BaseNCodecInputStream

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *      https://www.apache.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.
 */

package org.apache.commons.codec.binary;

import static org.apache.commons.codec.binary.BaseNCodec.EOF;

import java.io.ByteArrayInputStream;
import java.io.FilterInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.util.Objects;

import org.apache.commons.codec.binary.BaseNCodec.Context;

/**
 * Abstracts Base-N input streams.
 *
 * @param <C> A BaseNCodec subclass.
 * @param <T> A BaseNCodecInputStream subclass.
 * @param <B> A subclass.
 * @see Base16InputStream
 * @see Base32InputStream
 * @see Base58InputStream
 * @see Base64InputStream
 * @since 1.5
 */
public class BaseNCodecInputStream<C extends BaseNCodec, T extends BaseNCodecInputStream<C, T, B>, B extends BaseNCodecInputStream.AbstracBuilder<T, C, B>>
        extends FilterInputStream {

    /**
     * Builds input stream instances in {@link BaseNCodec} format.
     *
     * @param <T> The input stream type to build.
     * @param <C> A {@link BaseNCodec} subclass.
     * @param <B> The builder subclass.
     * @since 1.20.0
     */
    public abstract static class AbstracBuilder<T, C extends BaseNCodec, B extends AbstractBaseNCodecStreamBuilder<T, C, B>>
        extends AbstractBaseNCodecStreamBuilder<T, C, B> {

        private InputStream inputStream;

        /**
         * Constructs a new instance.
         */
        public AbstracBuilder() {
            // super
        }

        /**
         * Gets the input stream.
         *
         * @return The input stream.
         */
        protected InputStream getInputStream() {
            return inputStream;
        }

        /**
         * Sets the input bytes.
         *
         * @param inputBytes The input bytes.
         * @return {@code this} instance.
         * @since 1.22.0
         */
        public B setByteArray(final byte[] inputBytes) {
            return setInputStream(inputBytes == null ? null : new ByteArrayInputStream(inputBytes));
        }

        /**
         * Sets the input stream.
         *
         * @param inputStream The input stream.
         * @return {@code this} instance.
         */
        public B setInputStream(final InputStream inputStream) {
            this.inputStream = inputStream;
            return asThis();
        }
    }

    private final C baseNCodec;
    private final boolean doEncode;
    private final byte[] singleByte = new byte[1];
    private final byte[] buf;
    private final Context context = new Context();

    /**
     * Constructs a new instance.
     *
     * @param builder A builder.
     * @since 1.20.0
     */
    @SuppressWarnings("resource") // Caller closes.
    protected BaseNCodecInputStream(final AbstracBuilder<T, C, B> builder) {
        super(builder.getInputStream());
        this.baseNCodec = builder.getBaseNCodec();
        this.doEncode = builder.getEncode();
        this.buf = new byte[doEncode ? 4096 : 8192];
    }

    /**
     * Constructs a new instance.
     *
     * @param inputStream The input stream.
     * @param baseNCodec  The codec.
     * @param doEncode    set to true to perform encoding, else decoding.
     */
    protected BaseNCodecInputStream(final InputStream inputStream, final C baseNCodec, final boolean doEncode) {
        super(inputStream);
        this.doEncode = doEncode;
        this.baseNCodec = baseNCodec;
        this.buf = new byte[doEncode ? 4096 : 8192];
    }

    /**
     * {@inheritDoc}
     *
     * @return {@code 0} if the {@link InputStream} has reached {@code EOF}, {@code 1} otherwise.
     * @throws IOException Thrown if an I/O error occurs.
     * @since 1.7
     */
    @Override
    public int available() throws IOException {
        // Note: The logic is similar to the InflaterInputStream:
        // as long as we have not reached EOF, indicate that there is more
        // data available. As we do not know for sure how much data is left,
        // just return 1 as a safe guess.
        return context.eof ? 0 : 1;
    }

    /**
     * Tests whether decoding behavior is strict.
     *
     * <p>
     * Strict decoding rejects invalid trailing bits and, for Base32 and Base64, noncanonical input. Decoding errors are reported as {@link IOException}.
     * To complete validation, consume this stream to EOF. Decoded bytes can be emitted before a later validation error.
     * </p>
     *
     * @return true if using strict decoding.
     * @since 1.15
     */
    public boolean isStrictDecoding() {
        return baseNCodec.isStrictDecoding();
    }

