1 |
771 |
jeremybenn |
/* CharsetEncoder.java --
|
2 |
|
|
Copyright (C) 2002 Free Software Foundation, Inc.
|
3 |
|
|
|
4 |
|
|
This file is part of GNU Classpath.
|
5 |
|
|
|
6 |
|
|
GNU Classpath is free software; you can redistribute it and/or modify
|
7 |
|
|
it under the terms of the GNU General Public License as published by
|
8 |
|
|
the Free Software Foundation; either version 2, or (at your option)
|
9 |
|
|
any later version.
|
10 |
|
|
|
11 |
|
|
GNU Classpath is distributed in the hope that it will be useful, but
|
12 |
|
|
WITHOUT ANY WARRANTY; without even the implied warranty of
|
13 |
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
14 |
|
|
General Public License for more details.
|
15 |
|
|
|
16 |
|
|
You should have received a copy of the GNU General Public License
|
17 |
|
|
along with GNU Classpath; see the file COPYING. If not, write to the
|
18 |
|
|
Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
|
19 |
|
|
02110-1301 USA.
|
20 |
|
|
|
21 |
|
|
Linking this library statically or dynamically with other modules is
|
22 |
|
|
making a combined work based on this library. Thus, the terms and
|
23 |
|
|
conditions of the GNU General Public License cover the whole
|
24 |
|
|
combination.
|
25 |
|
|
|
26 |
|
|
As a special exception, the copyright holders of this library give you
|
27 |
|
|
permission to link this library with independent modules to produce an
|
28 |
|
|
executable, regardless of the license terms of these independent
|
29 |
|
|
modules, and to copy and distribute the resulting executable under
|
30 |
|
|
terms of your choice, provided that you also meet, for each linked
|
31 |
|
|
independent module, the terms and conditions of the license of that
|
32 |
|
|
module. An independent module is a module which is not derived from
|
33 |
|
|
or based on this library. If you modify this library, you may extend
|
34 |
|
|
this exception to your version of the library, but you are not
|
35 |
|
|
obligated to do so. If you do not wish to do so, delete this
|
36 |
|
|
exception statement from your version. */
|
37 |
|
|
|
38 |
|
|
package java.nio.charset;
|
39 |
|
|
|
40 |
|
|
import java.nio.ByteBuffer;
|
41 |
|
|
import java.nio.CharBuffer;
|
42 |
|
|
|
43 |
|
|
/**
|
44 |
|
|
* @author Jesse Rosenstock
|
45 |
|
|
* @since 1.4
|
46 |
|
|
*/
|
47 |
|
|
public abstract class CharsetEncoder
|
48 |
|
|
{
|
49 |
|
|
private static final int STATE_RESET = 0;
|
50 |
|
|
private static final int STATE_CODING = 1;
|
51 |
|
|
private static final int STATE_END = 2;
|
52 |
|
|
private static final int STATE_FLUSHED = 3;
|
53 |
|
|
|
54 |
|
|
private static final byte[] DEFAULT_REPLACEMENT = {(byte)'?'};
|
55 |
|
|
|
56 |
|
|
private final Charset charset;
|
57 |
|
|
private final float averageBytesPerChar;
|
58 |
|
|
private final float maxBytesPerChar;
|
59 |
|
|
private byte[] replacement;
|
60 |
|
|
|
61 |
|
|
private int state = STATE_RESET;
|
62 |
|
|
|
63 |
|
|
private CodingErrorAction malformedInputAction
|
64 |
|
|
= CodingErrorAction.REPORT;
|
65 |
|
|
private CodingErrorAction unmappableCharacterAction
|
66 |
|
|
= CodingErrorAction.REPORT;
|
67 |
|
|
|
68 |
|
|
protected CharsetEncoder (Charset cs, float averageBytesPerChar,
|
69 |
|
|
float maxBytesPerChar)
|
70 |
|
|
{
|
71 |
|
|
this (cs, averageBytesPerChar, maxBytesPerChar, DEFAULT_REPLACEMENT);
|
72 |
|
|
}
|
73 |
|
|
|
74 |
|
|
protected CharsetEncoder (Charset cs, float averageBytesPerChar,
|
75 |
|
|
float maxBytesPerChar, byte[] replacement)
|
76 |
|
|
{
|
77 |
|
|
if (averageBytesPerChar <= 0.0f)
|
78 |
|
|
throw new IllegalArgumentException ("Non-positive averageBytesPerChar");
|
79 |
|
|
if (maxBytesPerChar <= 0.0f)
|
80 |
|
|
throw new IllegalArgumentException ("Non-positive maxBytesPerChar");
