refactor: remove not used storage class and blob dataset

java/lance-jni/src/schema.rs

@@ -9,7 +9,7 @@ use arrow_schema::{TimeUnit, UnionFields};
 use jni::objects::{JObject, JValue};
 use jni::sys::{jboolean, jint};
 use jni::JNIEnv;
-use lance_core::datatypes::{Field, Schema, StorageClass};
+use lance_core::datatypes::{Field, Schema};
 
 impl IntoJava for Schema {
     fn into_java<'local>(self, env: &mut JNIEnv<'local>) -> Result<JObject<'local>> {
@@ -40,11 +40,8 @@ pub fn convert_to_java_field<'local>(
     let children = convert_children_fields(env, lance_field)?;
     let metadata = to_java_map(env, &lance_field.metadata)?;
     let arrow_type = convert_arrow_type(env, &lance_field.data_type())?;
-    let storage_type = convert_storage_type(env, &lance_field.storage_class)?;
-
     let ctor_sig = "(IILjava/lang/String;".to_owned()
         + "ZLorg/apache/arrow/vector/types/pojo/ArrowType;"
-        + "Lcom/lancedb/lance/schema/StorageType;"
         + "Lorg/apache/arrow/vector/types/pojo/DictionaryEncoding;"
         + "Ljava/util/Map;"
         + "Ljava/util/List;Z)V";
@@ -57,7 +54,6 @@ pub fn convert_to_java_field<'local>(
             JValue::Object(&JObject::from(name)),
             JValue::Bool(lance_field.nullable as jboolean),
             JValue::Object(&arrow_type),
-            JValue::Object(&storage_type),
             JValue::Object(&JObject::null()),
             JValue::Object(&metadata),
             JValue::Object(&children),
@@ -68,25 +64,6 @@ pub fn convert_to_java_field<'local>(
     Ok(field_obj)
 }
 
-fn convert_storage_type<'local>(
-    env: &mut JNIEnv<'local>,
-    storage_class: &StorageClass,
-) -> Result<JObject<'local>> {
-    let jname = match storage_class {
-        StorageClass::Blob => env.new_string("BLOB")?,
-        _ => env.new_string("DEFAULT")?,
-    };
-
-    Ok(env
-        .call_static_method(
-            "com/lancedb/lance/schema/StorageType",
-            "valueOf",
-            "(Ljava/lang/String;)Lcom/lancedb/lance/schema/StorageType;",
-            &[JValue::Object(&JObject::from(jname))],
-        )?
-        .l()?)
-}
-
 fn convert_children_fields<'local>(
     env: &mut JNIEnv<'local>,
     lance_field: &Field,

java/lance-jni/src/transaction.rs

@@ -397,18 +397,16 @@ fn convert_to_java_transaction<'local>(
         Some(properties) => to_java_map(env, &properties)?,
         _ => JObject::null(),
     };
-    let operation = convert_to_java_operation_inner(env, transaction.operation)?;
-    let blobs_op = convert_to_java_operation(env, transaction.blobs_op)?;
+    let operation = convert_to_java_operation(env, Some(transaction.operation))?;
 
     let java_transaction = env.new_object(
         "com/lancedb/lance/Transaction",
-        "(Lcom/lancedb/lance/Dataset;JLjava/lang/String;Lcom/lancedb/lance/operation/Operation;Lcom/lancedb/lance/operation/Operation;Ljava/util/Map;Ljava/util/Map;)V",
+        "(Lcom/lancedb/lance/Dataset;JLjava/lang/String;Lcom/lancedb/lance/operation/Operation;Ljava/util/Map;Ljava/util/Map;)V",
         &[
             JValue::Object(java_dataset),
             JValue::Long(transaction.read_version as i64),
             JValue::Object(&uuid),
             JValue::Object(&operation),
-            JValue::Object(&blobs_op),
             JValue::Object(&JObject::null()),
             JValue::Object(&transaction_properties),
         ],
@@ -707,11 +705,6 @@ fn convert_to_rust_transaction(
         .l()?;
     let op = convert_to_rust_operation(env, &op, java_dataset)?;
 
-    let blobs_op =
-        env.get_optional_from_method(&java_transaction, "blobsOperation", |env, blobs_op| {
-            convert_to_rust_operation(env, &blobs_op, java_dataset)
-        })?;
-
     let transaction_properties = env.get_optional_from_method(
         &java_transaction,
         "transactionProperties",
@@ -722,7 +715,6 @@ fn convert_to_rust_transaction(
     )?;
     Ok(TransactionBuilder::new(read_ver, op)
         .uuid(uuid)
-        .blobs_op(blobs_op)
         .transaction_properties(transaction_properties.map(Arc::new))
         .build())
 }

java/src/main/java/com/lancedb/lance/Transaction.java

@@ -37,21 +37,18 @@ public class Transaction {
   // Mainly for JNI usage
   private final Dataset dataset;
   private final Operation operation;
-  private final Optional<Operation> blobOp;
 
   private Transaction(
       Dataset dataset,
       long readVersion,
       String uuid,
       Operation operation,
-      Operation blobOp,
       Map<String, String> writeParams,
       Map<String, String> transactionProperties) {
     this.dataset = dataset;
     this.readVersion = readVersion;
     this.uuid = uuid;
     this.operation = operation;
-    this.blobOp = Optional.ofNullable(blobOp);
     this.writeParams = writeParams != null ? writeParams : new HashMap<>();
     this.transactionProperties = Optional.ofNullable(transactionProperties);
   }
@@ -68,10 +65,6 @@ public Operation operation() {
     return operation;
   }
 
