J. Duke | 319a3b9 | 2007-12-01 00:00:00 +0000 | [diff] [blame^] | 1 | /* |
| 2 | * Copyright 2001-2005 Sun Microsystems, Inc. All Rights Reserved. |
| 3 | * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. |
| 4 | * |
| 5 | * This code is free software; you can redistribute it and/or modify it |
| 6 | * under the terms of the GNU General Public License version 2 only, as |
| 7 | * published by the Free Software Foundation. Sun designates this |
| 8 | * particular file as subject to the "Classpath" exception as provided |
| 9 | * by Sun in the LICENSE file that accompanied this code. |
| 10 | * |
| 11 | * This code is distributed in the hope that it will be useful, but WITHOUT |
| 12 | * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
| 13 | * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
| 14 | * version 2 for more details (a copy is included in the LICENSE file that |
| 15 | * accompanied this code). |
| 16 | * |
| 17 | * You should have received a copy of the GNU General Public License version |
| 18 | * 2 along with this work; if not, write to the Free Software Foundation, |
| 19 | * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
| 20 | * |
| 21 | * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, |
| 22 | * CA 95054 USA or visit www.sun.com if you need additional information or |
| 23 | * have any questions. |
| 24 | */ |
| 25 | package sun.nio.cs.ext; |
| 26 | |
| 27 | import java.io.ByteArrayOutputStream; |
| 28 | import java.nio.ByteBuffer; |
| 29 | import java.nio.CharBuffer; |
| 30 | import java.nio.charset.*; |
| 31 | |
| 32 | import java.util.Collections; |
| 33 | import java.util.HashMap; |
| 34 | import java.util.Iterator; |
| 35 | import java.util.List; |
| 36 | import java.util.Map; |
| 37 | |
| 38 | public class COMPOUND_TEXT_Encoder extends CharsetEncoder { |
| 39 | |
| 40 | /** |
| 41 | * NOTE: The following four static variables should be used *only* for |
| 42 | * testing whether a encoder can encode a specific character. They |
| 43 | * cannot be used for actual encoding because they are shared across all |
| 44 | * COMPOUND_TEXT encoders and may be stateful. |
| 45 | */ |
| 46 | private static final Map encodingToEncoderMap = |
| 47 | Collections.synchronizedMap(new HashMap(21, 1.0f)); |
| 48 | private static final CharsetEncoder latin1Encoder; |
| 49 | private static final CharsetEncoder defaultEncoder; |
| 50 | private static final boolean defaultEncodingSupported; |
| 51 | |
| 52 | static { |
| 53 | CharsetEncoder encoder = Charset.defaultCharset().newEncoder(); |
| 54 | String encoding = encoder.charset().name(); |
| 55 | if ("ISO8859_1".equals(encoding)) { |
| 56 | latin1Encoder = encoder; |
| 57 | defaultEncoder = encoder; |
| 58 | defaultEncodingSupported = true; |
| 59 | } else { |
| 60 | try { |
| 61 | latin1Encoder = |
| 62 | Charset.forName("ISO8859_1").newEncoder(); |
| 63 | } catch (IllegalArgumentException e) { |
| 64 | throw new ExceptionInInitializerError |
| 65 | ("ISO8859_1 unsupported"); |
| 66 | } |
| 67 | defaultEncoder = encoder; |
| 68 | defaultEncodingSupported = CompoundTextSupport.getEncodings(). |
| 69 | contains(defaultEncoder.charset().name()); |
| 70 | } |
| 71 | } |
| 72 | |
| 73 | private CharsetEncoder encoder; |
| 74 | private char[] charBuf = new char[1]; |
| 75 | private CharBuffer charbuf = CharBuffer.wrap(charBuf); |
| 76 | private ByteArrayOutputStream nonStandardCharsetBuffer; |