|
81 |
|
|
|
82 |
|
|
this.charset = cs;
|
83 |
|
|
this.averageBytesPerChar
|
84 |
|
|
= averageBytesPerChar;
|
85 |
|
|
this.maxBytesPerChar
|
86 |
|
|
= maxBytesPerChar;
|
87 |
|
|
this.replacement = replacement;
|
88 |
|
|
implReplaceWith (replacement);
|
89 |
|
|
}
|
90 |
|
|
|
91 |
|
|
public final float averageBytesPerChar ()
|
92 |
|
|
{
|
93 |
|
|
return averageBytesPerChar;
|
94 |
|
|
}
|
95 |
|
|
|
96 |
|
|
public boolean canEncode (char c)
|
97 |
|
|
{
|
98 |
|
|
CharBuffer cb = CharBuffer.allocate (1).put (c);
|
99 |
|
|
cb.flip ();
|
100 |
|
|
return canEncode (cb);
|
101 |
|
|
}
|
102 |
|
|
|
103 |
|
|
public boolean canEncode (CharSequence cs)
|
104 |
|
|
{
|
105 |
|
|
CharBuffer cb;
|
106 |
|
|
if (cs instanceof CharBuffer)
|
107 |
|
|
cb = ((CharBuffer) cs).duplicate ();
|
108 |
|
|
else
|
109 |
|
|
cb = CharBuffer.wrap (cs);
|
110 |
|
|
return canEncode (cb);
|
111 |
|
|
}
|
112 |
|
|
|
113 |
|
|
private boolean canEncode (CharBuffer cb)
|
114 |
|
|
{
|
115 |
|
|
// It is an error if a coding operation is "in progress"
|
116 |
|
|
// I take that to mean the state is not reset or flushed.
|
117 |
|
|
// XXX: check "in progress" everywhere
|
118 |
|
|
if (state == STATE_FLUSHED)
|
119 |
|
|
reset ();
|
120 |
|
|
else if (state != STATE_RESET)
|
121 |
|
|
throw new IllegalStateException ();
|
122 |
|
|
|
123 |
|
|
CodingErrorAction oldMalformedInputAction = malformedInputAction;
|
124 |
|
|
CodingErrorAction oldUnmappableCharacterAction
|
125 |
|
|
= unmappableCharacterAction;
|
126 |
|
|
|
127 |
|
|
try
|
128 |
|
|
{
|
129 |
|
|
if (oldMalformedInputAction != CodingErrorAction.REPORT)
|
130 |
|
|
onMalformedInput (CodingErrorAction.REPORT);
|
131 |
|
|
if (oldUnmappableCharacterAction != CodingErrorAction.REPORT)
|
132 |
|
|
onUnmappableCharacter (CodingErrorAction.REPORT);
|
133 |
|
|
}
|
134 |
|
|
catch (Exception e)
|
135 |
|
|
{
|
136 |
|
|
return false;
|
137 |
|
|
}
|
138 |
|
|
finally
|
139 |
|
|
{
|
140 |
|
|
if (oldMalformedInputAction != CodingErrorAction.REPORT)
|
141 |
|
|
onMalformedInput (oldMalformedInputAction);
|
142 |
|
|
if (oldUnmappableCharacterAction != CodingErrorAction.REPORT)
|
143 |
|
|
onUnmappableCharacter (oldUnmappableCharacterAction);
|
144 |
|
|
}
|
145 |
|
|
|
146 |
|
|
return true;
|
147 |
|
|
}
|
148 |
|
|
|
149 |
|
|
public final Charset charset ()
|
150 |
|
|
{
|
151 |
|
|
return charset;
|
152 |
|
|
}
|
153 |
|
|
|
154 |
|
|
public final ByteBuffer encode (CharBuffer in)
|
155 |
|
|
throws CharacterCodingException
|
156 |
|
|
{
|
157 |
|
|
// XXX: Sun's Javadoc seems to contradict itself saying an
|
158 |
|
|
// IllegalStateException is thrown "if a decoding operation is already
|
159 |
|
|
// in progress" and also that "it resets this Encoder".
|
160 |
|
|
// Should we check to see that the state is reset, or should we
|
161 |
|
|
// call reset()?
|
162 |
|
|
if (state != STATE_RESET)
|
163 |
|
|
throw new IllegalStateException ();
|
164 |
|
|
|
165 |
|
|
// REVIEW: Using max instead of average may allocate a very large
|
166 |
|
|
// buffer. Maybe we should do something more efficient?
|
167 |
|
|
int remaining = in.remaining ();
|
168 |
|
|
int n = (int) (remaining * maxBytesPerChar ());
|
169 |
|
|
ByteBuffer out = ByteBuffer.allocate (n);
|
170 |
|
|
|
171 |
|
|
if (remaining == 0)
|
172 |
|
|
{
|
173 |
|
|
state = STATE_FLUSHED;
|
174 |
|
|
return out;
|
175 |
|
|
}
|
176 |
|
|
|
177 |
|
|
CoderResult cr = encode (in, out, true);
|
178 |
|
|
if (cr.isError ())
|
179 |
|
|
cr.throwException ();
|
180 |
|
|
|
181 |
|
|
cr = flush (out);
|
182 |
|
|
if (cr.isError ())
|
183 |
|
|
cr.throwException ();
|
184 |
|
|
|
185 |
|
|
out.flip ();
|
186 |
|
|
|
187 |
|
|
// Unfortunately, resizing the actual bytebuffer array is required.
