| 123456789101112131415161718192021222324252627 |
- from elasticsearch import Elasticsearch
- from typing import List, Dict
- import json
- from utils.es import ElasticsearchOperations
- def add_law_document(ops):
- json_data = None
- with open(r"D:\work\03\regulations.json","r",encoding="utf-8") as f:
- lines = f.readlines()
- json_data = json.loads(''.join(lines))
-
- print(">>> finished process document ")
-
- if json_data:
-
- index = 1
- total = len(json_data)
- for item in json_data:
- es_ops.index_document(item["article_text"],item["meta_data"]['ArticleTitle'])
- print(item["meta_data"]['ArticleTitle'],f"processed {index}/{total}")
- index = index + 1
- # 使用示例
- if __name__ == "__main__":
- es_ops = ElasticsearchOperations()
- add_law_document(es_ops)
-
|
