insert data error

[xxxrain]

0 Milvus version 2.5.11
1 create table python script
schema = client.create_schema(enable_dynamic_field=True, description="")
schema.add_field(field_name="kg_id", datatype=DataType.VARCHAR, description="The Primary Key", max_length=256, is_primary=True, auto_id=False)
schema.add_field(field_name="embedding", datatype=DataType.FLOAT_VECTOR, dim=768)
schema.add_field(field_name="node_id", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="data_id", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="name", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="ytenant_id", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="filter_p_key", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="table_type", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="describe", datatype=DataType.VARCHAR, max_length=768, nullable=True, default_value='')
schema.add_field(field_name="created", datatype=DataType.INT64)
schema.add_field(field_name="updated", datatype=DataType.INT64)

index_params = client.prepare_index_params()
index_params.add_index(field_name="embedding", metric_type="COSINE", index_type="AUTOINDEX")

client.create_collection(collection_name="filter_value", schema=schema, index_params=index_params)
2 table state desc
{'collection_name': 'filter_value', 'auto_id': False, 'num_shards': 1, 'description': '', 'fields': [{'field_id': 100, 'name': 'kg_id', 'description': 'The Primary Key', 'type': <DataType.VARCHAR: 21>, 'params': {'max_length': 256}, 'is_primary': True}, {'field_id': 101, 'name': 'embedding', 'description': '', 'type': <DataType.FLOAT_VECTOR: 101>, 'params': {'dim': 768}}, {'field_id': 102, 'name': 'node_id', 'description': '', 'type': <DataType.VARCHAR: 21>, 'params': {'max_length': 256}}, {'field_id': 103, 'name': 'data_id', 'description': '', 'type': <DataType.VARCHAR: 21>, 'params': {'max_length': 256}}, {'field_id': 104, 'name': 'name', 'description': '', 'type': <DataType.VARCHAR: 21>, 'params': {'max_length': 256}}, {'field_id': 105, 'name': 'ytenant_id', 'description': '', 'type': <DataType.VARCHAR: 21>, 'params': {'max_length': 256}}, {'field_id': 106, 'name': 'filter_p_key', 'description': '', 'type': <DataType.VARCHAR: 21>, 'params': {'max_length': 256}}, {'field_id': 107, 'name': 'table_type', 'description': '', 'type': <DataType.VARCHAR: 21>, 'params': {'max_length': 256}}, {'field_id': 108, 'name': 'describe', 'description': '', 'type': <DataType.VARCHAR: 21>, 'params': {'max_length': 768}, 'default_value': string_data: ""
, 'nullable': True}, {'field_id': 109, 'name': 'created', 'description': '', 'type': <DataType.INT64: 5>, 'params': {}}, {'field_id': 110, 'name': 'updated', 'description': '', 'type': <DataType.INT64: 5>, 'params': {}}], 'functions': [], 'aliases': [], 'collection_id': 458179491513393440, 'consistency_level': 2, 'properties': {}, 'num_partitions': 1, 'enable_dynamic_field': True, 'created_timestamp': 458180054501556226}
3 error
<DataNotMatchException: (code=1, message=Insert missed an field describe to collection without set nullable==true or set default_value)>

[yhmo]

I create a collection with the same schema, and inset some rows, seems it works well.

import random

from pymilvus import (
    MilvusClient, DataType, 
)

client = MilvusClient(uri="http://localhost:19530")
print(client.get_server_version())

collection_name = "filter_value"

def gen_embedding(dim):
    return [random.random()*10+1 for _ in range(dim)]

schema = client.create_schema(enable_dynamic_field=True, description="")
schema.add_field(field_name="kg_id", datatype=DataType.VARCHAR, description="The Primary Key", max_length=256, is_primary=True, auto_id=False)
schema.add_field(field_name="embedding", datatype=DataType.FLOAT_VECTOR, dim=768)
schema.add_field(field_name="node_id", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="data_id", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="name", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="ytenant_id", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="filter_p_key", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="table_type", datatype=DataType.VARCHAR, max_length=256)
schema.add_field(field_name="describe", datatype=DataType.VARCHAR, max_length=768, nullable=True, default_value='')
schema.add_field(field_name="created", datatype=DataType.INT64)
schema.add_field(field_name="updated", datatype=DataType.INT64)

index_params = client.prepare_index_params()
index_params.add_index(field_name="embedding", metric_type="COSINE", index_type="AUTOINDEX")

client.drop_collection(collection_name=collection_name)
client.create_collection(collection_name=collection_name, schema=schema, index_params=index_params)

# no "describe", default is applied
client.insert(collection_name=collection_name,
              data={"kg_id": "1", "embedding": gen_embedding(768), "node_id": "123", "data_id": "456", "name": "TOM",
                    "ytenant_id": "foo", "filter_p_key": "bar", "table_type": "ooo", "created": 123, "updated": 456})

# "describe" = "abcdef"
client.insert(collection_name=collection_name,
              data={"kg_id": "2", "embedding": gen_embedding(768), "node_id": "123", "data_id": "456", "name": "TOM",
                    "ytenant_id": "foo", "filter_p_key": "bar", "table_type": "ooo", "describe": "abcdef", "created": 123, "updated": 456})

# "describe" = None, default is applied
client.insert(collection_name=collection_name,
              data={"kg_id": "3", "embedding": gen_embedding(768), "node_id": "123", "data_id": "456", "name": "TOM",
                    "ytenant_id": "foo", "filter_p_key": "bar", "table_type": "ooo", "describe": None, "created": 123, "updated": 456})


res = client.query(collection_name=collection_name, filter="kg_id ==\"1\"", consistency_level="Strong")
print("describe value:", res[0]["describe"])

res = client.query(collection_name=collection_name, filter="kg_id ==\"2\"", consistency_level="Strong")
print("describe value:", res[0]["describe"])

res = client.query(collection_name=collection_name, filter="kg_id ==\"3\"", consistency_level="Strong")
print("describe value:", res[0]["describe"])

The output:

describe value: 
describe value: abcdef
describe value: