engine.schools)}校, 本区: {len(district_schools)}校") else: city_engine = None all_schools_for_scoring = data_engine if args.list_schools: print(f"\n可用学校 ({len(district_schools)}所):") for i, s in enumerate(district_schools, 1): print(f" {i:3d}. {s}") return school = args.school if school not in district_schools: # 尝试从映射查找 from config_era2 import SCHOOL_NAME_SHORT_TO_FULL, SCHOOL_NAME_FULL_TO_SHORT if school in SCHOOL_NAME_SHORT_TO_FULL: pass elif school in SCHOOL_NAME_FULL_TO_SHORT: school = SCHOOL_NAME_FULL_TO_SHORT[school] if school not in all_schools_for_scoring.schools: print(f"\n❌ 学校 '{args.school}' 不在数据中。本区可用学校:") for s in district_schools: print(f" - {s}") return print(f"\n🏫 目标学校: {school}") print(f"📍 数据范围: {args.district} ({len(district_schools)}所学校)") if city_engine: print(f"📊 标准化基准: 全市{len(city_engine.schools)}所学校") # ===== Step 2: PCA赋分（SPSS对齐，全市所有学校） ===== logger.info(f"[2/5] PCA赋分计算 ({len(all_schools_for_scoring.schools)}校, SPSS对齐)...") pca_engine = PcaScoringEngineEra2(all_schools_for_scoring) pca_sub_scores = pca_engine.compute_all() # ===== Step 3: 统计分析（全市基准标准化） ===== logger.info("[3/5] 全市基准标准化 + 统计分析...") stats_engine = StatsEngine() sub_scores = stats_engine.compute_dimension_scores_pca(pca_sub_scores) dim_scores = stats_engine.compute_dimension_aggregates(sub_scores) report_data = stats_engine.compute_school_report_data( school, sub_scores, dim_scores, district_schools=district_schools, ) # 注入区域信息（模板需要） report_data["district"] = args.district report_data["total_schools_in_district"] = len(district_schools) # 决定要生成的语言列表 langs = ["zh", "en"] if args.lang == "both" else [args.lang] # ===== Step 4: LLM生成（按语言分别生成；缓存按 lang 隔离） ===== llm_sections_by_lang = {} if args.enable_llm: sys.path.insert(0, str(Path(__file__).parent.parent.parent / "backend")) from app.engines.llm_engine import LLMEngine llm_engine = LLMEngine() llm_engine.set_progress_callback(progress_callback) for lg in langs: logger.info(f"[4/5] LLM并行生成报告文字 (lang={lg}) ...") llm_sections_by_lang[lg] = llm_engine.generate_report_segments( report_data, use_cache=not args.no_cache, lang=lg, ) print() else: logger.info("[4/5] 跳过LLM生成（--no-llm 模式）") for lg in langs: llm_sections_by_lang[lg] = {} # ===== Step 5: 渲染HTML ===== from engines.report_renderer_era2 import ReportRendererEra2 as ReportRenderer renderer = ReportRenderer() output_dir = Path(__file__).parent.parent.parent / "output" / "era2" / args.district output_dir.mkdir(parents=True, exist_ok=True) output_paths = {} for lg in langs: logger.info(f"[5/5] 渲染HTML报告 (lang={lg}) ...") suffix = "_report_en.html" if lg == "en" else "_报告.html" output_path = output_dir / f"{school}{suffix}" renderer.render_to_file( report_data, llm_sections_by_lang[lg], output_path, enable_agent=args.enable_agent, lang=lg, ) output_paths[lg] = output_path # 保存报告数据JSON（一份，与语言无关） json_path = output_dir / f"{school}_report_data.json" with open(json_path, "w", encoding="utf-8") as f: json.dump(clean_for_json(report_data), f, ensure_ascii=False, indent=2) elapsed = time.time() - start_time print(f"\n{'=' * 70}") print(f"✅ 报告生成完成!") print(f" 学校: {school}") print(f" 区域: {args.district} ({len(data_engine.schools)}所学校)") print(f" 总体得分: {report_data['overall']['score']}分 (第{report_data['overall']['rank_in_district']}名)") for lg in langs: n_sec = len(llm_sections_by_lang.get(lg, {})) print(f" LLM段落({lg}): {n_sec}个 {'(已禁用)' if not args.enable_llm else ''}") print(f" AI助手: {'✅ 已嵌入' if args.enable_agent else '❌ 未启用'}") print(f" 耗时: {elapsed:.1f}秒") for lg, p in output_paths.items(): print(f" HTML({lg}): {p}") print(f" JSON: {json_path}") print(f"{'=' * 70}") if __name__ == "__main__": main()