Spaces:
Sleeping
Sleeping
import json | |
import logging | |
import time | |
from pathlib import Path | |
from elasticsearch import Elasticsearch | |
from tqdm import tqdm | |
def create_index_elastic_segmentation( | |
path: str, | |
logger: logging.Logger | None = None, | |
): | |
if logger is None: | |
logger = logging.getLogger(__name__) | |
# Подключение к Elasticsearch | |
es = Elasticsearch(hosts='localhost:9200') | |
INDEX_NAME = 'segmentation_search_elastic' | |
# Удаление старого индекса, если он существует | |
if es.indices.exists(index=INDEX_NAME): | |
es.indices.delete(index=INDEX_NAME) | |
mapping = { | |
"mappings": { | |
"properties": { | |
"segmentation_model": {"type": "text", "analyzer": "standard"}, | |
"segmentation_model2": {"type": "text", "analyzer": "standard"}, | |
"company_name": {"type": "text", "analyzer": "standard"}, | |
} | |
} | |
} | |
# Создание индекса с указанным маппингом | |
es.indices.create(index=INDEX_NAME, body=mapping) | |
for ind, path in tqdm(enumerate(Path(path).iterdir())): | |
# Открываем файл и читаем его содержимое | |
with open(path, 'r', encoding='utf-8') as file: | |
data = json.load(file) | |
# Индексирование документа в Elasticsearch | |
es.index(index=INDEX_NAME, id=ind + 1, body=data) | |
# Подсчет количества документов в индексе | |
count_response = es.count(index=INDEX_NAME) | |
logger.info( | |
f"{ind}, Total documents in '{INDEX_NAME}': {count_response['count']}" | |
) | |
time.sleep(1.0) | |
if es.indices.exists(index=INDEX_NAME): | |
logger.info(f"Index '{INDEX_NAME}' exists.") | |
# Подсчет количества документов в индексе | |
count_response = es.count(index=INDEX_NAME) | |
logger.info(f"Total documents in '{INDEX_NAME}': {count_response['count']}") | |
query = "К какой модели сегментации относится ООО ГРК Быстринское?" | |
query_ = { | |
"query": { | |
"bool": { | |
"should": [ | |
{ | |
"multi_match": { | |
"query": f"{query}", | |
"fields": [ | |
"segmentation_model", | |
"segmentation_model2", | |
"company_name", | |
], | |
"fuzziness": "AUTO", | |
"analyzer": "standard", | |
} | |
}, | |
{ | |
"multi_match": { | |
"query": "модели сегментации модель сегментации", | |
"fields": ["segmentation_model", "segmentation_model2"], | |
"operator": "or", | |
"boost": 0.1, | |
} | |
}, | |
] | |
} | |
} | |
} | |
# Выполнение поиска в Elasticsearch | |
response = es.search(index=INDEX_NAME, body=query_, size=1) | |
logger.info(f"Number of hits: {response['hits']['total']['value']}") | |
# Вывод результата поиска | |
for hit in response['hits']['hits']: | |
logger.info(hit['_source']) | |
if __name__ == '__main__': | |
path = '/mnt/ntr_work/project/nmd800/data/segmentation_card' | |
create_index_elastic_segmentation(path) | |