/* * Copyright (c) 2012, 2016, Oracle and/or its affiliates. All rights reserved. * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This code is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License version 2 only, as * published by the Free Software Foundation. Oracle designates this * particular file as subject to the "Classpath" exception as provided * by Oracle in the LICENSE file that accompanied this code. * * This code is distributed in the hope that it will be useful, but WITHOUT * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License * version 2 for more details (a copy is included in the LICENSE file that * accompanied this code). * * You should have received a copy of the GNU General Public License version * 2 along with this work; if not, write to the Free Software Foundation, * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. * * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA * or visit www.oracle.com if you need additional information or have any * questions. */ package java.util; import java.io.FilterOutputStream; import java.io.InputStream; import java.io.IOException; import java.io.OutputStream; import java.nio.ByteBuffer; import java.nio.charset.StandardCharsets; /** * This class consists exclusively of static methods for obtaining * encoders and decoders for the Base64 encoding scheme. The * implementation of this class supports the following types of Base64 * as specified in * RFC 4648 and * RFC 2045. * * * *

Unless otherwise noted, passing a {@code null} argument to a * method of this class will cause a {@link java.lang.NullPointerException * NullPointerException} to be thrown. * * @author Xueming Shen * @since 1.8 */ public class Base64 { private Base64() {} /** * Returns a {@link Encoder} that encodes using the * Basic type base64 encoding scheme. * * @return A Base64 encoder. */ public static Encoder getEncoder() { return Encoder.RFC4648; } /** * Returns a {@link Encoder} that encodes using the * URL and Filename safe type base64 * encoding scheme. * * @return A Base64 encoder. */ public static Encoder getUrlEncoder() { return Encoder.RFC4648_URLSAFE; } /** * Returns a {@link Encoder} that encodes using the * MIME type base64 encoding scheme. * * @return A Base64 encoder. */ public static Encoder getMimeEncoder() { return Encoder.RFC2045; } /** * Returns a {@link Encoder} that encodes using the * MIME type base64 encoding scheme * with specified line length and line separators. * * @param lineLength * the length of each output line (rounded down to nearest multiple * of 4). If {@code lineLength <= 0} the output will not be separated * in lines * @param lineSeparator * the line separator for each output line * * @return A Base64 encoder. * * @throws IllegalArgumentException if {@code lineSeparator} includes any * character of "The Base64 Alphabet" as specified in Table 1 of * RFC 2045. */ public static Encoder getMimeEncoder(int lineLength, byte[] lineSeparator) { Objects.requireNonNull(lineSeparator); int[] base64 = Decoder.fromBase64; for (byte b : lineSeparator) { if (base64[b & 0xff] != -1) throw new IllegalArgumentException( "Illegal base64 line separator character 0x" + Integer.toString(b, 16)); } if (lineLength <= 0) { return Encoder.RFC4648; } return new Encoder(false, lineSeparator, lineLength >> 2 << 2, true); } /** * Returns a {@link Decoder} that decodes using the * Basic type base64 encoding scheme. * * @return A Base64 decoder. */ public static Decoder getDecoder() { return Decoder.RFC4648; } /** * Returns a {@link Decoder} that decodes using the * URL and Filename safe type base64 * encoding scheme. * * @return A Base64 decoder. */ public static Decoder getUrlDecoder() { return Decoder.RFC4648_URLSAFE; } /** * Returns a {@link Decoder} that decodes using the * MIME type base64 decoding scheme. * * @return A Base64 decoder. */ public static Decoder getMimeDecoder() { return Decoder.RFC2045; } /** * This class implements an encoder for encoding byte data using * the Base64 encoding scheme as specified in RFC 4648 and RFC 2045. * *

Instances of {@link Encoder} class are safe for use by * multiple concurrent threads. * *

