chore: import upstream snapshot with attribution
This commit is contained in:
wehub-resource-sync
@@ -0,0 +1,21 @@
|
||||
# Copyright 2025-present the zvec project
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
from __future__ import annotations
|
||||
|
||||
from zvec._zvec.schema import CollectionStats
|
||||
|
||||
from .collection_schema import CollectionSchema
|
||||
from .field_schema import FieldSchema, VectorSchema
|
||||
|
||||
__all__ = ["CollectionSchema", "CollectionStats", "FieldSchema", "VectorSchema"]
|
||||
@@ -0,0 +1,109 @@
|
||||
"""
|
||||
This module contains the schema of Zvec
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import collections.abc
|
||||
import typing
|
||||
|
||||
import zvec._zvec.param
|
||||
import zvec._zvec.typing
|
||||
|
||||
from .collection_schema import CollectionSchema
|
||||
from .field_schema import FieldSchema, VectorSchema
|
||||
|
||||
__all__: list[str] = [
|
||||
"CollectionSchema",
|
||||
"CollectionStats",
|
||||
"FieldSchema",
|
||||
"VectorSchema",
|
||||
]
|
||||
|
||||
class CollectionStats:
|
||||
def __init__(self) -> None: ...
|
||||
def __repr__(self) -> str: ...
|
||||
@property
|
||||
def doc_count(self) -> int: ...
|
||||
@property
|
||||
def index_completeness(self) -> dict[str, float]: ...
|
||||
|
||||
class _CollectionSchema:
|
||||
__hash__: typing.ClassVar[None] = None
|
||||
|
||||
def __eq__(self, arg0: _CollectionSchema) -> bool: ...
|
||||
def __init__(
|
||||
self, name: str, fields: collections.abc.Sequence[_FieldSchema]
|
||||
) -> None:
|
||||
"""
|
||||
Construct with name and list of fields
|
||||
"""
|
||||
|
||||
def __ne__(self, arg0: _CollectionSchema) -> bool: ...
|
||||
def fields(self) -> list[_FieldSchema]:
|
||||
"""
|
||||
Return list of all field schemas.
|
||||
"""
|
||||
|
||||
def forward_fields(self) -> list[_FieldSchema]:
|
||||
"""
|
||||
Return list of forward-indexed fields.
|
||||
"""
|
||||
|
||||
def get_field(self, field_name: str) -> _FieldSchema:
|
||||
"""
|
||||
Get field by name (const pointer), returns None if not found.
|
||||
"""
|
||||
|
||||
def get_forward_field(self, field_name: str) -> _FieldSchema:
|
||||
"""
|
||||
Get forward field (used for filtering).
|
||||
"""
|
||||
|
||||
def get_vector_field(self, field_name: str) -> _FieldSchema:
|
||||
"""
|
||||
Get vector field by name.
|
||||
"""
|
||||
|
||||
def has_field(self, field_name: str) -> bool:
|
||||
"""
|
||||
Check if a field exists.
|
||||
"""
|
||||
|
||||
def vector_fields(self) -> list[_FieldSchema]:
|
||||
"""
|
||||
Return list of vector fields.
|
||||
"""
|
||||
|
||||
@property
|
||||
def name(self) -> str: ...
|
||||
|
||||
class _FieldSchema:
|
||||
__hash__: typing.ClassVar[None] = None
|
||||
|
||||
def __eq__(self, arg0: _FieldSchema) -> bool: ...
|
||||
def __init__(
|
||||
self,
|
||||
name: str,
|
||||
data_type: zvec._zvec.typing.DataType,
|
||||
nullable: bool = False,
|
||||
dimension: typing.SupportsInt = 0,
|
||||
index_param: zvec._zvec.param.IndexParam = None,
|
||||
) -> None: ...
|
||||
def __ne__(self, arg0: _FieldSchema) -> bool: ...
|
||||
@property
|
||||
def data_type(self) -> zvec._zvec.typing.DataType: ...
|
||||
@property
|
||||
def dimension(self) -> int: ...
|
||||
@property
|
||||
def index_param(self) -> typing.Any: ...
|
||||
@property
|
||||
def index_type(self) -> zvec._zvec.typing.IndexType: ...
|
||||
@property
|
||||
def is_dense_vector(self) -> bool: ...
