root / trunk / libraries / libFMap / src / com / iver / cit / gvsig / fmap / drivers / shp / DbaseFileReaderNIO.java @ 1100
History | View | Annotate | Download (15.3 KB)
1 | 377 | fjp | /*
|
---|---|---|---|
2 | * Geotools - OpenSource mapping toolkit
|
||
3 | * (C) 2002, Centre for Computational Geography
|
||
4 | *
|
||
5 | * This library is free software; you can redistribute it and/or
|
||
6 | * modify it under the terms of the GNU Lesser General Public
|
||
7 | * License as published by the Free Software Foundation;
|
||
8 | * version 2.1 of the License.
|
||
9 | *
|
||
10 | * This library is distributed in the hope that it will be useful,
|
||
11 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||
12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
||
13 | * Lesser General Public License for more details.
|
||
14 | *
|
||
15 | * You should have received a copy of the GNU Lesser General Public
|
||
16 | * License along with this library; if not, write to the Free Software
|
||
17 | * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
||
18 | *
|
||
19 | * This file is based on an origional contained in the GISToolkit project:
|
||
20 | * http://gistoolkit.sourceforge.net/
|
||
21 | *
|
||
22 | */
|
||
23 | 1100 | fjp | /* gvSIG. Sistema de Informaci?n Geogr?fica de la Generalitat Valenciana
|
24 | *
|
||
25 | * Copyright (C) 2004 IVER T.I. and Generalitat Valenciana.
|
||
26 | *
|
||
27 | * This program is free software; you can redistribute it and/or
|
||
28 | * modify it under the terms of the GNU General Public License
|
||
29 | * as published by the Free Software Foundation; either version 2
|
||
30 | * of the License, or (at your option) any later version.
|
||
31 | *
|
||
32 | * This program is distributed in the hope that it will be useful,
|
||
33 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||
34 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
||
35 | * GNU General Public License for more details.
|
||
36 | *
|
||
37 | * You should have received a copy of the GNU General Public License
|
||
38 | * along with this program; if not, write to the Free Software
|
||
39 | * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307,USA.
|
||
40 | *
|
||
41 | * For more information, contact:
|
||
42 | *
|
||
43 | * Generalitat Valenciana
|
||
44 | * Conselleria d'Infraestructures i Transport
|
||
45 | * Av. Blasco Ib??ez, 50
|
||
46 | * 46010 VALENCIA
|
||
47 | * SPAIN
|
||
48 | *
|
||
49 | * +34 963862235
|
||
50 | * gvsig@gva.es
|
||
51 | * www.gvsig.gva.es
|
||
52 | *
|
||
53 | * or
|
||
54 | *
|
||
55 | * IVER T.I. S.A
|
||
56 | * Salamanca 50
|
||
57 | * 46005 Valencia
|
||
58 | * Spain
|
||
59 | *
|
||
60 | * +34 963163400
|
||
61 | * dac@iver.es
|
||
62 | */
|
||
63 | 377 | fjp | package com.iver.cit.gvsig.fmap.drivers.shp; |
64 | |||
65 | |||
66 | import java.io.*; |
||
67 | import java.util.*; |
||
68 | import java.nio.*; |
||
69 | import java.nio.channels.ReadableByteChannel; |
||
70 | import java.nio.channels.FileChannel; |
||
71 | import java.nio.charset.Charset; |
||
72 | import java.nio.charset.CharsetDecoder; |
||
73 | import org.geotools.resources.NumberParser; |
||
74 | import org.geotools.resources.NIOUtilities; |
||
75 | |||
76 | |||
77 | /** A DbaseFileReader is used to read a dbase III format file.
|
||
78 | * <br>
|
||
79 | * The general use of this class is:
|
||
80 | * <CODE><PRE>
|
||
81 | * FileChannel in = new FileInputStream("thefile.dbf").getChannel();
|
||
82 | * DbaseFileReader r = new DbaseFileReader( in )
|
||
83 | * Object[] fields = new Object[r.getHeader().getNumFields()];
|
||
84 | * while (r.hasNext()) {
|
||
85 | * r.readEntry(fields);
|
||
86 | * // do stuff
|
||
87 | * }
|
||
88 | * r.close();
|
||
89 | * </PRE></CODE>
|
||
90 | * For consumers who wish to be a bit more selective with their reading of rows,
|
||
91 | * the Row object has been added. The semantics are the same as using the
|
||
92 | * readEntry method, but remember that the Row object is always the same. The
|
||
93 | * values are parsed as they are read, so it pays to copy them out (as each call
|
||
94 | * to Row.read() will result in an expensive String parse).
