pdf_code/zzb_data_prod/Mil_unit.py

87 lines
3.3 KiB
Python

from pymilvus import connections, FieldSchema, CollectionSchema, DataType, Collection,MilvusClient
from config import MILVUS_CLIENT
import time
from datetime import datetime, timedelta
def create_partition_by_hour():
# 连接到 Milvus 服务器
connections.connect("default",uri=MILVUS_CLIENT)
# 获取集合
collection_name = "pdf_measure_v4"
collection = Collection(collection_name)
# 获取当前时间
now = datetime.now()
current_hour = now.strftime("%Y%m%d%H")
# 创建当前小时的分区
partition_name = f"partition_{current_hour}"
if not collection.has_partition(partition_name):
collection.create_partition(partition_name)
print(f"Created partition: {partition_name}")
# 删除前一个小时的分区
previous_hour = (now - timedelta(hours=1)).strftime("%Y%m%d%H")
previous_partition_name = f"partition_{previous_hour}"
if collection.has_partition(previous_partition_name):
pre_partition = collection.partition(previous_partition_name)
pre_partition.release()
collection.drop_partition(previous_partition_name)
print(f"Dropped partition: {previous_partition_name}")
partition = collection.partition(partition_name)
partition.load()
return collection, partition
# res = partition.search(
# # collection_name="pdf_measure_v4", # Replace with the actual name of your collection
# # Replace with your query vector
# data=data,
# limit=3, # Max. number of search results to return
# anns_field="vector",
# param={"metric_type": "COSINE", "params": {}}, # Search parameters
# output_fields=["measure_name","measure_value","table_num","table_index","measure_unit"],
# # filter=filter_str,
# expr=query
# )
# from pymilvus import connections, CollectionSchema, Collection,utility,FieldSchema,DataType
# # 连接到 B 服务器上的 Milvus
# # connections.connect(host='124.70.129.232', port='19530')# 测试服务器
# connections.connect(host='1.94.60.103', port='19530')# 测试服务器
# # # 获取集合列表
# utility.drop_collection("pdf_measure_v4")
#
# # 定义字段
# fields = [
# FieldSchema(name="id", dtype=DataType.INT64, is_primary=True, auto_id=True),
# FieldSchema(name="vector", dtype=DataType.FLOAT_VECTOR, dim=1536),
# FieldSchema(name="table_num", dtype=DataType.INT16),
# FieldSchema(name="table_index", dtype=DataType.INT16),
# FieldSchema(name="measure_name", dtype=DataType.VARCHAR, max_length=200),
# FieldSchema(name="measure_value", dtype=DataType.VARCHAR, max_length=200),
# FieldSchema(name="file_id", dtype=DataType.VARCHAR, max_length=200),
# FieldSchema(name="measure_unit", dtype=DataType.VARCHAR, max_length=200)
# ]
#
# # 定义集合的 schema
# schema = CollectionSchema(fields=fields, description="My Milvus collection")
#
# # 创建集合
# collection = Collection(name="pdf_measure_v4", schema=schema)
#
# collection = Collection("pdf_measure_v4")
# index_params = {
# "index_type": "IVF_FLAT",
# "metric_type": "COSINE",
# "params": {"nlist": 128}
# }
# collection.create_index(field_name="vector", index_params=index_params)
# collection.load()