ci: auto release java

.github/workflows/bump-version/action.yml

@@ -2,25 +2,61 @@ name: Bump version for releases
 description: "Call bumpversion"
 inputs:
   part:
-    description: 'What kind of release is this?'
+    description: "What kind of release is this?"
     required: true
-    default: 'release'
+    default: "release"
 runs:
   using: "composite"
   steps:
-  - name: Set git configs for bumpversion
-    shell: bash
-    run: |
-      git config user.name 'Lance Release'
-      git config user.email '[email protected]'
-  - name: Create release version and tags
-    working-directory: python
-    shell: bash
-    run: |
-      cargo install cargo-bump
-      cargo bump ${{ inputs.part }}
-  - name: Synchronize rust version
-    shell: bash
-    run: |
-      cargo install cargo-workspaces --version 0.2.44
-      cargo ws version --no-git-commit -y --exact --force 'lance*' ${{ inputs.part }}
+    - name: Set git configs for bumpversion
+      shell: bash
+      run: |
+        git config user.name 'Lance Release'
+        git config user.email '[email protected]'
+    - name: Create release version and tags
+      working-directory: python
+      shell: bash
+      run: |
+        cargo install cargo-bump
+        cargo bump ${{ inputs.part }}
+    - name: Synchronize rust version
+      shell: bash
+      run: |
+        cargo install cargo-workspaces --version 0.2.44
+        cargo ws version --no-git-commit -y --exact --force 'lance*' ${{ inputs.part }}
+    - name: Bump java version
+      working-directory: java
+      shell: bash
+      run: |
+        # Get current version
+        current_version=$(mvn help:evaluate -Dexpression=project.version -q -DforceStdout)
+        current_version=${current_version%\%}
+
+        # Split the version into components using parameter expansion
+        major=${current_version%%.*}
+        minor=${current_version#*.}
+        minor=${minor%%.*}
+        patch=${current_version##*.}
+
+        case "${{ inputs.part }}" in
+          patch)
+            patch=$((patch + 1))
+            ;;
+          minor)
+            minor=$((minor + 1))
+            patch=0
+            ;;
+          major)
+            major=$((major + 1))
+            minor=0
+            patch=0
+            ;;
+          *)
+            echo "Invalid part specified"
+            exit 1
+            ;;
+        esac
+
+        new_version="${major}.${minor}.${patch}"
+
+        mvn versions:set versions:commit -DnewVersion=$new_version

.github/workflows/java-publish.yml

@@ -1,6 +1,8 @@
 name: Build and publish Java packages
 on:
   release:
+    # Use released instead of published, since we don't publish preview/beta
+    # versions
     types: [released]
   pull_request:
     paths:
@@ -99,7 +101,12 @@ jobs:
         run: |
           git config --global user.email "Lance Github Runner"
           git config --global user.name "[email protected]"
+      - name: Dry run
+        if: github.event_name == 'pull_request'
+        run: |
+          mvn --batch-mode -DskipTests package
       - name: Publish with Java 8
+        if: github.event_name == 'release'
         run: |
           echo "use-agent" >> ~/.gnupg/gpg.conf
           echo "pinentry-mode loopback" >> ~/.gnupg/gpg.conf

.github/workflows/java.yml

@@ -93,4 +93,4 @@ jobs:
             -Djdk.reflect.useDirectMethodHandle=false \
             -Dio.netty.tryReflectionSetAccessible=true"
           fi
-          mvn clean test
+          mvn clean install

java/core/src/main/java/com/lancedb/lance/Dataset.java

@@ -237,6 +237,8 @@ public LanceScanner newScan(ScanOptions options) {
 
   /**
    * Gets the currently checked out version of the dataset.
+   *
+   * @return the version of the dataset
    */
   public long version() {
     try (LockManager.ReadLock readLock = lockManager.acquireReadLock()) {
@@ -248,7 +250,7 @@ public long version() {
   private native long nativeVersion();
 
   /**
-   * Gets the latest version of the dataset.
+   * @return the latest version of the dataset.
    */
   public long latestVersion() {
     try (LockManager.ReadLock readLock = lockManager.acquireReadLock()) {

java/core/src/main/java/com/lancedb/lance/DatasetFragment.java

@@ -78,7 +78,9 @@ public int getId() {
     return metadata.getId();
   }
 
-  /** Count rows in this Fragment. */
+  /**
+   * @return row counts in this Fragment
+   */
   public int countRows() {
     return countRowsNative(dataset, metadata.getId());
   }

java/core/src/main/java/com/lancedb/lance/Fragment.java

@@ -29,7 +29,16 @@ public class Fragment {
     JniLoader.ensureLoaded();
   }
 
-  /** Create a fragment from the given data in vector schema root. */
+  /**
+   * Create a fragment from the given data.
+   *
+   * @param datasetUri the dataset uri
+   * @param allocator the buffer allocator
+   * @param root the vector schema root
+   * @param fragmentId the fragment id
+   * @param params the write params
+   * @return the fragment metadata
+   */
   public static FragmentMetadata create(String datasetUri, BufferAllocator allocator,
       VectorSchemaRoot root, Optional<Integer> fragmentId, WriteParams params) {
     Preconditions.checkNotNull(datasetUri);
@@ -46,7 +55,14 @@ public static FragmentMetadata create(String datasetUri, BufferAllocator allocat
     }
   }
 
