nl.topicus.jdbc.shaded.io.netty.buffer.package-info Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of spanner-jdbc Show documentation
JDBC Driver for Google Cloud Spanner
There is a newer version: 1.1.6
/*
 * Copyright 2012 The Netty Project
 *
 * The Netty Project licenses this file to you under the Apache License,
 * version 2.0 (the "License"); you may not use this file except in nl.topicus.jdbc.shaded.com.liance
 * with the License. You may obtain a copy of the License at:
 *
 *   http://www.apache.nl.topicus.jdbc.shaded.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.
 */

/**
 * Abstraction of a byte buffer - the fundamental data structure
 * to represent a low-level binary and text message.
 *
 * Netty uses its own buffer API instead of NIO {@link java.nio.ByteBuffer} to
 * represent a sequence of bytes. This approach has significant advantage over
 * using {@link java.nio.ByteBuffer}.  Netty's new buffer type,
 * {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf}, has been designed from ground
 * up to address the problems of {@link java.nio.ByteBuffer} and to meet the
 * daily needs of nl.topicus.jdbc.shaded.net.ork application developers.  To list a few cool features:
 * 
 *   You can define your buffer type if necessary.
 *   Transparent zero copy is achieved by built-in nl.topicus.jdbc.shaded.com.osite buffer type.
 *   A dynamic buffer type is provided out-of-the-box, whose capacity is
 *       expanded on demand, just like {@link java.lang.StringBuffer}.
 *   There's no need to call the {@code flip()} method anymore.
 *   It is often faster than {@link java.nio.ByteBuffer}.
 * 
 *
 * Extensibility
 *
 * {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf} has rich set of operations
 * optimized for rapid protocol implementation.  For example,
 * {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf} provides various operations
 * for accessing unsigned values and strings and searching for certain byte
 * sequence in a buffer.  You can also extend or wrap existing buffer type
 * to add convenient accessors.  The custom buffer type still implements
 * {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf} interface rather than
 * introducing an incompatible type.
 *
 * Transparent Zero Copy
 *
 * To lift up the performance of a nl.topicus.jdbc.shaded.net.ork application to the extreme, you need
 * to reduce the number of memory copy operation.  You might have a set of
 * buffers that could be sliced and nl.topicus.jdbc.shaded.com.ined to nl.topicus.jdbc.shaded.com.ose a whole message.  Netty
 * provides a nl.topicus.jdbc.shaded.com.osite buffer which allows you to create a new buffer from the
 * arbitrary number of existing buffers with no memory copy.  For example, a
 * message could be nl.topicus.jdbc.shaded.com.osed of two parts; header and body.  In a modularized
 * application, the two parts could be produced by different modules and
 * assembled later when the message is sent out.
 *  * +--------+----------+
 * | header |   body   |
 * +--------+----------+
 * 
 * If {@link java.nio.ByteBuffer} were used, you would have to create a new big
 * buffer and copy the two parts into the new buffer.   Alternatively, you can
 * perform a gathering write operation in NIO, but it restricts you to represent
 * the nl.topicus.jdbc.shaded.com.osite of buffers as an array of {@link java.nio.ByteBuffer}s rather
 * than a single buffer, breaking the abstraction and introducing nl.topicus.jdbc.shaded.com.licated
 * state management.  Moreover, it's of no use if you are not going to read or
 * write from an NIO channel.
 *  * // The nl.topicus.jdbc.shaded.com.osite type is incompatible with the nl.topicus.jdbc.shaded.com.onent type.
 * ByteBuffer[] message = new ByteBuffer[] { header, body };
 * 
 * By contrast, {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf} does not have such
 * caveats because it is fully extensible and has a built-in nl.topicus.jdbc.shaded.com.osite buffer
 * type.
 *  * // The nl.topicus.jdbc.shaded.com.osite type is nl.topicus.jdbc.shaded.com.atible with the nl.topicus.jdbc.shaded.com.onent type.
 * {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf} message = {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.Unpooled}.wrappedBuffer(header, body);
 *
 * // Therefore, you can even create a nl.topicus.jdbc.shaded.com.osite by mixing a nl.topicus.jdbc.shaded.com.osite and an
 * // ordinary buffer.
 * {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf} messageWithFooter = {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.Unpooled}.wrappedBuffer(message, footer);
 *
 * // Because the nl.topicus.jdbc.shaded.com.osite is still a {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf}, you can access its content
 * // easily, and the accessor method will behave just like it's a single buffer
 * // even if the region you want to access spans over multiple nl.topicus.jdbc.shaded.com.onents.  The
 * // unsigned integer being read here is located across body and footer.
 * messageWithFooter.getUnsignedInt(
 *     messageWithFooter.readableBytes() - footer.readableBytes() - 1);
 * 
 *
 * Automatic Capacity Extension
 *
 * Many protocols define variable length messages, which means there's no way to
 * determine the length of a message until you construct the message or it is
 * difficult and inconvenient to calculate the length precisely.  It is just
 * like when you build a {@link java.lang.String}. You often estimate the length
 * of the resulting string and let {@link java.lang.StringBuffer} expand itself
 * on demand.
 *  * // A new dynamic buffer is created.  Internally, the actual buffer is created
 * // lazily to avoid potentially wasted memory space.
 * {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf} b = {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.Unpooled}.buffer(4);
 *
 * // When the first write attempt is made, the internal buffer is created with
 * // the specified initial capacity (4).
 * b.writeByte('1');
 *
 * b.writeByte('2');
 * b.writeByte('3');
 * b.writeByte('4');
 *
 * // When the number of written bytes exceeds the initial capacity (4), the
 * // internal buffer is reallocated automatically with a larger capacity.
 * b.writeByte('5');
 * 
 *
 * Better Performance
 *
 * Most frequently used buffer implementation of
 * {@link nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer.ByteBuf} is a very thin wrapper of a
 * byte array (i.e. {@code byte[]}).  Unlike {@link java.nio.ByteBuffer}, it has
 * no nl.topicus.jdbc.shaded.com.licated boundary check and index nl.topicus.jdbc.shaded.com.ensation, and therefore it is
 * easier for a JVM to optimize the buffer access.  More nl.topicus.jdbc.shaded.com.licated buffer
 * implementation is used only for sliced or nl.topicus.jdbc.shaded.com.osite buffers, and it performs
 * as well as {@link java.nio.ByteBuffer}.
 */
package nl.topicus.jdbc.shaded.io.nl.topicus.jdbc.shaded.net.y.buffer;