|
||||
@property
|
||||
def is_sparse_vector(self) -> bool: ...
|
||||
@property
|
||||
def name(self) -> str: ...
|
||||
@property
|
||||
def nullable(self) -> bool: ...
|
||||
@@ -0,0 +1,215 @@
|
||||
# Copyright 2025-present the zvec project
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
from __future__ import annotations
|
||||
|
||||
import json
|
||||
from typing import Optional, Union
|
||||
|
||||
from zvec._zvec.schema import _CollectionSchema, _FieldSchema
|
||||
|
||||
from .field_schema import FieldSchema, VectorSchema
|
||||
|
||||
__all__ = [
|
||||
"CollectionSchema",
|
||||
]
|
||||
|
||||
|
||||
class CollectionSchema:
|
||||
"""Defines the structure of a collection in Zvec.
|
||||
|
||||
A collection schema specifies the name of the collection and its fields,
|
||||
including both scalar fields (e.g., int, string) and vector fields.
|
||||
Field names must be unique across both scalar and vector fields.
|
||||
|
||||
Args:
|
||||
name (str): Name of the collection.
|
||||
fields (Optional[Union[FieldSchema, list[FieldSchema]]], optional):
|
||||
One or more scalar field definitions. Defaults to None.
|
||||
vectors (Optional[Union[VectorSchema, list[VectorSchema]]], optional):
|
||||
One or more vector field definitions. Defaults to None.
|
||||
|
||||
Raises:
|
||||
TypeError: If `fields` or `vectors` are of unsupported types.
|
||||
ValueError: If any field or vector name is duplicated.
|
||||
|
||||
Examples:
|
||||
>>> from zvec import FieldSchema, VectorSchema, DataType, IndexType
|
||||
>>> id_field = FieldSchema("id", DataType.INT64, is_primary=True)
|
||||
>>> emb_field = VectorSchema("embedding", dim=128, data_type=DataType.VECTOR_FP32)
|
||||
>>> schema = CollectionSchema(
|
||||
... name="my_collection",
|
||||
... fields=id_field,
|
||||
... vectors=emb_field
|
||||
... )
|
||||
>>> print(schema.name)
|
||||
my_collection
|
||||
"""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
name: str,
|
||||
fields: Optional[Union[FieldSchema, list[FieldSchema]]] = None,
|
||||
vectors: Optional[Union[VectorSchema, list[VectorSchema]]] = None,
|
||||
):
|
||||
if name is None or not isinstance(name, str):
|
||||
raise ValueError(
|
||||
f"schema validate failed: collection name must be str, got {type(name).__name__}"
|
||||
)
|
||||
|
||||
# handle fields
|
||||
_fields_name: list[str] = []
|
||||
_fields_list: list[_FieldSchema] = []
|
||||
|
||||
self._check_fields(fields, _fields_name, _fields_list)
|
||||
self._check_vectors(vectors, _fields_name, _fields_list)
|
||||
|
||||
# init
|
||||
self._cpp_obj = _CollectionSchema(
|
||||
name=name,
|
||||
fields=_fields_list,
|
||||
)
|
||||
|
||||
def _check_fields(
|
||||
self,
|
||||
fields: Optional[Union[FieldSchema, list[FieldSchema]]],
|
||||
_fields_name: list[str],
|
||||
_fields_list: list[_FieldSchema],
|
||||
) -> None:
|
||||
field_items = []
|
||||
|
||||
if isinstance(fields, FieldSchema):
|
||||
field_items = [fields]
|
||||
elif isinstance(fields, list):
|
||||
field_items = fields
|
||||
elif fields is None:
|
||||
field_items = []
|
||||
else:
|
||||
raise TypeError(
|
||||
f"schema validate failed: invalid 'fields' type, expected FieldSchema or list[FieldSchema], "
|
||||
f"got {type(fields).__name__}"
|
||||
)
|
||||
|
||||
for idx, field in enumerate(field_items):
|
||||