    /**
     * Marks the current position in this input stream.
     * <p>
     * The {@link #mark} method of {@link BaseNCodecInputStream} does nothing.
     * </p>
     *
     * @param readLimit The maximum limit of bytes that can be read before the mark position becomes invalid.
     * @see #markSupported()
     * @since 1.7
     */
    @Override
    public synchronized void mark(final int readLimit) {
        // noop
    }

    /**
     * {@inheritDoc}
     *
     * @return Always returns {@code false}.
     */
    @Override
    public boolean markSupported() {
        return false; // not an easy job to support marks
    }

    /**
     * Reads one {@code byte} from this input stream.
     *
     * @return The byte as an integer in the range 0 to 255. Returns -1 if EOF has been reached.
     * @throws IOException Thrown if an I/O error occurs.
     */
    @Override
    public int read() throws IOException {
        int r = read(singleByte, 0, 1);
        while (r == 0) {
            r = read(singleByte, 0, 1);
        }
        if (r > 0) {
            final byte b = singleByte[0];
            return b < 0 ? 256 + b : b;
        }
        return EOF;
    }

    /**
     * Attempts to read {@code len} bytes into the specified {@code b} array starting at {@code offset} from this InputStream.
     *
     * @param array  destination byte array.
     * @param offset where to start writing the bytes.
     * @param len    maximum number of bytes to read.
     * @return number of bytes read.
     * @throws IOException               Thrown if an I/O error occurs.
     * @throws NullPointerException      Thrown if the byte array parameter is null.
     * @throws IndexOutOfBoundsException Thrown if the offset, length, or buffer size is invalid.
     */
    @Override
    public int read(final byte[] array, final int offset, final int len) throws IOException {
        Objects.requireNonNull(array, "array");
        if (offset < 0 || len < 0 || offset > array.length || offset + len > array.length) {
            throw new IndexOutOfBoundsException();
        }
        if (len == 0) {
            return 0;
        }
        int readLen = 0;
        /*
         * Rationale for while-loop on (readLen == 0): ----- Base32.readResults() usually returns > 0 or EOF (-1). In the rare case where it returns 0, we just
         * keep trying.
         *
         * This is essentially an undocumented contract for InputStream implementers that want their code to work properly with java.io.InputStreamReader, since
         * the latter hates it when InputStream.read(byte[]) returns a zero. Unfortunately our readResults() call must return 0 if a large amount of the data
         * being decoded was non-base32, so this while-loop enables proper interop with InputStreamReader for that scenario. ----- This is a fix for CODEC-101
         */
        // Attempt to read the request length
        while (readLen < len) {
            if (!baseNCodec.hasData(context)) {
                // Obtain more data.
                // buf is reused across calls to read to avoid repeated allocations
                BaseNCodec.code(doEncode, baseNCodec, buf, 0, in.read(buf), context);
            }
            final int read = baseNCodec.readResults(array, offset + readLen, len - readLen, context);
            if (read < 0) {
                // Return the amount read or EOF
                return readLen != 0 ? readLen : -1;
            }
            readLen += read;
        }
        return readLen;
    }

    /**
     * Always throws {@link IOException} because this stream does not support resetting.
     *
     * @throws IOException Thrown if this method is invoked.
     * @since 1.7
     */
    @Override
    public synchronized void reset() throws IOException {
        throw new IOException("mark/reset not supported");
    }

    /**
     * {@inheritDoc}
     *
     * @throws IllegalArgumentException Thrown if the provided skip length is negative.
     * @throws IOException Thrown if an I/O error occurs.
     * @since 1.7
     */
    @Override
    public long skip(final long n) throws IOException {
        if (n < 0) {
            throw new IllegalArgumentException("Negative skip length: " + n);
        }
        // skip in chunks of 512 bytes
        final byte[] b = new byte[512];
        long todo = n;
        while (todo > 0) {
            int len = (int) Math.min(b.length, todo);
            len = this.read(b, 0, len);
            if (len == EOF) {
                break;
            }
            todo -= len;
        }
        return n - todo;
    }
}