Unless otherwise noted, passing a {@code null} argument to * a method of this class will cause a * {@link java.lang.NullPointerException NullPointerException} to * be thrown. * * @see Decoder * @since 1.8 */ public static class Encoder { private final byte[] newline; private final int linemax; private final boolean isURL; private final boolean doPadding; private Encoder(boolean isURL, byte[] newline, int linemax, boolean doPadding) { this.isURL = isURL; this.newline = newline; this.linemax = linemax; this.doPadding = doPadding; } /** * This array is a lookup table that translates 6-bit positive integer * index values into their "Base64 Alphabet" equivalents as specified * in "Table 1: The Base64 Alphabet" of RFC 2045 (and RFC 4648). */ private static final char[] toBase64 = { 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', '/' }; /** * It's the lookup table for "URL and Filename safe Base64" as specified * in Table 2 of the RFC 4648, with the '+' and '/' changed to '-' and * '_'. This table is used when BASE64_URL is specified. */ private static final char[] toBase64URL = { 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '-', '_' }; private static final int MIMELINEMAX = 76; private static final byte[] CRLF = new byte[] {'\r', '\n'}; static final Encoder RFC4648 = new Encoder(false, null, -1, true); static final Encoder RFC4648_URLSAFE = new Encoder(true, null, -1, true); static final Encoder RFC2045 = new Encoder(false, CRLF, MIMELINEMAX, true); private final int outLength(int srclen) { int len = 0; if (doPadding) { len = 4 * ((srclen + 2) / 3); } else { int n = srclen % 3; len = 4 * (srclen / 3) + (n == 0 ? 0 : n + 1); } if (linemax > 0) // line separators len += (len - 1) / linemax * newline.length; return len; } /** * Encodes all bytes from the specified byte array into a newly-allocated * byte array using the {@link Base64} encoding scheme. The returned byte * array is of the length of the resulting bytes. * * @param src * the byte array to encode * @return A newly-allocated byte array containing the resulting * encoded bytes. */ public byte[] encode(byte[] src) { int len = outLength(src.length); // dst array size byte[] dst = new byte[len]; int ret = encode0(src, 0, src.length, dst); if (ret != dst.length) return Arrays.copyOf(dst, ret); return dst; } /** * Encodes all bytes from the specified byte array using the * {@link Base64} encoding scheme, writing the resulting bytes to the * given output byte array, starting at offset 0. * *

It is the responsibility of the invoker of this method to make * sure the output byte array {@code dst} has enough space for encoding * all bytes from the input byte array. No bytes will be written to the * output byte array if the output byte array is not big enough. * * @param src * the byte array to encode * @param dst * the output byte array * @return The number of bytes written to the output byte array * * @throws IllegalArgumentException if {@code dst} does not have enough * space for encoding all input bytes. */ public int encode(byte[] src, byte[] dst) { int len = outLength(src.length); // dst array size if (dst.length < len) throw new IllegalArgumentException( "Output byte array is too small for encoding all input bytes"); return encode0(src, 0, src.length, dst); } /** * Encodes the specified byte array into a String using the {@link Base64} * encoding scheme. * *

This method first encodes all input bytes into a base64 encoded * byte array and then constructs a new String by using the encoded byte * array and the {@link java.nio.charset.StandardCharsets#ISO_8859_1 * ISO-8859-1} charset. * *

In other words, an invocation of this method has exactly the same * effect as invoking * {@code new String(encode(src), StandardCharsets.ISO_8859_1)}. * * @param src * the byte array to encode * @return A String containing the resulting Base64 encoded characters */ @SuppressWarnings("deprecation") public String encodeToString(byte[] src) { byte[] encoded = encode(src); return new String(encoded, 0, 0, encoded.length); } /** * Encodes all remaining bytes from the specified byte buffer into * a newly-allocated ByteBuffer using the {@link Base64} encoding * scheme. * * Upon return, the source buffer's position will be updated to * its limit; its limit will not have been changed. The returned * output buffer's position will be zero and its limit will be the * number of resulting encoded bytes. * * @param buffer * the source ByteBuffer to encode * @return A newly-allocated byte buffer containing the encoded bytes. */ public ByteBuffer encode(ByteBuffer buffer) { int len = outLength(buffer.remaining()); byte[] dst = new byte[len]; int ret = 0; if (buffer.hasArray()) { ret = encode0(buffer.array(), buffer.arrayOffset() + buffer.position(), buffer.arrayOffset() + buffer.limit(), dst); buffer.position(buffer.limit()); } else { byte[] src = new byte[buffer.remaining()]; buffer.get(src); ret = encode0(src, 0, src.length, dst); } if (ret != dst.length) dst = Arrays.copyOf(dst, ret); return ByteBuffer.wrap(dst); } /** * Wraps an output stream for encoding byte data using the {@link Base64} * encoding scheme. * *