|
188 |
|
|
byte[] resized = new byte[out.remaining()];
|
189 |
|
|
out.get(resized);
|
190 |
|
|
return ByteBuffer.wrap(resized);
|
191 |
|
|
}
|
192 |
|
|
|
193 |
|
|
public final CoderResult encode (CharBuffer in, ByteBuffer out,
|
194 |
|
|
boolean endOfInput)
|
195 |
|
|
{
|
196 |
|
|
int newState = endOfInput ? STATE_END : STATE_CODING;
|
197 |
|
|
// XXX: Need to check for "previous step was an invocation [not] of
|
198 |
|
|
// this method with a value of true for the endOfInput parameter but
|
199 |
|
|
// a return value indicating an incomplete decoding operation"
|
200 |
|
|
// XXX: We will not check the previous return value, just
|
201 |
|
|
// that the previous call passed true for endOfInput
|
202 |
|
|
if (state != STATE_RESET && state != STATE_CODING
|
203 |
|
|
&& !(endOfInput && state == STATE_END))
|
204 |
|
|
throw new IllegalStateException ();
|
205 |
|
|
state = newState;
|
206 |
|
|
|
207 |
|
|
for (;;)
|
208 |
|
|
{
|
209 |
|
|
CoderResult cr;
|
210 |
|
|
try
|
211 |
|
|
{
|
212 |
|
|
cr = encodeLoop (in, out);
|
213 |
|
|
}
|
214 |
|
|
catch (RuntimeException e)
|
215 |
|
|
{
|
216 |
|
|
throw new CoderMalfunctionError (e);
|
217 |
|
|
}
|
218 |
|
|
|
219 |
|
|
if (cr.isOverflow ())
|
220 |
|
|
return cr;
|
221 |
|
|
|
222 |
|
|
if (cr.isUnderflow ())
|
223 |
|
|
{
|
224 |
|
|
if (endOfInput && in.hasRemaining ())
|
225 |
|
|
cr = CoderResult.malformedForLength (in.remaining ());
|
226 |
|
|
else
|
227 |
|
|
return cr;
|
228 |
|
|
}
|
229 |
|
|
|
230 |
|
|
CodingErrorAction action = cr.isMalformed ()
|
231 |
|
|
? malformedInputAction
|
232 |
|
|
: unmappableCharacterAction;
|
233 |
|
|
|
234 |
|
|
if (action == CodingErrorAction.REPORT)
|
235 |
|
|
return cr;
|
236 |
|
|
|
237 |
|
|
if (action == CodingErrorAction.REPLACE)
|
238 |
|
|
{
|
239 |
|
|
if (out.remaining () < replacement.length)
|
240 |
|
|
return CoderResult.OVERFLOW;
|
241 |
|
|
out.put (replacement);
|
242 |
|
|
}
|
243 |
|
|
|
244 |
|
|
in.position (in.position () + cr.length ());
|
245 |
|
|
}
|
246 |
|
|
}
|
247 |
|
|
|
248 |
|
|
protected abstract CoderResult encodeLoop (CharBuffer in, ByteBuffer out);
|
249 |
|
|
|
250 |
|
|
public final CoderResult flush (ByteBuffer out)
|
251 |
|
|
{
|
252 |
|
|
// It seems weird that you can flush after reset, but Sun's javadoc
|
253 |
|
|
// says an IllegalStateException is thrown "If the previous step of the
|
254 |
|
|
// current decoding operation was an invocation neither of the reset
|
255 |
|
|
// method nor ... of the three-argument encode method with a value of
|
256 |
|
|
// true for the endOfInput parameter."
|
257 |
|
|
// Further note that flush() only requires that there not be
|
258 |
|
|
// an IllegalStateException if the previous step was a call to
|
259 |
|
|
// encode with true as the last argument. It does not require
|
260 |
|
|
// that the call succeeded. encode() does require that it succeeded.
