#!/usr/bin/env python3 """从BOSS直聘本地DB提取简历为JSONL 用法: python3 extract_resumes.py --job "戏文" --output data/xiwen_resumes.jsonl python3 extract_resumes.py --job "全栈" --output data/fullstack_resumes.jsonl python3 extract_resumes.py --account 5889492 --job "戏文" --output data/xiwen_resumes.jsonl """ import sqlite3, javaobj, json, argparse, sys, os SCRIPT_DIR = os.path.dirname(os.path.abspath(__file__)) DATA_DIR = os.path.join('/sdcard/bosszhipin', 'data') DB_PATH = os.path.join(DATA_DIR, 'hpbr_bosszhipin_client.db') def flatten(x, depth=0): if depth > 10: return str(x)[:200] if isinstance(x, (str, int, float, bool)) or x is None: return x if isinstance(x, (list, tuple)): return [flatten(v, depth+1) for v in x] if isinstance(x, dict): return {k: flatten(v, depth+1) for k, v in x.items()} if hasattr(x, '__dict__'): return {k: flatten(v, depth+1) for k, v in vars(x).items() if k not in ('classdesc', 'annotations')} return str(x)[:200] def main(): parser = argparse.ArgumentParser(description='提取BOSS直聘简历') parser.add_argument('--job', default=None, help='岗位关键词(模糊匹配 bossJobPosition)') parser.add_argument('--job-id', default=None, help='精确按 jobId 提取(优先于 --job)') parser.add_argument('--account', default=None, help='指定账号 myId(不指定则查所有账号)') parser.add_argument('--output', required=True, help='输出 JSONL 文件路径') args = parser.parse_args() if not args.job and not args.job_id: print('错误: --job 或 --job-id 至少传一个', file=sys.stderr) sys.exit(1) if not os.path.exists(DB_PATH): print(f'错误: DB不存在 {DB_PATH}', file=sys.stderr) print('请先运行: bash pull_dbs_local.sh', file=sys.stderr) sys.exit(1) con = sqlite3.connect(DB_PATH) con.row_factory = sqlite3.Row # 构建查询 if args.job_id: # 精确 jobId 匹配 query = ''' SELECT m.fromUid, c.friendName, c.bossJobPosition, c.jobId, m.resume FROM MessageTable m JOIN Contact c ON c.friendId = m.fromUid WHERE c.jobId = ? AND m.resume IS NOT NULL ORDER BY c.bossJobPosition, c.friendName ''' params = [args.job_id] if args.account: query = ''' SELECT m.fromUid, c.friendName, c.bossJobPosition, c.jobId, m.resume FROM MessageTable m JOIN Contact c ON c.friendId = m.fromUid WHERE c.myId = ? AND c.jobId = ? AND m.resume IS NOT NULL ORDER BY c.bossJobPosition, c.friendName ''' params = [args.account, args.job_id] rows = con.execute(query, params).fetchall() elif args.account: query = ''' SELECT m.fromUid, c.friendName, c.bossJobPosition, c.jobId, m.resume FROM MessageTable m JOIN Contact c ON c.friendId = m.fromUid WHERE c.myId = ? AND c.bossJobPosition LIKE ? AND m.resume IS NOT NULL ORDER BY c.bossJobPosition, c.friendName ''' rows = con.execute(query, (args.account, f'%{args.job}%')).fetchall() else: query = ''' SELECT m.fromUid, c.friendName, c.bossJobPosition, c.jobId, m.resume FROM MessageTable m JOIN Contact c ON c.friendId = m.fromUid WHERE c.bossJobPosition LIKE ? AND m.resume IS NOT NULL ORDER BY c.bossJobPosition, c.friendName ''' rows = con.execute(query, (f'%{args.job}%',)).fetchall() os.makedirs(os.path.dirname(args.output) or '.', exist_ok=True) with open(args.output, 'w', encoding='utf-8') as f: for row in rows: obj = javaobj.loads(row['resume']) resume = flatten(obj) if not isinstance(resume, dict): continue record = { 'friendId': row['fromUid'], 'name': row['friendName'] or resume.get('userInfo', {}).get('name', ''), 'jobPosition': row['bossJobPosition'], 'jobId': row['jobId'], 'age': resume.get('age', ''), 'education': resume.get('education', ''), 'applyStatus': resume.get('applyStatus', ''), 'city': resume.get('city', ''), 'expSalary': resume.get('expSalary', ''), 'workAge': resume.get('workAge', ''), 'firstText': resume.get('firstText', ''), 'secondText': resume.get('secondText', ''), 'positionName': resume.get('positionName', ''), 'jobSalary': resume.get('jobSalary', ''), 'experiences': [], } for exp in resume.get('experiences', []): if isinstance(exp, dict): record['experiences'].append({ 'type': exp.get('type', 0), 'startDate': exp.get('startDate', ''), 'endDate': exp.get('endDate', ''), 'organization': exp.get('organization', ''), 'occupation': exp.get('occupation', ''), }) f.write(json.dumps(record, ensure_ascii=False) + '\n') print(f'导出 {len(rows)} 份简历到 {args.output}') # 统计 from collections import Counter edu = Counter(r['education'] for r in [json.loads(l) for l in open(args.output, encoding='utf-8')]) print(f'学历分布: {edu.most_common()}') con.close() if __name__ == '__main__': main()