| 77 | private byte[] byteBuf; |
| 78 | private ByteBuffer bytebuf; |
| 79 | private int numNonStandardChars, nonStandardEncodingLen; |
| 80 | |
| 81 | public COMPOUND_TEXT_Encoder(Charset cs) { |
| 82 | super(cs, |
| 83 | (float)(CompoundTextSupport.MAX_CONTROL_SEQUENCE_LEN + 2), |
| 84 | (float)(CompoundTextSupport.MAX_CONTROL_SEQUENCE_LEN + 2)); |
| 85 | try { |
| 86 | encoder = Charset.forName("ISO8859_1").newEncoder(); |
| 87 | } catch (IllegalArgumentException cannotHappen) {} |
| 88 | initEncoder(encoder); |
| 89 | } |
| 90 | |
| 91 | protected CoderResult encodeLoop(CharBuffer src, ByteBuffer des) { |
| 92 | CoderResult cr = CoderResult.UNDERFLOW; |
| 93 | char[] input = src.array(); |
| 94 | int inOff = src.arrayOffset() + src.position(); |
| 95 | int inEnd = src.arrayOffset() + src.limit(); |
| 96 | |
| 97 | try { |
| 98 | while (inOff < inEnd && cr.isUnderflow()) { |
| 99 | charBuf[0] = input[inOff]; |
| 100 | if (charBuf[0] <= '\u0008' || |
| 101 | (charBuf[0] >= '\u000B' && charBuf[0] <= '\u001F') || |
| 102 | (charBuf[0] >= '\u0080' && charBuf[0] <= '\u009F')) { |
| 103 | // The compound text specification only permits the octets |
| 104 | // 0x09, 0x0A, 0x1B, and 0x9B in C0 and C1. Of these, 1B and |
| 105 | // 9B must also be removed because they initiate control |
| 106 | // sequences. |
| 107 | charBuf[0] = '?'; |
| 108 | } |
| 109 | |
| 110 | CharsetEncoder enc = getEncoder(charBuf[0]); |
| 111 | //System.out.println("char=" + charBuf[0] + ", enc=" + enc); |
| 112 | if (enc == null) { |
| 113 | if (unmappableCharacterAction() |
| 114 | == CodingErrorAction.REPORT) { |
| 115 | charBuf[0] = '?'; |
| 116 | enc = latin1Encoder; |
| 117 | } else { |
| 118 | return CoderResult.unmappableForLength(1); |
| 119 | } |
| 120 | } |
| 121 | if (enc != encoder) { |
| 122 | if (nonStandardCharsetBuffer != null) { |
| 123 | cr = flushNonStandardCharsetBuffer(des); |
| 124 | } else { |
| 125 | //cr= encoder.flush(des); |
| 126 | flushEncoder(encoder, des); |
| 127 | } |
| 128 | if (!cr.isUnderflow()) |
| 129 | return cr; |
| 130 | byte[] escSequence = CompoundTextSupport. |
| 131 | getEscapeSequence(enc.charset().name()); |
| 132 | if (escSequence == null) { |
| 133 | throw new InternalError("Unknown encoding: " + |
| 134 | enc.charset().name()); |
| 135 | } else if (escSequence[1] == (byte)0x25 && |
| 136 | escSequence[2] == (byte)0x2F) { |
| 137 | initNonStandardCharsetBuffer(enc, escSequence); |
| 138 | } else if (des.remaining() >= escSequence.length) { |
| 139 | des.put(escSequence, 0, escSequence.length); |
| 140 | } else { |
| 141 | return CoderResult.OVERFLOW; |
| 142 | } |
| 143 | encoder = enc; |
| 144 | continue; |
| 145 | } |
| 146 | charbuf.rewind(); |
| 147 | if (nonStandardCharsetBuffer == null) { |
| 148 | cr = encoder.encode(charbuf, des, false); |
| 149 | } else { |
| 150 | bytebuf.clear(); |
| 151 | cr = encoder.encode(charbuf, bytebuf, false); |
| 152 | bytebuf.flip(); |
| 153 | nonStandardCharsetBuffer.write(byteBuf, |
| 154 | 0, bytebuf.limit()); |
| 155 | numNonStandardChars++; |
| 156 | } |
| 157 | inOff++; |
| 158 | } |
| 159 | return cr; |
| 160 | } finally { |
| 161 | src.position(inOff - src.arrayOffset()); |
| 162 | } |
| 163 | } |
| 164 | |
| 165 | protected CoderResult implFlush(ByteBuffer out) { |
| 166 | CoderResult cr = (nonStandardCharsetBuffer != null) |
| 167 | ? flushNonStandardCharsetBuffer(out) |