if not isinstance(field, FieldSchema):
|
||||
raise TypeError(
|
||||
f"schema validate failed: invalid field type in 'fields' list, expected FieldSchema, "
|
||||
f"got {type(field).__name__} at index {idx}"
|
||||
)
|
||||
|
||||
if field.name in _fields_name:
|
||||
raise ValueError(
|
||||
f"schema validate failed: duplicate field name '{field.name}': field names must be unique"
|
||||
)
|
||||
_fields_name.append(field.name)
|
||||
_fields_list.append(field._get_object())
|
||||
|
||||
def _check_vectors(
|
||||
self,
|
||||
vectors: Optional[Union[VectorSchema, list[VectorSchema]]],
|
||||
_fields_name: list[str],
|
||||
_fields_list: list[_FieldSchema],
|
||||
) -> None:
|
||||
# handle vector
|
||||
if isinstance(vectors, VectorSchema):
|
||||
vectors_items = [vectors]
|
||||
elif isinstance(vectors, list):
|
||||
vectors_items = vectors
|
||||
elif vectors is None:
|
||||
vectors_items = []
|
||||
else:
|
||||
raise TypeError(
|
||||
f"schema validate failed: invalid 'vectors' type, expected VectorSchema or list[VectorSchema], "
|
||||
f"got {type(vectors).__name__}"
|
||||
)
|
||||
|
||||
for idx, vector in enumerate(vectors_items):
|
||||
if not isinstance(vector, VectorSchema):
|
||||
raise TypeError(
|
||||
f"schema validate failed: invalid vector type in 'vectors' list, expected VectorSchema, "
|
||||
f"got {type(vector).__name__} at index {idx}"
|
||||
)
|
||||
|
||||
if vector.name in _fields_name:
|
||||
raise ValueError(
|
||||
f"schema validate failed: duplicate vector name '{vector.name}', vector names must be unique "
|
||||
f"(conflicts with existing field or vector)"
|
||||
)
|
||||
_fields_name.append(vector.name)
|
||||
_fields_list.append(vector._get_object())
|
||||
|
||||
@classmethod
|
||||
def _from_core(cls, core_collection_schema: _CollectionSchema):
|
||||
inst = cls.__new__(cls)
|
||||
if not core_collection_schema:
|
||||
raise ValueError("schema validate failed: schema is null")
|
||||
inst._cpp_obj = core_collection_schema
|
||||
return inst
|
||||
|
||||
@property
|
||||
def name(self) -> str:
|
||||
"""str: The name of the collection."""
|
||||
return self._cpp_obj.name
|
||||
|
||||
def field(self, name: str) -> Optional[FieldSchema]:
|
||||
"""Retrieve a scalar field by name.
|
||||
|
||||
Args:
|
||||
name (str): Name of the field.
|
||||
|
||||
Returns:
|
||||
Optional[FieldSchema]: The field if found, otherwise None.
|
||||
"""
|
||||
_field = self._cpp_obj.get_forward_field(name)
|
||||
return FieldSchema._from_core(_field) if _field else None
|
||||
|
||||
def vector(self, name: str) -> Optional[VectorSchema]:
|
||||
"""Retrieve a vector field by name.
|
||||
|
||||
Args:
|
||||
name (str): Name of the vector field.
|
||||
|
||||
Returns:
|
||||
Optional[VectorSchema]: The vector field if found, otherwise None.
|
||||
"""
|
||||
_field = self._cpp_obj.get_vector_field(name)
|
||||
return VectorSchema._from_core(_field) if _field else None
|
||||
|
||||
@property
|
||||
def fields(self) -> list[FieldSchema]:
|
||||
"""list[FieldSchema]: All scalar (non-vector) fields in the schema."""
|
||||
_fields = self._cpp_obj.forward_fields()
|
||||
return [FieldSchema._from_core(_field) for _field in _fields]
|
||||
|
||||
@property
|
||||
def vectors(self) -> list[VectorSchema]:
|
||||
"""list[VectorSchema]: All vector fields in the schema."""