|
||
95 | * <br><b>EACH CALL TO readEntry OR readRow ADVANCES THE FILE!</b><br>
|
||
96 | * An example of using the Row method of reading:
|
||
97 | * <CODE><PRE>
|
||
98 | * FileChannel in = new FileInputStream("thefile.dbf").getChannel();
|
||
99 | * DbaseFileReader r = new DbaseFileReader( in )
|
||
100 | * int fields = r.getHeader().getNumFields();
|
||
101 | * while (r.hasNext()) {
|
||
102 | * DbaseFileReader.Row row = r.readRow();
|
||
103 | * for (int i = 0; i < fields; i++) {
|
||
104 | * // do stuff
|
||
105 | * Foo.bar( row.read(i) );
|
||
106 | * }
|
||
107 | * }
|
||
108 | * r.close();
|
||
109 | * </PRE></CODE>
|
||
110 | *
|
||
111 | * @author Ian Schneider
|
||
112 | */
|
||
113 | public class DbaseFileReaderNIO { |
||
114 | |||
115 | public final class Row { |
||
116 | public Object read(int column) throws IOException { |
||
117 | int offset = getOffset(column);
|
||
118 | return readObject(offset, column);
|
||
119 | } |
||
120 | } |
||
121 | |||
122 | DbaseFileHeaderNIO header; |
||
123 | ByteBuffer buffer;
|
||
124 | ReadableByteChannel channel;
|
||
125 | CharBuffer charBuffer;
|
||
126 | CharsetDecoder decoder;
|
||
127 | char[] fieldTypes; |
||
128 | int[] fieldLengths; |
||
129 | int cnt = 1; |
||
130 | Row row; |
||
131 | NumberParser numberParser = new NumberParser();
|
||
132 | |||
133 | /** Creates a new instance of DBaseFileReader
|
||
134 | * @param channel The readable channel to use.
|
||
135 | * @throws IOException If an error occurs while initializing.
|
||
136 | */
|
||
137 | public DbaseFileReaderNIO(ReadableByteChannel channel) throws IOException { |
||
138 | this.channel = channel;
|
||
139 | |||
140 | header = new DbaseFileHeaderNIO();
|
||
141 | ///header.readHeader(channel);
|
||
142 | |||
143 | init(); |
||
144 | } |
||
145 | |||
146 | private int fill(ByteBuffer buffer,ReadableByteChannel channel) throws IOException { |
||
147 | int r = buffer.remaining();
|
||
148 | // channel reads return -1 when EOF or other error
|
||
149 | // because they a non-blocking reads, 0 is a valid return value!!