-  public Optional<Operation> blobsOperation() {
-    return blobOp;
-  }
-
   public Map<String, String> writeParams() {
     return writeParams;
   }
@@ -89,7 +82,6 @@ public Dataset commit() {
 
   public void release() {
     operation.release();
-    blobOp.ifPresent(Operation::release);
   }
 
   @Override
@@ -99,7 +91,6 @@ public String toString() {
         .add("uuid", uuid)
         .add("operation", operation)
         .add("writeParams", writeParams)
-        .add("blobOp", blobOp)
         .add("transactionProperties", transactionProperties)
         .toString();
   }
@@ -116,7 +107,6 @@ public boolean equals(Object o) {
     return readVersion == that.readVersion
         && uuid.equals(that.uuid)
         && Objects.equals(operation, that.operation)
-        && Objects.equals(blobOp, that.blobOp)
         && Objects.equals(writeParams, that.writeParams)
         && Objects.equals(transactionProperties, that.transactionProperties);
   }
@@ -126,7 +116,6 @@ public static class Builder {
     private final Dataset dataset;
     private long readVersion;
     private Operation operation;
-    private Operation blobOp;
     private Map<String, String> writeParams;
     private Map<String, String> transactionProperties;
 
@@ -156,11 +145,6 @@ public Builder operation(Operation operation) {
       return this;
     }
 
-    public Builder blobsOperation(Operation blobOp) {
-      this.blobOp = blobOp;
-      return this;
-    }
-
     private void validateState() {
       if (operation != null) {
         throw new IllegalStateException(
@@ -171,7 +155,7 @@ private void validateState() {
     public Transaction build() {
       Preconditions.checkState(operation != null, "TransactionBuilder has no operations");
       return new Transaction(
-          dataset, readVersion, uuid, operation, blobOp, writeParams, transactionProperties);
+          dataset, readVersion, uuid, operation, writeParams, transactionProperties);
     }
   }
 }

java/src/main/java/com/lancedb/lance/schema/LanceField.java

@@ -30,7 +30,6 @@ public class LanceField {
   private final String name;
   private final boolean nullable;
   private final ArrowType type;
-  private final StorageType storageType;
   private final DictionaryEncoding dictionaryEncoding;
   private final Map<String, String> metadata;
   private final List<LanceField> children;
@@ -42,7 +41,6 @@ public class LanceField {
       String name,
       boolean nullable,
       ArrowType type,
-      StorageType storageType,
       DictionaryEncoding dictionaryEncoding,
       Map<String, String> metadata,
       List<LanceField> children,
@@ -52,7 +50,6 @@ public class LanceField {
     this.name = name;
     this.nullable = nullable;
     this.type = type;
-    this.storageType = storageType;
     this.dictionaryEncoding = dictionaryEncoding;
     this.metadata = metadata;
     this.children = children;
@@ -79,10 +76,6 @@ public ArrowType getType() {
     return type;
   }
 
-  public StorageType getStorageType() {
-    return storageType;
-  }
-
   public Optional<DictionaryEncoding> getDictionaryEncoding() {
     return Optional.ofNullable(dictionaryEncoding);
   }
@@ -114,7 +107,6 @@ public String toString() {
         .add("name", name)
         .add("nullable", nullable)
         .add("type", type)
-        .add("storageType", storageType)
         .add("dictionaryEncoding", dictionaryEncoding)
         .add("children", children)
         .add("isUnenforcedPrimaryKey", isUnenforcedPrimaryKey)

java/src/test/java/com/lancedb/lance/TestUtils.java

@@ -628,11 +628,8 @@ public static final class BlobTestDataset {
     /**
      * Build the Arrow schema with a filter column and a blob column marked as blob storage.
      *
-     * <p>Columns: - filterer: Int64 (not nullable) - blobs: Binary (nullable) with metadata
-     * {"lance-schema:storage-class":"blob"}
-     *
-     * <p>Note: ArrowType.LargeBinary may not be available in our Arrow Java version; Binary is
-     * sufficient for tests and aligns with Lance blob storage when annotated via metadata.
+     * <p>Columns: - filterer: Int64 (not nullable) - blobs: LargeBinary (nullable) annotated with
+     * metadata {"lance-encoding:blob":"true"}
      */
     public Schema getSchema() {
       Map<String, String> blobMeta = Maps.newHashMap();

protos/file.proto

@@ -15,16 +15,16 @@ message FileDescriptor {
 
 // A schema which describes the data type of each of the columns
 message Schema {
-    // All fields in this file, including the nested fields.
-    repeated lance.file.Field fields = 1;
-    // Schema metadata.
-    map<string, bytes> metadata = 5;
+  // All fields in this file, including the nested fields.
+  repeated lance.file.Field fields = 1;
+  // Schema metadata.
+  map<string, bytes> metadata = 5;
 }
 
