#!/usr/bin/env python3 """ 生成预算关键字分析HTML报告 读取analysis_results.json，按市州→区县→组织部/党校分层 关键字合并：同一区县同一单位下相同关键字合并为一行，金额累加，位置全列出 """ import json import os from collections import defaultdict, OrderedDict BASE_DIR = "/data/www/files/2026-yjs-cyl" JSON_PATH = os.path.join(BASE_DIR, "analysis_results.json") OUTPUT_HTML = os.path.join(BASE_DIR, "2026年 曾耀龙 负责市场区域预决算分析报告.html") # 关键字颜色映射 KW_COLORS = { "网络培训": "#FF6B6B", "培训": "#4ECDC4", "能力提升": "#45B7D1", "视频拍摄": "#FFA07A", "党员教育片": "#98D8C8", "课程制作": "#F7DC6F", "直播": "#BB8FCE", "视频": "#85C1E9", "课件": "#F8B88B", "系统": "#82E0AA", "信息化": "#F1948A", "网络": "#AED6F1", "数字": "#D7BDE2", "设备": "#A3E4D7", "平台": "#F9E79F", "维护": "#D5A6BD", "软件": "#A9CCE3", } def load_data(): with open(JSON_PATH, 'r', encoding='utf-8') as f: return json.load(f) def merge_keywords(data): """ 按市州→区县→单位类型分层 同一区县同一单位下相同关键字合并为一行，金额累加，位置全列出 """ # 构建分层结构 cities = OrderedDict() for key, val in data.items(): city = val['city'] district = val['district'] if val['district'] else "未分类" unit_type = val['unit_type'] if val['unit_type'] else "未分类" if city not in cities: cities[city] = OrderedDict() if district not in cities[city]: cities[city][district] = OrderedDict() if unit_type not in cities[city][district]: cities[city][district][unit_type] = { 'files': [], 'keywords_merged': OrderedDict() } unit = cities[city][district][unit_type] for file_info in val['files']: file_entry = { 'filename': file_info['filename'], 'rel_path': file_info['rel_path'], 'match_count': len(file_info['matches']) } unit['files'].append(file_entry) # 合并关键字 for m in file_info['matches']: kw = m['keyword'] if kw not in unit['keywords_merged']: unit['keywords_merged'][kw] = { 'keyword': kw, 'amount': 0.0, 'locations': [], 'contents': [], 'count': 0, 'files': set() } entry = unit['keywords_merged'][kw] entry['amount'] += m['amount'] entry['count'] += 1 entry['locations'].append(m['location']) entry['contents'].append(m['content']) entry['files'].add(file_info['filename']) return cities def calc_totals(cities): """计算汇总数据""" total_files = 0 total_matches = 0 total_amount = 0.0 total_cities = len(cities) total_districts = 0 total_units = 0 kw_dist = defaultdict(lambda: {'count': 0, 'amount': 0.0}) for city, districts in cities.items(): total_districts += len(districts) for district, units in districts.items(): for unit_type, unit_data in units.items(): total_files += len(unit_data['files']) for kw, kw_data in unit_data['keywords_merged'].items(): total_matches += kw_data['count'] total_amount += kw_data['amount'] kw_dist[kw]['count'] += kw_data['count'] kw_dist[kw]['amount'] += kw_data['amount'] total_units += 1 return { 'total_files': total_files, 'total_matches': total_matches, 'total_amount': total_amount, 'total_cities': total_cities, 'total_districts': total_districts, 'total_units': total_units, 'kw_dist': dict(sorted(kw_dist.items(), key=lambda x: x[1]['count'], reverse=True)) } def generate_html(cities, totals): """生成完整HTML报告""" # 关键字分布条形图数据 kw_dist = totals['kw_dist'] max_kw_count = max([v['count'] for v in kw_dist.values()]) if kw_dist else 1 kw_bars = "" for kw, data in kw_dist.items(): pct = (data['count'] / max_kw_count * 100) if max_kw_count > 0 else 0 color = KW_COLORS.get(kw, "#888") kw_bars += f'''
{kw}
{data['count']}次
{data['amount']:.2f}万元
''' # 市州导航卡片 city_cards = "" city_sections = "" for city_idx, (city, districts) in enumerate(cities.items()): city_id = f"city_{city_idx}" city_amount = 0 city_matches = 0 city_files = 0 for district, units in districts.items(): for unit_type, unit_data in units.items(): city_files += len(unit_data['files']) for kw, kw_data in unit_data['keywords_merged'].items(): city_matches += kw_data['count'] city_amount += kw_data['amount'] city_cards += f'''
{city}
{len(districts)}区县 {city_files}文件 {city_amount:.2f}万元
''' # 市州详情区块 district_blocks = "" for dist_idx, (district, units) in enumerate(districts.items()): dist_id = f"{city_id}_dist_{dist_idx}" # 组织部和党校分开展示 unit_blocks = "" for unit_type, unit_data in units.items(): unit_id = f"{dist_id}_unit_{unit_type}" # 汇总金额 unit_amount = sum(kw_data['amount'] for kw_data in unit_data['keywords_merged'].values()) unit_match_count = sum(kw_data['count'] for kw_data in unit_data['keywords_merged'].values()) icon = "📋" if "组织部" in unit_type else ("🎓" if "党校" in unit_type else "📄") # 文件链接 file_links = "" for fi in unit_data['files']: safe_path = fi['rel_path'].replace("\\", "/") file_links += f'📂{fi["filename"]}\n' # 关键字表格行 table_rows = "" for kw, kw_data in unit_data['keywords_merged'].items(): color = KW_COLORS.get(kw, "#888") # 位置合并显示（最多显示10个，超出显示...） locs = kw_data['locations'] if len(locs) > 10: loc_str = ', '.join(locs[:10]) + f' ...等{len(locs)}处' else: loc_str = ', '.join(locs) # 内容合并显示 contents = kw_data['contents'] unique_contents = list(dict.fromkeys(contents)) # 去重保持顺序 if len(unique_contents) > 3: content_str = '
'.join(unique_contents[:3]) + f'
...等{len(unique_contents)}条内容' else: content_str = '
'.join(unique_contents) # 文件列表 kw_files = sorted(kw_data['files']) file_list = ', '.join(kw_files) table_rows += f''' {kw} {content_str} {loc_str} {kw_data["amount"]:.2f} {file_list} ''' if not table_rows: table_rows = '无匹配数据' unit_blocks += f'''
{icon} {unit_type} 汇总：{unit_amount:.2f}万元 | {unit_match_count}条匹配
{file_links}
{table_rows}
关键字 匹配内容 文件位置 预算金额(万元) 来源文件
''' district_blocks += f'''
▶ {district}
{unit_blocks}
''' city_sections += f'''
▶ {city} {len(districts)}个区县 | {city_files}个文件 | {city_matches}条匹配 | {city_amount:.2f}万元
{district_blocks}
''' html = f''' 2026年 曾耀龙 负责市场区域预决算分析报告