|
||||
_vectors = self._cpp_obj.vector_fields()
|
||||
return [VectorSchema._from_core(_vector) for _vector in _vectors]
|
||||
|
||||
def _get_object(self) -> _CollectionSchema:
|
||||
return self._cpp_obj
|
||||
|
||||
def __repr__(self) -> str:
|
||||
try:
|
||||
schema = {
|
||||
"name": self.name,
|
||||
"fields": {field.name: field.__dict__() for field in self.fields},
|
||||
"vectors": {vector.name: vector.__dict__() for vector in self.vectors},
|
||||
}
|
||||
return json.dumps(schema, indent=2, ensure_ascii=False)
|
||||
except Exception as e:
|
||||
return f"<CollectionSchema error during repr: {e}>"
|
||||
|
||||
def __str__(self) -> str:
|
||||
return self.__repr__()
|
||||
@@ -0,0 +1,310 @@
|
||||
# Copyright 2025-present the zvec project
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
from __future__ import annotations
|
||||
|
||||
import json
|
||||
from typing import Any, Optional, Union
|
||||
|
||||
from zvec._zvec.schema import _FieldSchema
|
||||
from zvec.model.param import (
|
||||
FlatIndexParam,
|
||||
FtsIndexParam,
|
||||
HnswIndexParam,
|
||||
HnswRabitqIndexParam,
|
||||
InvertIndexParam,
|
||||
IVFIndexParam,
|
||||
)
|
||||
from zvec.typing import DataType
|
||||
|
||||
__all__ = [
|
||||
"FieldSchema",
|
||||
"VectorSchema",
|
||||
]
|
||||
|
||||
SUPPORT_VECTOR_DATA_TYPE = [
|
||||
DataType.VECTOR_FP16,
|
||||
DataType.VECTOR_FP32,
|
||||
DataType.VECTOR_FP64,
|
||||
DataType.VECTOR_INT8,
|
||||
DataType.SPARSE_VECTOR_FP16,
|
||||
DataType.SPARSE_VECTOR_FP32,
|
||||
]
|
||||
|
||||
SUPPORT_SCALAR_DATA_TYPE = [
|
||||
DataType.INT32,
|
||||
DataType.INT64,
|
||||
DataType.UINT32,
|
||||
DataType.UINT64,
|
||||
DataType.FLOAT,
|
||||
DataType.DOUBLE,
|
||||
DataType.STRING,
|
||||
DataType.BOOL,
|
||||
DataType.ARRAY_INT32,
|
||||
DataType.ARRAY_INT64,
|
||||
DataType.ARRAY_UINT32,
|
||||
DataType.ARRAY_UINT64,
|
||||
DataType.ARRAY_FLOAT,
|
||||
DataType.ARRAY_DOUBLE,
|
||||
DataType.ARRAY_STRING,
|
||||
DataType.ARRAY_BOOL,
|
||||
]
|
||||
|
||||
|
||||
class FieldSchema:
|
||||
"""Represents a scalar (non-vector) field in a collection schema.
|
||||
|
||||
A `FieldSchema` defines the name, data type, nullability, and optional
|
||||
inverted index configuration for a regular field (e.g., ID, timestamp, category).
|
||||
|
||||
Args:
|
||||
name (str): Name of the field. Must be unique within the collection.
|
||||
data_type (DataType): Data type of the field (e.g., INT64, STRING).
|
||||
nullable (bool, optional): Whether the field can contain null values.
|
||||
Defaults to False.
|
||||
index_param (Optional[Union[InvertIndexParam, FtsIndexParam]], optional):
|
||||
Index parameters for this field. Use ``InvertIndexParam`` for scalar
|
||||
inverted indexing, or ``FtsIndexParam`` for full-text search indexing
|
||||
on STRING fields. Defaults to None.