-  /** Create a fragment from the given data. */
+  /**
+   * Create a fragment from the given arrow stream.
+   * @param datasetUri  the dataset uri
+   * @param stream  the arrow stream
+   * @param fragmentId  the fragment id
+   * @param params  the write params
+   * @return  the fragment metadata
+   */
   public static FragmentMetadata create(String datasetUri, ArrowArrayStream stream,
       Optional<Integer> fragmentId, WriteParams params) {
     Preconditions.checkNotNull(datasetUri);

java/core/src/main/java/com/lancedb/lance/ReadOptions.java

@@ -70,6 +70,9 @@ public static class Builder {
     /**
      * Set the version of the dataset to read.
      * If not set, read from latest version.
+     *
+     * @param version the version of the dataset
+     * @return this builder
      */
     public Builder setVersion(int version) {
       this.version = Optional.of(version);
@@ -80,6 +83,9 @@ public Builder setVersion(int version) {
      * Block size in bytes.
      * Provide a hint for the size of the minimal I/O request.
      * Recommended to be set to value bigger than 4KB.
+     *
+     * @param blockSize the block size in bytes
+     * @return this builder
      */
     public Builder setBlockSize(int blockSize) {
       this.blockSize = Optional.of(blockSize);
@@ -97,6 +103,9 @@ public Builder setBlockSize(int blockSize) {
      * / number of IVF partitions.
      * pq = number of PQ sub-vectors.
      * Default is 256.
+     *
+     * @param indexCacheSize the index cache size
+     * @return this builder
      */
     public Builder setIndexCacheSize(int indexCacheSize) {
       this.indexCacheSize = indexCacheSize;
@@ -107,6 +116,9 @@ public Builder setIndexCacheSize(int indexCacheSize) {
      * Metadata cache size for the fragment metadata. If it is zero, metadata
      * cache is disabled.
      * Default is 256.
+     *
+     * @param metadataCacheSize the metadata cache size
+     * @return this builder
      */
     public Builder setMetadataCacheSize(int metadataCacheSize) {
       this.metadataCacheSize = metadataCacheSize;

java/core/src/main/java/com/lancedb/lance/Utils.java

@@ -25,14 +25,23 @@
 /** Utility. */
 public class Utils {
 
-  /** Convert Arrow Schema to FFI Arrow Schema. */
+  /**
+   * Convert schema to ArrowSchema for JNI processing.
+   * @param schema schema
+   * @param allocator buffer allocator
+   * @return ArrowSchema
+   */
   public static ArrowSchema toFfi(Schema schema, BufferAllocator allocator) {
     ArrowSchema arrowSchema = ArrowSchema.allocateNew(allocator);
     Data.exportSchema(allocator, schema, null, arrowSchema);
     return arrowSchema;
   }
 
-  /** Convert optional array to optional list for JNI processing. */
+  /**
+   * Convert optional array to optional list for JNI processing.
+   * @param optionalArray  Optional array
+   * @return Optional list
+   */
   public static Optional<List<String>> convert(Optional<String[]> optionalArray) {
     if (optionalArray.isPresent()) {
       return Optional.of(Arrays.asList(optionalArray.get()));

java/core/src/main/java/com/lancedb/lance/WriteParams.java

@@ -57,7 +57,10 @@ public Optional<Long> getMaxBytesPerFile() {
     return maxBytesPerFile;
   }
 
-  /** Get Mode with name. */
+  /**
+   * Get Mode with name.
+   * @return mode
+   */
   public Optional<String> getMode() {
     return mode.map(Enum::name);
   }

java/core/src/main/java/com/lancedb/lance/index/IndexParams.java

@@ -39,6 +39,9 @@ public Builder() {}
     /**
      * Set the distance type for calculating the distance between vectors.
      * Default to L2.
+     *
+     * @param distanceType distance type
+     * @return this builder
      */
     public Builder setDistanceType(DistanceType distanceType) {
       this.distanceType = distanceType;
@@ -47,6 +50,8 @@ public Builder setDistanceType(DistanceType distanceType) {
 
