# ADR-0002 - Avoid using Java serialization

## Date

2012-12-01

## Context

In Gradle we often need to serialize in-memory objects for caching, or to transmit them across process barriers, etc.
Java serialization is one way to implement this, however, despite its simplicity of implementation, it has several drawbacks:

- **Performance:**
Java's built-in serialization mechanism is often slower compared to other serialization solutions.

  public abstract HashCode hashBytes(byte[] input, int off, int len);

  @Override
  public HashCode hashBytes(ByteBuffer input) {
    return newHasher(input.remaining()).putBytes(input).hash();
  }

  /** In-memory stream-based implementation of Hasher. */
  private final class BufferingHasher extends AbstractHasher {
    final ExposedByteArrayOutputStream stream;

    BufferingHasher(int expectedInputSize) {

  /**
   * Flush a {@link Flushable}, with control over whether an {@code IOException} may be thrown.
   *
   * <p>If {@code swallowIOException} is true, then we don't rethrow {@code IOException}, but merely
   * log it.
   *
   * @param flushable the {@code Flushable} object to be flushed.
   * @param swallowIOException if true, don't propagate IO exceptions thrown by the {@code flush}
   *     method

Each <code>DATA</code> directive initializes a section of the
corresponding memory.
The memory not explicitly initialized is zeroed.
The general form of the <code>DATA</code> directive is

<pre>
DATA	symbol+offset(SB)/width, value
</pre>

<p>
which initializes the symbol memory at the given offset and width with the given value.

    return byteSink.wasStreamClosed();
  }

  @Override
  public Writer openStream() throws IOException {
    // using TestByteSink's output stream to get option behavior, so flush to it on every write
    return new FilterWriter(new OutputStreamWriter(byteSink.openStream(), UTF_8)) {
      @Override
      public void write(int c) throws IOException {
        super.write(c);
        flush();
      }

//     in `input_shapes`.
//
// The results are returned in `output_shapes` and
// `output_resource_shapes_and_types`. The caller is responsible for freeing the
// memory in these buffers by calling `TF_DeleteShapeAndTypeList`.
TF_CAPI_EXPORT extern void TFE_InferShapes(
    TFE_Op* op, TF_ShapeAndTypeList* input_shapes, TF_Tensor** input_tensors,
    TF_ShapeAndTypeList* input_tensor_as_shapes,

import static org.apache.maven.api.services.BaseRequest.nonNull;

/**
 * Provides access to the contents of a source independently of the
 * backing store (e.g. file system, database, memory).
 * <p>
 * This is mainly used to parse files into objects such as
 * {@link org.apache.maven.api.Project},
 * {@link org.apache.maven.api.model.Model},
 * {@link org.apache.maven.api.settings.Settings}, or

Code Contributions
------------------

Get working code on a personal branch with tests passing before you submit a PR:

```
./gradlew clean check
```

Please make every effort to follow existing conventions and style in order to keep the code as
readable as possible.

Contribute code changes through GitHub by forking the repository and sending a pull request. We
squash all pull requests on merge.

Code Contributions
------------------

Get working code on a personal branch with tests passing before you submit a PR:

```
./gradlew clean check
```

Please make every effort to follow existing conventions and style in order to keep the code as
readable as possible.

Contribute code changes through GitHub by forking the repository and sending a pull request. We
squash all pull requests on merge.

    - [Quality of Service for memory resources](#quality-of-service-for-memory-resources)
    - [API changes and improvements for ephemeral containers](#api-changes-and-improvements-for-ephemeral-containers)
  - [Known Issues](#known-issues)