|
||||
|
||||
Examples:
|
||||
>>> from zvec.typing import DataType
|
||||
>>> from zvec.model.param import InvertIndexParam, FtsIndexParam
|
||||
>>> id_field = FieldSchema(
|
||||
... name="id",
|
||||
... data_type=DataType.INT64,
|
||||
... nullable=False,
|
||||
... index_param=InvertIndexParam(enable_range_optimization=True)
|
||||
... )
|
||||
>>> content_field = FieldSchema(
|
||||
... name="content",
|
||||
... data_type=DataType.STRING,
|
||||
... nullable=False,
|
||||
... index_param=FtsIndexParam(tokenizer_name="standard")
|
||||
... )
|
||||
"""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
name: str,
|
||||
data_type: DataType,
|
||||
nullable: bool = False,
|
||||
index_param: Optional[Union[InvertIndexParam, FtsIndexParam]] = None,
|
||||
):
|
||||
if name is None or not isinstance(name, str):
|
||||
raise ValueError(
|
||||
f"schema validate failed: field name must be str, got {type(name).__name__}"
|
||||
)
|
||||
|
||||
if data_type not in SUPPORT_SCALAR_DATA_TYPE:
|
||||
raise ValueError(
|
||||
f"schema validate failed: scalar_field's data_type must be one of "
|
||||
f"{', '.join(str(dt) for dt in SUPPORT_SCALAR_DATA_TYPE)}, "
|
||||
f"but field[{name}]'s data_type is {data_type}"
|
||||
)
|
||||
|
||||
self._cpp_obj = _FieldSchema(
|
||||
name=name,
|
||||
data_type=data_type,
|
||||
dimension=0,
|
||||
nullable=nullable,
|
||||
index_param=index_param,
|
||||
)
|
||||
|
||||
@classmethod
|
||||
def _from_core(cls, core_field_schema: _FieldSchema):
|
||||
if core_field_schema is None:
|
||||
raise ValueError("schema validate failed: field schema is None")
|
||||
inst = cls.__new__(cls)
|
||||
inst._cpp_obj = core_field_schema
|
||||
return inst
|
||||
|
||||
def _get_object(self) -> _FieldSchema:
|
||||
return self._cpp_obj
|
||||
|
||||
@property
|
||||
def name(self) -> str:
|
||||
"""str: The name of the field."""
|
||||
return self._cpp_obj.name
|
||||
|
||||
@property
|
||||
def data_type(self) -> DataType:
|
||||
"""DataType: The data type of the field (e.g., INT64, STRING)."""
|
||||
return self._cpp_obj.data_type
|
||||
|
||||
@property
|
||||
def nullable(self) -> bool:
|
||||
"""bool: Whether the field allows null values."""
|
||||
return self._cpp_obj.nullable
|
||||
|
||||
@property
|
||||
def index_param(self) -> Optional[Union[InvertIndexParam, FtsIndexParam]]:
|
||||
"""Optional[Union[InvertIndexParam, FtsIndexParam]]: Index configuration, if any."""
|
||||
return self._cpp_obj.index_param
|
||||
|
||||
def __dict__(self) -> dict[str, Any]:
|
||||
return {
|
||||
"name": self.name,
|
||||
"data_type": (
|
||||
self.data_type.name
|
||||
if hasattr(self.data_type, "name")
|
||||
else str(self.data_type)
|
||||
),
|
||||
"nullable": self.nullable,
|
||||
"index_param": (
|
||||
self.index_param.to_dict() if self.index_param is not None else None
|
||||
),
|
||||
}
|
||||
|
||||
def __repr__(self) -> str:
|
||||
try:
|
||||
schema = self.__dict__()
|
||||
return json.dumps(schema, indent=2, ensure_ascii=False)
|
||||
except Exception as e:
|
||||
return f"<FieldSchema error during repr: {e}>"
|
||||
|
||||
def __str__(self) -> str:
|
||||
return self.__repr__()
|
||||
|
||||
def __eq__(self, other: object) -> bool:
|
||||
if not isinstance(other, FieldSchema):
|
||||
return False
|
||||
return self._cpp_obj == other._cpp_obj
|
||||
|
||||
def __hash__(self) -> int:
|
||||
return hash((self.name, self.data_type, self.nullable))
|
||||
|
||||
|
||||
class VectorSchema:
|
||||
"""Represents a vector field in a collection schema.
|
||||
|
||||
A `VectorSchema` defines the name, data type, dimensionality, and index
|
||||
configuration for a vector field used in similarity search.
|
||||
|
||||
Args:
|
||||
name (str): Name of the vector field. Must be unique within the collection.
|
||||
data_type (DataType): Vector data type (e.g., VECTOR_FP32, VECTOR_INT8).
|
||||
dimension (int, optional): Dimensionality of the vector. Must be > 0 for dense vectors;
|
||||
may be `None` for sparse vectors.
|
||||
index_param (Union[HnswIndexParam, IVFIndexParam, FlatIndexParam], optional):
|
||||
Index configuration for this vector field. Defaults to
|
||||
``HnswIndexParam()``.