     /**
      * Vector index parameters for creating a vector index.
+     * @param vectorIndexParams vector index parameters
+     * @return this builder
      */
     public Builder setVectorIndexParams(VectorIndexParams vectorIndexParams) {
       this.vectorIndexParams = Optional.of(vectorIndexParams);

java/core/src/main/java/com/lancedb/lance/index/vector/IvfBuildParams.java

@@ -115,7 +115,7 @@ public Builder setShufflePartitionBatches(int shufflePartitionBatches) {
      * By making this value smaller, this shuffle will consume less memory
      * but will take longer to complete, and vice versa.
      *
-     * @param shufflePartitionConcurrency
+     * @param shufflePartitionConcurrency the number of shuffle partitions to process concurrently
      * @return Builder
      */
     public Builder setShufflePartitionConcurrency(int shufflePartitionConcurrency) {
@@ -125,7 +125,7 @@ public Builder setShufflePartitionConcurrency(int shufflePartitionConcurrency) {
 
     /**
      * Set whether to use residual for k-means clustering. Default value is true.
-     * @param useResidual
+     * @param useResidual whether to use residual for k-means clustering
      * @return Builder
      */
     public Builder setUseResidual(boolean useResidual) {

java/core/src/main/java/com/lancedb/lance/index/vector/PQBuildParams.java

@@ -58,7 +58,7 @@ public Builder() {
      * The number of subvectors to divide the source vectors into.
      * This must be a divisor of the vector dimension.
      *
-     * @param numSubVectors
+     * @param numSubVectors the number of subvectors
      * @return Builder
      */
     public Builder setNumSubVectors(int numSubVectors) {
@@ -95,7 +95,7 @@ public Builder setKmeansRedos(int kmeansRedos) {
 
     /**
      * @param sampleRate sample rate to train PQ codebook
-     * @return
+     * @return Builder
      */
     public Builder setSampleRate(int sampleRate) {
       this.sampleRate = sampleRate;

java/core/src/main/java/com/lancedb/lance/index/vector/SQBuildParams.java

@@ -46,7 +46,7 @@ public Builder setNumBits(short numBits) {
 
     /**
      * @param sampleRate sample rate for training
-     * @return
+     * @return Builder
      */
     public Builder setSampleRate(int sampleRate) {
       this.sampleRate = sampleRate;

java/core/src/main/java/com/lancedb/lance/index/vector/VectorIndexParams.java

@@ -54,6 +54,7 @@ private void validate() {
    *
    * @param numPartitions the number of partitions of IVF (Inverted File Index)
    * @param distanceType the distance type for calculating the distance between vectors
+   * @return the VectorIndexParams
    */
   public static VectorIndexParams ivfFlat(int numPartitions, DistanceType distanceType) {
     return new Builder(new IvfBuildParams.Builder().setNumPartitions(numPartitions).build())
@@ -92,8 +93,8 @@ public static VectorIndexParams ivfPq(int numPartitions, int numBits, int numSub
    * Create a new IVF index with PQ quantizer.
    *
    * @param distanceType the distance type for calculating the distance between vectors
-   * @param IvfBuildParams the IVF build parameters
-   * @param PQBuildParams the PQ build parameters
+   * @param ivf the IVF build parameters
+   * @param pq the PQ build parameters
    * @return the VectorIndexParams
    */
   public static VectorIndexParams withIvfPqParams(DistanceType distanceType,
@@ -110,9 +111,9 @@ public static VectorIndexParams withIvfPqParams(DistanceType distanceType,
    * The dataset is partitioned into IVF partitions, and each partition builds an HNSW graph.
    *
    * @param distanceType the distance type for calculating the distance between vectors
-   * @param IvfBuildParams the IVF build parameters
-   * @param hnswBuildParams the HNSW build parameters
-   * @param PQBuildParams the PQ build parameters
+   * @param ivf the IVF build parameters
+   * @param hnsw the HNSW build parameters
+   * @param pq the PQ build parameters
    * @return the VectorIndexParams
    */
   public static VectorIndexParams withIvfHnswPqParams(DistanceType distanceType,
@@ -131,9 +132,9 @@ public static VectorIndexParams withIvfHnswPqParams(DistanceType distanceType,
    * The dataset is partitioned into IVF partitions, and each partition builds an HNSW graph.
    *
    * @param distanceType the distance type for calculating the distance between vectors
-   * @param IvfBuildParams the IVF build parameters
-   * @param hnswBuildParams the HNSW build parameters
-   * @param SQBuildParams the SQ build parameters
+   * @param ivf the IVF build parameters
+   * @param hnsw the HNSW build parameters
+   * @param sq the SQ build parameters
    * @return the VectorIndexParams
    */
   public static VectorIndexParams withIvfHnswSqParams(DistanceType distanceType,
@@ -173,7 +174,7 @@ public Builder setDistanceType(DistanceType distanceType) {
     }
 
     /**
-     * @param sqParams the SQ quantizer build parameters
+     * @param pqParams the PQ quantizer build parameters
      * @return Builder
      */
     public Builder setPqParams(PQBuildParams pqParams) {