| 168 | //: encoder.flush(out); |
| 169 | : flushEncoder(encoder, out); |
| 170 | reset(); |
| 171 | return cr; |
| 172 | } |
| 173 | |
| 174 | private void initNonStandardCharsetBuffer(CharsetEncoder c, |
| 175 | byte[] escSequence) |
| 176 | { |
| 177 | nonStandardCharsetBuffer = new ByteArrayOutputStream(); |
| 178 | byteBuf = new byte[(int)c.maxBytesPerChar()]; |
| 179 | bytebuf = ByteBuffer.wrap(byteBuf); |
| 180 | nonStandardCharsetBuffer.write(escSequence, 0, escSequence.length); |
| 181 | nonStandardCharsetBuffer.write(0); // M placeholder |
| 182 | nonStandardCharsetBuffer.write(0); // L placeholder |
| 183 | byte[] encoding = CompoundTextSupport. |
| 184 | getEncoding(c.charset().name()); |
| 185 | if (encoding == null) { |
| 186 | throw new InternalError |
| 187 | ("Unknown encoding: " + encoder.charset().name()); |
| 188 | } |
| 189 | nonStandardCharsetBuffer.write(encoding, 0, encoding.length); |
| 190 | nonStandardCharsetBuffer.write(0x02); // divider |
| 191 | nonStandardEncodingLen = encoding.length + 1; |
| 192 | } |
| 193 | |
| 194 | private CoderResult flushNonStandardCharsetBuffer(ByteBuffer out) { |
| 195 | if (numNonStandardChars > 0) { |
| 196 | byte[] flushBuf = new byte[(int)encoder.maxBytesPerChar() * |
| 197 | numNonStandardChars]; |
| 198 | ByteBuffer bb = ByteBuffer.wrap(flushBuf); |
| 199 | flushEncoder(encoder, bb); |
| 200 | bb.flip(); |
| 201 | nonStandardCharsetBuffer.write(flushBuf, 0, bb.limit()); |
| 202 | numNonStandardChars = 0; |
| 203 | } |
| 204 | |
| 205 | int numBytes = nonStandardCharsetBuffer.size(); |
| 206 | int nonStandardBytesOff = 6 + nonStandardEncodingLen; |
| 207 | |
| 208 | if (out.remaining() < (numBytes - nonStandardBytesOff) + |
| 209 | nonStandardBytesOff * (((numBytes - nonStandardBytesOff) / |
| 210 | ((1 << 14) - 1)) + 1)) |
| 211 | { |
| 212 | return CoderResult.OVERFLOW; |
| 213 | } |
| 214 | |
| 215 | byte[] nonStandardBytes = |
| 216 | nonStandardCharsetBuffer.toByteArray(); |
| 217 | |
| 218 | // The non-standard charset header only supports 2^14-1 bytes of data. |
| 219 | // If we have more than that, we have to repeat the header. |
| 220 | do { |
| 221 | out.put((byte)0x1B); |
| 222 | out.put((byte)0x25); |
| 223 | out.put((byte)0x2F); |
| 224 | out.put((byte)nonStandardBytes[3]); |
| 225 | |
| 226 | int toWrite = Math.min(numBytes - nonStandardBytesOff, |
| 227 | (1 << 14) - 1 - nonStandardEncodingLen); |
| 228 | |
| 229 | out.put((byte) |
| 230 | (((toWrite + nonStandardEncodingLen) / 0x80) | 0x80)); // M |
| 231 | out.put((byte) |
| 232 | (((toWrite + nonStandardEncodingLen) % 0x80) | 0x80)); // L |
| 233 | out.put(nonStandardBytes, 6, nonStandardEncodingLen); |
| 234 | out.put(nonStandardBytes, nonStandardBytesOff, toWrite); |
| 235 | nonStandardBytesOff += toWrite; |
| 236 | } while (nonStandardBytesOff < numBytes); |
| 237 | |
| 238 | nonStandardCharsetBuffer = null; |
| 239 | byteBuf = null; |
| 240 | nonStandardEncodingLen = 0; |
| 241 | return CoderResult.UNDERFLOW; |
| 242 | } |
| 243 | |
| 244 | /** |
| 245 | * Resets the encoder. |
| 246 | * Call this method to reset the encoder to its initial state |
| 247 | */ |
| 248 | protected void implReset() { |
| 249 | numNonStandardChars = nonStandardEncodingLen = 0; |
| 250 | nonStandardCharsetBuffer = null; |
| 251 | byteBuf = null; |
| 252 | try { |
| 253 | encoder = Charset.forName("ISO8859_1").newEncoder(); |