It is recommended to promptly close the returned output stream after * use, during which it will flush all possible leftover bytes to the underlying * output stream. Closing the returned output stream will close the underlying * output stream. * * @param os * the output stream. * @return the output stream for encoding the byte data into the * specified Base64 encoded format */ public OutputStream wrap(OutputStream os) { Objects.requireNonNull(os); return new EncOutputStream(os, isURL ? toBase64URL : toBase64, newline, linemax, doPadding); } /** * Returns an encoder instance that encodes equivalently to this one, * but without adding any padding character at the end of the encoded * byte data. * *

The encoding scheme of this encoder instance is unaffected by * this invocation. The returned encoder instance should be used for * non-padding encoding operation. * * @return an equivalent encoder that encodes without adding any * padding character at the end */ public Encoder withoutPadding() { if (!doPadding) return this; return new Encoder(isURL, newline, linemax, false); } private int encode0(byte[] src, int off, int end, byte[] dst) { char[] base64 = isURL ? toBase64URL : toBase64; int sp = off; int slen = (end - off) / 3 * 3; int sl = off + slen; if (linemax > 0 && slen > linemax / 4 * 3) slen = linemax / 4 * 3; int dp = 0; while (sp < sl) { int sl0 = Math.min(sp + slen, sl); for (int sp0 = sp, dp0 = dp ; sp0 < sl0; ) { int bits = (src[sp0++] & 0xff) << 16 | (src[sp0++] & 0xff) << 8 | (src[sp0++] & 0xff); dst[dp0++] = (byte)base64[(bits >>> 18) & 0x3f]; dst[dp0++] = (byte)base64[(bits >>> 12) & 0x3f]; dst[dp0++] = (byte)base64[(bits >>> 6) & 0x3f]; dst[dp0++] = (byte)base64[bits & 0x3f]; } int dlen = (sl0 - sp) / 3 * 4; dp += dlen; sp = sl0; if (dlen == linemax && sp < end) { for (byte b : newline){ dst[dp++] = b; } } } if (sp < end) { // 1 or 2 leftover bytes int b0 = src[sp++] & 0xff; dst[dp++] = (byte)base64[b0 >> 2]; if (sp == end) { dst[dp++] = (byte)base64[(b0 << 4) & 0x3f]; if (doPadding) { dst[dp++] = '='; dst[dp++] = '='; } } else { int b1 = src[sp++] & 0xff; dst[dp++] = (byte)base64[(b0 << 4) & 0x3f | (b1 >> 4)]; dst[dp++] = (byte)base64[(b1 << 2) & 0x3f]; if (doPadding) { dst[dp++] = '='; } } } return dp; } } /** * This class implements a decoder for decoding byte data using the * Base64 encoding scheme as specified in RFC 4648 and RFC 2045. * *

The Base64 padding character {@code '='} is accepted and * interpreted as the end of the encoded byte data, but is not * required. So if the final unit of the encoded byte data only has * two or three Base64 characters (without the corresponding padding * character(s) padded), they are decoded as if followed by padding * character(s). If there is a padding character present in the * final unit, the correct number of padding character(s) must be * present, otherwise {@code IllegalArgumentException} ( * {@code IOException} when reading from a Base64 stream) is thrown * during decoding. * *

Instances of {@link Decoder} class are safe for use by * multiple concurrent threads. * *

Unless otherwise noted, passing a {@code null} argument to * a method of this class will cause a * {@link java.lang.NullPointerException NullPointerException} to * be thrown. * * @see Encoder * @since 1.8 */ public static class Decoder { private final boolean isURL; private final boolean isMIME; private Decoder(boolean isURL, boolean isMIME) { this.isURL = isURL; this.isMIME = isMIME; } /** * Lookup table for decoding unicode characters drawn from the * "Base64 Alphabet" (as specified in Table 1 of RFC 2045) into * their 6-bit positive integer equivalents. Characters that * are not in the Base64 alphabet but fall within the bounds of * the array are encoded to -1. * */ private static final int[] fromBase64 = new int[256]; static { Arrays.fill(fromBase64, -1); for (int i = 0; i < Encoder.toBase64.length; i++) fromBase64[Encoder.toBase64[i]] = i; fromBase64['='] = -2; } /** * Lookup table for decoding "URL and Filename safe Base64 Alphabet" * as specified in Table2 of the RFC 4648. */ private static final int[] fromBase64URL = new int[256]; static { Arrays.fill(fromBase64URL, -1); for (int i = 0; i < Encoder.toBase64URL.length; i++) fromBase64URL[Encoder.toBase64URL[i]] = i; fromBase64URL['='] = -2; } static final Decoder RFC4648 = new Decoder(false, false); static final Decoder RFC4648_URLSAFE = new Decoder(true, false); static final Decoder RFC2045 = new Decoder(false, true); /** * Decodes all bytes from the input byte array using the {@link Base64} * encoding scheme, writing the results into a newly-allocated output * byte array. The returned byte array is of the length of the resulting * bytes. * * @param src * the byte array to decode * * @return A newly-allocated byte array containing the decoded bytes. * * @throws IllegalArgumentException * if {@code src} is not in valid Base64 scheme */ public byte[] decode(byte[] src) { byte[] dst = new byte[outLength(src, 0, src.length)]; int ret = decode0(src, 0, src.length, dst); if (ret != dst.length) { dst = Arrays.copyOf(dst, ret); } return dst; } /** * Decodes a Base64 encoded String into a newly-allocated byte array * using the {@link Base64} encoding scheme. * *