|
||||
|
||||
Examples:
|
||||
>>> from zvec.typing import DataType
|
||||
>>> from zvec.model.param import HnswIndexParam
|
||||
>>> emb_field = VectorSchema(
|
||||
... name="embedding",
|
||||
... data_type=DataType.VECTOR_FP32,
|
||||
... dimension=128,
|
||||
... index_param=HnswIndexParam(ef_construction=200, m=16)
|
||||
... )
|
||||
"""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
name: str,
|
||||
data_type: DataType,
|
||||
dimension: Optional[int] = 0,
|
||||
index_param: Optional[
|
||||
Union[HnswIndexParam, HnswRabitqIndexParam, FlatIndexParam, IVFIndexParam]
|
||||
] = None,
|
||||
):
|
||||
if name is None or not isinstance(name, str):
|
||||
raise ValueError(
|
||||
f"schema validate failed: field name must be str, got {type(name).__name__}"
|
||||
)
|
||||
|
||||
if not isinstance(dimension, int) or dimension < 0:
|
||||
raise ValueError("schema validate failed: vector's dimension must be >= 0")
|
||||
|
||||
if data_type not in SUPPORT_VECTOR_DATA_TYPE:
|
||||
raise ValueError(
|
||||
f"schema validate failed: vector's data_type must be one of "
|
||||
f"{', '.join(str(dt) for dt in SUPPORT_VECTOR_DATA_TYPE)}, "
|
||||
f"but field[{name}]'s data_type is {data_type}"
|
||||
)
|
||||
|
||||
if index_param is None:
|
||||
index_param = FlatIndexParam()
|
||||
|
||||
self._cpp_obj = _FieldSchema(
|
||||
name=name,
|
||||
data_type=data_type,
|
||||
dimension=dimension,
|
||||
nullable=False,
|
||||
index_param=index_param,
|
||||
)
|
||||
|
||||
@classmethod
|
||||
def _from_core(cls, core_field_schema: _FieldSchema):
|
||||
inst = cls.__new__(cls)
|
||||
inst._cpp_obj = core_field_schema
|
||||
return inst
|
||||
|
||||
def _get_object(self) -> _FieldSchema:
|
||||
return self._cpp_obj
|
||||
|
||||
@property
|
||||
def name(self) -> str:
|
||||
"""str: The name of the vector field."""
|
||||
return self._cpp_obj.name
|
||||
|
||||
@property
|
||||
def data_type(self) -> DataType:
|
||||
"""DataType: The vector data type (e.g., VECTOR_FP32)."""
|
||||
return self._cpp_obj.data_type
|
||||
|
||||
@property
|
||||
def dimension(self) -> int:
|
||||
"""int: The dimensionality of the vector."""
|
||||
return self._cpp_obj.dimension
|
||||
|
||||
@property
|
||||
def index_param(
|
||||
self,
|
||||
) -> Union[HnswIndexParam, HnswRabitqIndexParam, IVFIndexParam, FlatIndexParam]:
|
||||
"""Union[HnswIndexParam, HnswRabitqIndexParam, IVFIndexParam, FlatIndexParam]: Index configuration for the vector."""
|
||||
return self._cpp_obj.index_param
|
||||
|
||||
def __dict__(self) -> dict[str, Any]:
|
||||
return {
|
||||
"name": self.name,
|
||||
"data_type": (
|
||||
self.data_type.name
|
||||
if hasattr(self.data_type, "name")
|
||||
else str(self.data_type)
|
||||
),
|
||||
"dimension": self.dimension,
|
||||
"index_param": (
|
||||
self.index_param.to_dict() if self.index_param is not None else None
|
||||
),
|
||||
}
|
||||
|
||||
def __repr__(self) -> str:
|
||||
try:
|
||||
schema = self.__dict__()
|
||||
return json.dumps(schema, indent=2, ensure_ascii=False)
|
||||
except Exception as e:
|
||||
return f"<FieldSchema error during repr: {e}>"
|
||||
|
||||
def __str__(self) -> str:
|
||||
return self.__repr__()
|
||||
|
||||
def __eq__(self, other: object) -> bool:
|
||||
if not isinstance(other, VectorSchema):
|
||||
return False
|
||||
return self._cpp_obj == other._cpp_obj
|
||||
|
||||
def __hash__(self) -> int:
|
||||
return hash((self.name, self.data_type, self.dimension))
|
||||
Reference in New Issue
Block a user