2026年 曾耀龙 负责市场区域预决算分析报告

预算文件关键字分析 | 17项关键字全量匹配
生成时间：2026-07-22 | 分析范围：内江市、自贡市、成都市东部新区
{totals['total_files']}
预算文件数
{totals['total_matches']}
总匹配项
{totals['total_amount']:.2f}万元
预算总金额
{totals['total_cities']}
市州数
{totals['total_districts']}
区县数
{len(kw_dist)}
命中关键字数

关键字分布

{kw_bars}
{city_cards}
{city_sections}
2026年 曾耀龙 负责市场区域预决算分析报告 | 数据来源：政府预算公开文件 | 17项关键字全量匹配分析
''' return html def main(): data = load_data() cities = merge_keywords(data) totals = calc_totals(cities) html = generate_html(cities, totals) with open(OUTPUT_HTML, 'w', encoding='utf-8') as f: f.write(html) print(f"报告已生成: {OUTPUT_HTML}") print(f"文件大小: {os.path.getsize(OUTPUT_HTML) / 1024:.1f} KB") # 打印汇总 print(f"\n=== 汇总 ===") print(f"文件数: {totals['total_files']}") print(f"总匹配: {totals['total_matches']}") print(f"总金额: {totals['total_amount']:.2f}万元") print(f"市州数: {totals['total_cities']}") print(f"区县数: {totals['total_districts']}") print(f"单位数: {totals['total_units']}") print(f"\n关键字分布:") for kw, d in totals['kw_dist'].items(): print(f" {kw}: {d['count']}次, {d['amount']:.2f}万元") if __name__ == '__main__': main()