| 254 | } catch (IllegalArgumentException cannotHappen) { |
| 255 | } |
| 256 | initEncoder(encoder); |
| 257 | } |
| 258 | |
| 259 | /** |
| 260 | * Return whether a character is mappable or not |
| 261 | * @return true if a character is mappable |
| 262 | */ |
| 263 | public boolean canEncode(char ch) { |
| 264 | return getEncoder(ch) != null; |
| 265 | } |
| 266 | |
| 267 | protected void implOnMalformedInput(CodingErrorAction newAction) { |
| 268 | encoder.onUnmappableCharacter(newAction); |
| 269 | } |
| 270 | |
| 271 | protected void implOnUnmappableCharacter(CodingErrorAction newAction) { |
| 272 | encoder.onUnmappableCharacter(newAction); |
| 273 | } |
| 274 | |
| 275 | protected void implReplaceWith(byte[] newReplacement) { |
| 276 | if (encoder != null) |
| 277 | encoder.replaceWith(newReplacement); |
| 278 | } |
| 279 | |
| 280 | /** |
| 281 | * Try to figure out which CharsetEncoder to use for conversion |
| 282 | * of the specified Unicode character. The target character encoding |
| 283 | * of the returned encoder is approved to be used with Compound Text. |
| 284 | * |
| 285 | * @param ch Unicode character |
| 286 | * @return CharsetEncoder to convert the given character |
| 287 | */ |
| 288 | private CharsetEncoder getEncoder(char ch) { |
| 289 | // 1. Try the current encoder. |
| 290 | if (encoder.canEncode(ch)) { |
| 291 | return encoder; |
| 292 | } |
| 293 | |
| 294 | // 2. Try the default encoder. |
| 295 | if (defaultEncodingSupported && defaultEncoder.canEncode(ch)) { |
| 296 | CharsetEncoder retval = null; |
| 297 | try { |
| 298 | retval = defaultEncoder.charset().newEncoder(); |
| 299 | } catch (UnsupportedOperationException cannotHappen) { |
| 300 | } |
| 301 | initEncoder(retval); |
| 302 | return retval; |
| 303 | } |
| 304 | |
| 305 | // 3. Try ISO8859-1. |
| 306 | if (latin1Encoder.canEncode(ch)) { |
| 307 | CharsetEncoder retval = null; |
| 308 | try { |
| 309 | retval = latin1Encoder.charset().newEncoder(); |
| 310 | } catch (UnsupportedOperationException cannotHappen) {} |
| 311 | initEncoder(retval); |
| 312 | return retval; |
| 313 | } |
| 314 | |
| 315 | // 4. Brute force search of all supported encodings. |
| 316 | for (Iterator iter = CompoundTextSupport.getEncodings().iterator(); |
| 317 | iter.hasNext();) |
| 318 | { |
| 319 | String encoding = (String)iter.next(); |
| 320 | CharsetEncoder enc = |
| 321 | (CharsetEncoder)encodingToEncoderMap.get(encoding); |
| 322 | if (enc == null) { |
| 323 | enc = CompoundTextSupport.getEncoder(encoding); |
| 324 | if (enc == null) { |
| 325 | throw new InternalError("Unsupported encoding: " + |
| 326 | encoding); |
| 327 | } |
| 328 | encodingToEncoderMap.put(encoding, enc); |
| 329 | } |
| 330 | if (enc.canEncode(ch)) { |
| 331 | CharsetEncoder retval = CompoundTextSupport.getEncoder(encoding); |
| 332 | initEncoder(retval); |
| 333 | return retval; |
| 334 | } |
| 335 | } |
| 336 | |
| 337 | return null; |
| 338 | } |
| 339 | |
| 340 | private void initEncoder(CharsetEncoder enc) { |
| 341 | try { |
| 342 | enc.onUnmappableCharacter(CodingErrorAction.REPLACE) |
| 343 | .replaceWith(replacement()); |
| 344 | } catch (IllegalArgumentException x) {} |
| 345 | } |
| 346 | |
| 347 | private CharBuffer fcb= CharBuffer.allocate(0); |
| 348 | private CoderResult flushEncoder(CharsetEncoder enc, ByteBuffer bb) { |
| 349 | enc.encode(fcb, bb, true); |
| 350 | return enc.flush(bb); |
| 351 | } |
| 352 | } |