An invocation of this method has exactly the same effect as invoking * {@code decode(src.getBytes(StandardCharsets.ISO_8859_1))} * * @param src * the string to decode * * @return A newly-allocated byte array containing the decoded bytes. * * @throws IllegalArgumentException * if {@code src} is not in valid Base64 scheme */ public byte[] decode(String src) { return decode(src.getBytes(StandardCharsets.ISO_8859_1)); } /** * Decodes all bytes from the input byte array using the {@link Base64} * encoding scheme, writing the results into the given output byte array, * starting at offset 0. * *

It is the responsibility of the invoker of this method to make * sure the output byte array {@code dst} has enough space for decoding * all bytes from the input byte array. No bytes will be be written to * the output byte array if the output byte array is not big enough. * *

If the input byte array is not in valid Base64 encoding scheme * then some bytes may have been written to the output byte array before * IllegalargumentException is thrown. * * @param src * the byte array to decode * @param dst * the output byte array * * @return The number of bytes written to the output byte array * * @throws IllegalArgumentException * if {@code src} is not in valid Base64 scheme, or {@code dst} * does not have enough space for decoding all input bytes. */ public int decode(byte[] src, byte[] dst) { int len = outLength(src, 0, src.length); if (dst.length < len) throw new IllegalArgumentException( "Output byte array is too small for decoding all input bytes"); return decode0(src, 0, src.length, dst); } /** * Decodes all bytes from the input byte buffer using the {@link Base64} * encoding scheme, writing the results into a newly-allocated ByteBuffer. * *

Upon return, the source buffer's position will be updated to * its limit; its limit will not have been changed. The returned * output buffer's position will be zero and its limit will be the * number of resulting decoded bytes * *

{@code IllegalArgumentException} is thrown if the input buffer * is not in valid Base64 encoding scheme. The position of the input * buffer will not be advanced in this case. * * @param buffer * the ByteBuffer to decode * * @return A newly-allocated byte buffer containing the decoded bytes * * @throws IllegalArgumentException * if {@code src} is not in valid Base64 scheme. */ public ByteBuffer decode(ByteBuffer buffer) { int pos0 = buffer.position(); try { byte[] src; int sp, sl; if (buffer.hasArray()) { src = buffer.array(); sp = buffer.arrayOffset() + buffer.position(); sl = buffer.arrayOffset() + buffer.limit(); buffer.position(buffer.limit()); } else { src = new byte[buffer.remaining()]; buffer.get(src); sp = 0; sl = src.length; } byte[] dst = new byte[outLength(src, sp, sl)]; return ByteBuffer.wrap(dst, 0, decode0(src, sp, sl, dst)); } catch (IllegalArgumentException iae) { buffer.position(pos0); throw iae; } } /** * Returns an input stream for decoding {@link Base64} encoded byte stream. * *

The {@code read} methods of the returned {@code InputStream} will * throw {@code IOException} when reading bytes that cannot be decoded. * *