 // Metadata of one Lance file.
 message Metadata {
-  // 4 was used for StatisticsMetadata in the past, but has been moved to prevent
-  // a bug in older readers.
+  // 4 was used for StatisticsMetadata in the past, but has been moved to
+  // prevent a bug in older readers.
   reserved 4;
 
   // Position of the manifest in the file. If it is zero, the manifest is stored
@@ -44,7 +44,7 @@ message Metadata {
   // contiguously stored.
   //
   // Every field that is a part of the file will have a run in the page table.
-  // This includes struct columns, which will have a run of length 0 since 
+  // This includes struct columns, which will have a run of length 0 since
   // they don't store any actual data.
   //
   // For example, for the column 5 and batch 4, we have:
@@ -57,7 +57,7 @@ message Metadata {
   message StatisticsMetadata {
     // The schema of the statistics.
     //
-    // This might be empty, meaning there are no statistics. It also might not 
+    // This might be empty, meaning there are no statistics. It also might not
     // contain statistics for every field.
     repeated Field schema = 1;
 
@@ -70,20 +70,20 @@ message Metadata {
 
     // The file position of the statistics page table
     //
-    // The page table is a matrix of N x 2, where N = length of stats_fields. This is
-    // the same layout as the main page table, except there is always only one
-    // batch.
+    // The page table is a matrix of N x 2, where N = length of stats_fields.
+    // This is the same layout as the main page table, except there is always
+    // only one batch.
     //
     // For example, to get the stats column 5, we have:
     // ```text
     //   position = stats_page_table[5][0];
     //   length = stats_page_table[5][1];
     // ```
-      uint64 page_table_position = 3;
+    uint64 page_table_position = 3;
   }
 
   StatisticsMetadata statistics = 5;
-} // Metadata
+}  // Metadata
 
 // Supported encodings.
 enum Encoding {
@@ -154,7 +154,8 @@ message Field {
   // * "date32:day"
   // * "date64:ms"
   // * "decimal:128:{precision}:{scale}" / "decimal:256:{precision}:{scale}"
-  // * "time:{unit}" / "timestamp:{unit}" / "duration:{unit}", where unit is "s", "ms", "us", "ns"
+  // * "time:{unit}" / "timestamp:{unit}" / "duration:{unit}", where unit is
+  // "s", "ms", "us", "ns"
   // * "dict:{value_type}:{index_type}:false"
   string logical_type = 5;
   // If this field is nullable.
@@ -168,24 +169,18 @@ message Field {
   /// The logic type presents the value type of the column, i.e., string value.
   Dictionary dictionary = 8;
 
-  // Deprecated: optional extension type name, use metadata field ARROW:extension:name
+  // Deprecated: optional extension type name, use metadata field
+  // ARROW:extension:name
   string extension_name = 9;
 
   // optional field metadata (e.g. extension type name/parameters)
   map<string, bytes> metadata = 10;
 
-  /// The storage class of the field
-  ///
-  /// This determines the rate at which the field is compacted.
-  ///
-  /// Currently, there are only two storage classes:
-  /// 
-  /// "" - The default storage class.
-  /// "blob" - The field is compacted into fewer rows per fragment.
-  ///
-  /// Fields that have non-default storage classes are stored in different
-  /// datasets (e.g. blob fields are stored in the nested "_blobs" dataset)
-  string storage_class = 11;
+  // Field number 11 was previously `string storage_class`.
+  // Keep it reserved so older manifests remain compatible while new writers
+  // avoid reusing the slot.
+  reserved 11;
+  reserved "storage_class";
 
   bool unenforced_primary_key = 12;
 }

protos/table.proto

@@ -183,12 +183,9 @@ message Manifest {
   // data itself and is attached to the output schema of scans.
   map<string, string> table_metadata = 19;
 
-  // The version of the blob dataset associated with this table.  Changes to
-  // blob fields will modify the blob dataset and update this version in the parent
-  // table.
-  //
-  // If this value is 0 then there are no blob fields.
-  uint64 blob_dataset_version = 17;
+  // Field number 17 (`blob_dataset_version`) was used for a secondary blob dataset.
+  reserved 17;
+  reserved "blob_dataset_version";
 
   // The base paths of data files.
   //

protos/transaction.proto

@@ -298,9 +298,7 @@ message Transaction {
     UpdateBases update_bases = 114;
   }
 
-  // An operation to apply to the blob dataset
-  oneof blob_operation {
-    Append blob_append = 200;
-    Overwrite blob_overwrite = 202;
-  }  
+  // Fields 200/202 (`blob_append` / `blob_overwrite`) previously represented blob dataset ops.
+  reserved 200, 202;
+  reserved "blob_append", "blob_overwrite";
 }