FieldSchema

Defines the schema of a single field in a CollectionSchema.

public sealed class FieldSchema
{
    public FieldSchema(string name, DataType dataType, bool isPrimaryKey = false, bool autoId = false, bool isPartitionKey = false, string description = "")

    public static FieldSchema CreateVarchar(string name, int maxLength, bool isPrimaryKey = false, bool autoId = false, bool isPartitionKey = false, string description = "", bool nullable = false, string? defaultValue = null, bool enableAnalyzer = false, IReadOnlyDictionary<string, object>? analyzerParams = null, bool enableMatch = false, IReadOnlyDictionary<string, object>? multiAnalyzerParams = null)

    public static FieldSchema CreateFloatVector(string name, int dimension, string description = "")

    public static FieldSchema CreateSparseFloatVector(string name, string description = "")

    public string Name { get; }
    public DataType DataType { get; }
    public bool IsPrimaryKey { get; }
    public bool AutoId { get; }
    public bool IsPartitionKey { get; }
    public string Description { get; }
    public int? MaxLength { get; set; }
    public int? Dimension { get; set; }
    public int? MaxCapacity { get; set; }
    public DataType? ElementDataType { get; set; }
    public bool Nullable { get; set; }
    public object? DefaultValue { get; set; }
    public bool EnableAnalyzer { get; set; }
    public IReadOnlyDictionary<string, object>? AnalyzerParams { get; set; }
    public bool EnableMatch { get; set; }
    public IReadOnlyDictionary<string, object>? MultiAnalyzerParams { get; set; }
    public bool IsFunctionOutput { get; internal set; }
    public bool IsClusteringKey { get; set; }
    public bool IsDynamic { get; internal set; }
    public long FieldId { get; internal set; }
    public IReadOnlyDictionary<string, string>? RawTypeParams { get; internal set; }
    public IDictionary<string, string> TypeParams { get; }
}

PARAMETERS:

  • Name (string) -

    The field name.

  • DataType (DataType) -

    The data type of the field. See DataType.

  • IsPrimaryKey (bool) -

    Whether the field is the primary key of the collection.

  • AutoId (bool) -

    Whether primary key values are automatically generated by Milvus.

  • IsPartitionKey (bool) -

    Whether the field is a partition key.

  • Description (string) -

    An optional description for the field.

  • MaxLength (int?) -

    The maximum length of a VarChar field.

  • Dimension (int?) -

    The dimensionality of a vector field.

  • MaxCapacity (int?) -

    The maximum capacity of an array field.

  • ElementDataType (DataType?) -

    The element data type of an array field. See DataType.

  • Nullable (bool) -

    Whether the field is nullable.

  • DefaultValue (object?) -

    The default value for the field.

  • EnableAnalyzer (bool) -

    Whether the analyzer is enabled for this field (BM25 full-text search input fields).

  • AnalyzerParams (IReadOnlyDictionary?) -

    The analyzer parameters, e.g. new Dictionary<string, object> { ["type"] = "english" }. Mutually exclusive with MultiAnalyzerParams.

  • EnableMatch (bool) -

    Whether match (BM25 keyword) search is enabled for this field.

  • MultiAnalyzerParams (IReadOnlyDictionary?) -

    The multi-language analyzer parameters for the field. Mutually exclusive with AnalyzerParams.

  • IsFunctionOutput (bool) -

    Whether this field is the output of a function (e.g. a BM25 sparse vector field).

  • IsClusteringKey (bool) -

    Whether this field is a clustering key.

  • IsDynamic (bool) -

    Whether this field is the dynamic ($meta) field. Only populated when the schema is read back via DescribeCollection.

  • FieldId (long) -

    The server-assigned field id, populated when the schema is read back via DescribeCollection.

  • RawTypeParams (IReadOnlyDictionary?) -

    The raw type_params key-value map as returned by DescribeCollection, preserving any keys the SDK does not parse into typed properties.

  • TypeParams (IDictionary) -

    Additional per-field type_params to send when creating the collection. Keys that map to a typed property (e.g. max_length, dim) are ignored in favor of the typed property.

METHODS:

  • FieldSchema(string name, DataType dataType, bool isPrimaryKey = false, bool autoId = false, bool isPartitionKey = false, string description = "") -

    Creates a new field schema. See DataType.

  • CreateVarchar(...) (static) -

    Creates a new VarChar field schema. maxLength is the maximum length of the varchar field and must be greater than zero. See DataType.

  • CreateFloatVector(...) (static) -

    Creates a new FloatVector field schema. dimension is the dimensionality of the vector and must be at least 2. See DataType.

  • CreateSparseFloatVector(...) (static) -

    Creates a new SparseFloatVector field schema. See DataType.

Example

using Milvus.Client.V2;
using Milvus.Client.V2.Requests.Collection;
using Milvus.Client.V2.Types;

using var client = new MilvusClientV2(new ConnectConfig
{
    Uri = "localhost:19530",
    Username = "root",
    Password = "Milvus"
});

var schema = new CollectionSchema { Name = "my_collection" };
schema.Fields.Add(new FieldSchema("id", DataType.Int64, isPrimaryKey: true, autoId: true));
schema.Fields.Add(FieldSchema.CreateVarchar("title", maxLength: 256, nullable: true, defaultValue: "untitled"));
schema.Fields.Add(FieldSchema.CreateVarchar("body", maxLength: 1024, enableAnalyzer: true,
    analyzerParams: new Dictionary<string, object> { ["type"] = "english" }));
schema.Fields.Add(FieldSchema.CreateFloatVector("embedding", dimension: 128));

await client.CreateCollectionAsync(new CreateCollectionReq
{
    CollectionName = "my_collection",
    Schema = schema
});

Try Managed Milvus for Free

Zilliz Cloud is hassle-free, powered by Milvus and 10x faster.

Get Started
Feedback

Was this page helpful?