|
261 |
|
|
// XXX: test this to see if reality matches javadoc
|
262 |
|
|
if (state != STATE_RESET && state != STATE_END)
|
263 |
|
|
throw new IllegalStateException ();
|
264 |
|
|
|
265 |
|
|
state = STATE_FLUSHED;
|
266 |
|
|
return implFlush (out);
|
267 |
|
|
}
|
268 |
|
|
|
269 |
|
|
protected CoderResult implFlush (ByteBuffer out)
|
270 |
|
|
{
|
271 |
|
|
return CoderResult.UNDERFLOW;
|
272 |
|
|
}
|
273 |
|
|
|
274 |
|
|
protected void implOnMalformedInput (CodingErrorAction newAction)
|
275 |
|
|
{
|
276 |
|
|
// default implementation does nothing
|
277 |
|
|
}
|
278 |
|
|
|
279 |
|
|
protected void implOnUnmappableCharacter (CodingErrorAction newAction)
|
280 |
|
|
{
|
281 |
|
|
// default implementation does nothing
|
282 |
|
|
}
|
283 |
|
|
|
284 |
|
|
protected void implReplaceWith (byte[] newReplacement)
|
285 |
|
|
{
|
286 |
|
|
// default implementation does nothing
|
287 |
|
|
}
|
288 |
|
|
|
289 |
|
|
protected void implReset ()
|
290 |
|
|
{
|
291 |
|
|
// default implementation does nothing
|
292 |
|
|
}
|
293 |
|
|
|
294 |
|
|
public boolean isLegalReplacement (byte[] replacement)
|
295 |
|
|
{
|
296 |
|
|
// TODO: cache the decoder
|
297 |
|
|
// error actions will be REPORT after construction
|
298 |
|
|
CharsetDecoder decoder = charset.newDecoder ();
|
299 |
|
|
ByteBuffer bb = ByteBuffer.wrap (replacement);
|
300 |
|
|
CharBuffer cb
|
301 |
|
|
= CharBuffer.allocate ((int) (replacement.length
|
302 |
|
|
* decoder.maxCharsPerByte ()));
|
303 |
|
|
return !decoder.decode (bb, cb, true).isError ();
|
304 |
|
|
}
|
305 |
|
|
|
306 |
|
|
public CodingErrorAction malformedInputAction ()
|
307 |
|
|
{
|
308 |
|
|
return malformedInputAction;
|
309 |
|
|
}
|
310 |
|
|
|
311 |
|
|
public final float maxBytesPerChar ()
|
312 |
|
|
{
|
313 |
|
|
return maxBytesPerChar;
|
314 |
|
|
}
|
315 |
|
|
|
316 |
|
|
public final CharsetEncoder onMalformedInput (CodingErrorAction newAction)
|
317 |
|
|
{
|
318 |
|
|
if (newAction == null)
|
319 |
|
|
throw new IllegalArgumentException ("Null action");
|
320 |
|
|
|
321 |
|
|
malformedInputAction = newAction;
|
322 |
|
|
implOnMalformedInput (newAction);
|
323 |
|
|
return this;
|
324 |
|
|
}
|
325 |
|
|
|
326 |
|
|
public CodingErrorAction unmappableCharacterAction ()
|
327 |
|
|
{
|
328 |
|
|
return unmappableCharacterAction;
|
329 |
|
|
}
|
330 |
|
|
|
331 |
|
|
public final CharsetEncoder onUnmappableCharacter
|
332 |
|
|
(CodingErrorAction newAction)
|
333 |
|
|
{
|
334 |
|
|
if (newAction == null)
|
335 |
|
|
throw new IllegalArgumentException ("Null action");
|
336 |
|
|
|
337 |
|
|
unmappableCharacterAction = newAction;
|
338 |
|
|
implOnUnmappableCharacter (newAction);
|
339 |
|
|
return this;
|
340 |
|
|
}
|
341 |
|
|
|
342 |
|
|
public final byte[] replacement ()
|
343 |
|
|
{
|
344 |
|
|
return replacement;
|
345 |
|
|
}
|
346 |
|
|
|
347 |
|
|
public final CharsetEncoder replaceWith (byte[] newReplacement)
|
348 |
|
|
{
|
349 |
|
|
if (newReplacement == null)
|
350 |
|
|
throw new IllegalArgumentException ("Null replacement");
|
351 |
|
|
if (newReplacement.length == 0)
|
352 |
|
|
throw new IllegalArgumentException ("Empty replacement");
|
353 |
|
|
// XXX: what about maxBytesPerChar?
|
354 |
|
|
|
355 |
|
|
if (!isLegalReplacement (newReplacement))
|
356 |
|
|
throw new IllegalArgumentException ("Illegal replacement");
|
357 |
|
|
|
358 |
|
|
this.replacement = newReplacement;
|
359 |
|
|
implReplaceWith (newReplacement);
|
360 |
|
|
return this;
|
361 |
|
|
}
|
362 |
|
|
|
363 |
|
|
public final CharsetEncoder reset ()
|
364 |
|
|
{
|
365 |
|
|
state = STATE_RESET;
|
366 |
|
|
implReset ();
|
367 |
|
|
return this;
|
368 |
|
|
}
|
369 |
|
|
}
|