Closing the returned input stream will close the underlying * input stream. * * @param is * the input stream * * @return the input stream for decoding the specified Base64 encoded * byte stream */ public InputStream wrap(InputStream is) { Objects.requireNonNull(is); return new DecInputStream(is, isURL ? fromBase64URL : fromBase64, isMIME); } private int outLength(byte[] src, int sp, int sl) { int[] base64 = isURL ? fromBase64URL : fromBase64; int paddings = 0; int len = sl - sp; if (len == 0) return 0; if (len < 2) { if (isMIME && base64[0] == -1) return 0; throw new IllegalArgumentException( "Input byte[] should at least have 2 bytes for base64 bytes"); } if (isMIME) { // scan all bytes to fill out all non-alphabet. a performance // trade-off of pre-scan or Arrays.copyOf int n = 0; while (sp < sl) { int b = src[sp++] & 0xff; if (b == '=') { len -= (sl - sp + 1); break; } if ((b = base64[b]) == -1) n++; } len -= n; } else { if (src[sl - 1] == '=') { paddings++; if (src[sl - 2] == '=') paddings++; } } if (paddings == 0 && (len & 0x3) != 0) paddings = 4 - (len & 0x3); return 3 * ((len + 3) / 4) - paddings; } private int decode0(byte[] src, int sp, int sl, byte[] dst) { int[] base64 = isURL ? fromBase64URL : fromBase64; int dp = 0; int bits = 0; int shiftto = 18; // pos of first byte of 4-byte atom while (sp < sl) { int b = src[sp++] & 0xff; if ((b = base64[b]) < 0) { if (b == -2) { // padding byte '=' // = shiftto==18 unnecessary padding // x= shiftto==12 a dangling single x // x to be handled together with non-padding case // xx= shiftto==6&&sp==sl missing last = // xx=y shiftto==6 last is not = if (shiftto == 6 && (sp == sl || src[sp++] != '=') || shiftto == 18) { throw new IllegalArgumentException( "Input byte array has wrong 4-byte ending unit"); } break; } if (isMIME) // skip if for rfc2045 continue; else throw new IllegalArgumentException( "Illegal base64 character " + Integer.toString(src[sp - 1], 16)); } bits |= (b << shiftto); shiftto -= 6; if (shiftto < 0) { dst[dp++] = (byte)(bits >> 16); dst[dp++] = (byte)(bits >> 8); dst[dp++] = (byte)(bits); shiftto = 18; bits = 0; } } // reached end of byte array or hit padding '=' characters. if (shiftto == 6) { dst[dp++] = (byte)(bits >> 16); } else if (shiftto == 0) { dst[dp++] = (byte)(bits >> 16); dst[dp++] = (byte)(bits >> 8); } else if (shiftto == 12) { // dangling single "x", incorrectly encoded. throw new IllegalArgumentException( "Last unit does not have enough valid bits"); } // anything left is invalid, if is not MIME. // if MIME, ignore all non-base64 character while (sp < sl) { if (isMIME && base64[src[sp++]] < 0) continue; throw new IllegalArgumentException( "Input byte array has incorrect ending byte at " + sp); } return dp; } } /* * An output stream for encoding bytes into the Base64. */ private static class EncOutputStream extends FilterOutputStream { private int leftover = 0; private int b0, b1, b2; private boolean closed = false; private final char[] base64; // byte->base64 mapping private final byte[] newline; // line separator, if needed private final int linemax; private final boolean doPadding;// whether or not to pad private int linepos = 0; EncOutputStream(OutputStream os, char[] base64, byte[] newline, int linemax, boolean doPadding) { super(os); this.base64 = base64; this.newline = newline; this.linemax = linemax; this.doPadding = doPadding; } @Override public void write(int b) throws IOException { byte[] buf = new byte[1]; buf[0] = (byte)(b & 0xff); write(buf, 0, 1); } private void checkNewline() throws IOException { if (linepos == linemax) { out.write(newline); linepos = 0; } } @Override public void write(byte[] b, int off, int len) throws IOException { if (closed) throw new IOException("Stream is closed"); // Android-changed: Upstream fix to avoid overflow. // This upstream fix is from beyond OpenJDK8u121-b13. http://b/62368386 // if (off < 0 || len < 0 || off + len > b.length) if (off < 0 || len < 0 || len > b.length - off) throw new ArrayIndexOutOfBoundsException(); if (len == 0) return; if (leftover != 0) { if (leftover == 1) { b1 = b[off++] & 0xff; len--; if (len == 0) { leftover++; return; } } b2 = b[off++] & 0xff; len--; checkNewline(); out.write(base64[b0 >> 2]); out.write(base64[(b0 << 4) & 0x3f | (b1 >> 4)]); out.write(base64[(b1 << 2) & 0x3f | (b2 >> 6)]); out.write(base64[b2 & 0x3f]); linepos += 4; } int nBits24 = len / 3; leftover = len - (nBits24 * 3); while (nBits24-- > 0) { checkNewline(); int bits = (b[off++] & 0xff) << 16 | (b[off++] & 0xff) << 8 | (b[off++] & 0xff); out.write(base64[(bits >>> 18) & 0x3f]); out.write(base64[(bits >>> 12) & 0x3f]); out.write(base64[(bits >>> 6) & 0x3f]); out.write(base64[bits & 0x3f]); linepos += 4; } if (leftover == 1) { b0 = b[off++] & 0xff; } else if (leftover == 2) { b0 = b[off++] & 0xff; b1 = b[off++] & 0xff; } } @Override public void close() throws IOException { if (!closed) { closed = true; if (leftover == 1) { checkNewline(); out.write(base64[b0 >> 2]); out.write(base64[(b0 << 4) & 0x3f]); if (doPadding) { out.write('='); out.write('='); } } else if (leftover == 2) { checkNewline(); out.write(base64[b0 >> 2]); out.write(base64[(b0 << 4) & 0x3f | (b1 >> 4)]); out.write(base64[(b1 << 2) & 0x3f]); if (doPadding) { out.write('='); } } leftover = 0; out.close(); } } } /* * An input stream for decoding Base64 bytes */ private static class DecInputStream extends InputStream { private final InputStream is; private final boolean isMIME; private final int[] base64; // base64 -> byte mapping private int bits = 0; // 24-bit buffer for decoding private int nextin = 18; // next available "off" in "bits" for input; // -> 18, 12, 6, 0 private int nextout = -8; // next available "off" in "bits" for output; // -> 8, 0, -8 (no byte for output) private boolean eof = false; private boolean closed = false; DecInputStream(InputStream is, int[] base64, boolean isMIME) { this.is = is; this.base64 = base64; this.isMIME = isMIME; } private byte[] sbBuf = new byte[1]; @Override public int read() throws IOException { return read(sbBuf, 0, 1) == -1 ? -1 : sbBuf[0] & 0xff; } @Override public int read(byte[] b, int off, int len) throws IOException { if (closed) throw new IOException("Stream is closed"); if (eof && nextout < 0) // eof and no leftover return -1; if (off < 0 || len < 0 || len > b.length - off) throw new IndexOutOfBoundsException(); int oldOff = off; if (nextout >= 0) { // leftover output byte(s) in bits buf do { if (len == 0) return off - oldOff; b[off++] = (byte)(bits >> nextout); len--; nextout -= 8; } while (nextout >= 0); bits = 0; } while (len > 0) { int v = is.read(); if (v == -1) { eof = true; if (nextin != 18) { if (nextin == 12) throw new IOException("Base64 stream has one un-decoded dangling byte."); // treat ending xx/xxx without padding character legal. // same logic as v == '=' below b[off++] = (byte)(bits >> (16)); len--; if (nextin == 0) { // only one padding byte if (len == 0) { // no enough output space bits >>= 8; // shift to lowest byte nextout = 0; } else { b[off++] = (byte) (bits >> 8); } } } if (off == oldOff) return -1; else return off - oldOff; } if (v == '=') { // padding byte(s) // = shiftto==18 unnecessary padding // x= shiftto==12 dangling x, invalid unit // xx= shiftto==6 && missing last '=' // xx=y or last is not '=' if (nextin == 18 || nextin == 12 || nextin == 6 && is.read() != '=') { throw new IOException("Illegal base64 ending sequence:" + nextin); } b[off++] = (byte)(bits >> (16)); len--; if (nextin == 0) { // only one padding byte if (len == 0) { // no enough output space bits >>= 8; // shift to lowest byte nextout = 0; } else { b[off++] = (byte) (bits >> 8); } } eof = true; break; } if ((v = base64[v]) == -1) { if (isMIME) // skip if for rfc2045 continue; else throw new IOException("Illegal base64 character " + Integer.toString(v, 16)); } bits |= (v << nextin); if (nextin == 0) { nextin = 18; // clear for next nextout = 16; while (nextout >= 0) { b[off++] = (byte)(bits >> nextout); len--; nextout -= 8; if (len == 0 && nextout >= 0) { // don't clean "bits" return off - oldOff; } } bits = 0; } else { nextin -= 6; } } return off - oldOff; } @Override public int available() throws IOException { if (closed) throw new IOException("Stream is closed"); return is.available(); // TBD: } @Override public void close() throws IOException { if (!closed) { closed = true; is.close(); } } } }