|
||
150 | while (buffer.remaining() > 0 && r != -1) { |
||
151 | r = channel.read(buffer); |
||
152 | } |
||
153 | if (r == -1) { |
||
154 | buffer.limit(buffer.position()); |
||
155 | } |
||
156 | return r;
|
||
157 | } |
||
158 | |||
159 | private void bufferCheck() throws IOException { |
||
160 | // remaining is less than record length
|
||
161 | // compact the remaining data and read again
|
||
162 | if (!buffer.isReadOnly() && buffer.remaining() < header.getRecordLength()) {
|
||
163 | buffer.compact(); |
||
164 | fill(buffer,channel); |
||
165 | buffer.position(0);
|
||
166 | } |
||
167 | } |
||
168 | |||
169 | private int getOffset(int column) { |
||
170 | int offset = 0; |
||
171 | for (int i = 0, ii = column; i < ii; i++) { |
||
172 | offset += fieldLengths[i]; |
||
173 | } |
||
174 | return offset;
|
||
175 | } |
||
176 | |||
177 | private void init() throws IOException { |
||
178 | // create the ByteBuffer
|
||
179 | // if we have a FileChannel, lets map it
|
||
180 | if (channel instanceof FileChannel) { |
||
181 | FileChannel fc = (FileChannel) channel; |
||
182 | buffer = fc.map(FileChannel.MapMode.READ_ONLY,0,fc.size()); |
||
183 | buffer.position((int) fc.position());
|
||
184 | } else {
|
||
185 | // Some other type of channel
|
||
186 | // start with a 8K buffer, should be more than adequate
|
||
187 | int size = 8 * 1024; |
||
188 | // if for some reason its not, resize it
|
||
189 | size = header.getRecordLength() > size ? header.getRecordLength() : size; |
||
190 | buffer = ByteBuffer.allocateDirect(size);
|
||
191 | // fill it and reset
|
||
192 | fill(buffer,channel); |
||
193 | buffer.flip(); |
||
194 | } |
||
195 | |||
196 | // The entire file is in little endian
|
||
197 | buffer.order(ByteOrder.LITTLE_ENDIAN);
|
||
198 | |||
199 | // Set up some buffers and lookups for efficiency
|
||
200 | fieldTypes = new char[header.getNumFields()]; |
||
201 | fieldLengths = new int[header.getNumFields()]; |
||
202 | for (int i = 0, ii = header.getNumFields(); i < ii; i++) { |
||
203 | fieldTypes[i] = header.getFieldType(i); |
||
204 | fieldLengths[i] = header.getFieldLength(i); |
||
205 | } |
||
206 | |||
207 | charBuffer = CharBuffer.allocate(header.getRecordLength() - 1); |
||
208 | Charset chars = Charset.forName("ISO-8859-1"); |
||
209 | decoder = chars.newDecoder(); |
||
210 | |||
211 | row = new Row();
|
||
212 | } |
||
213 | |||
214 | |||
215 | |||
216 | /** Get the header from this file. The header is read upon instantiation.
|
||
217 | * @return The header associated with this file or null if an error occurred.
|
||
218 | */
|
||
219 | public DbaseFileHeaderNIO getHeader() {
|
||
220 | return header;
|
||
221 | } |
||
222 | |||
223 | /** Clean up all resources associated with this reader.<B>Highly recomended.</B>
|
||
224 | * @throws IOException If an error occurs.
|
||
225 | */
|
||
226 | public void close() throws IOException { |
||
227 | if (channel.isOpen()) {
|
||
228 | channel.close(); |
||
229 | } |
||
230 | if (buffer instanceof MappedByteBuffer) { |
||
231 | NIOUtilities.clean(buffer); |
||
232 | } |
||
233 | buffer = null;
|
||
234 | channel = null;
|
||
235 | charBuffer = null;
|
||
236 | decoder = null;
|
||
237 | header = null;
|
||
238 | row = null;
|
||
239 | } |
||
240 | |||
241 | /** Query the reader as to whether there is another record.
|
||
242 | * @return True if more records exist, false otherwise.
|
||
243 | */
|
||
244 | public boolean hasNext() { |
||
245 | return cnt < header.getNumRecords() + 1; |
||
246 | } |
||
247 | |||
248 | /** Get the next record (entry). Will return a new array of values.
|
||
249 | * @throws IOException If an error occurs.
|
||
250 | * @return A new array of values.
|
||
251 | */
|
||
252 | public Object[] readEntry() throws IOException { |
||
253 | return readEntry(new Object[header.getNumFields()]); |
||
254 | } |
||
255 | |||
256 | public Row readRow() throws IOException { |
||
257 | read(); |
||
258 | return row;
|
||
259 | } |
||
260 | |||
261 | /** Skip the next record.
|
||
262 | * @throws IOException If an error occurs.
|
||
263 | */
|
||
264 | public void skip() throws IOException { |
||
265 | boolean foundRecord = false; |
||
266 | while (!foundRecord) {
|
||
267 | |||
268 | bufferCheck(); |
||
269 | |||
270 | // read the deleted flag
|
||
271 | char tempDeleted = (char) buffer.get(); |
||
272 | |||
273 | // skip the next bytes
|
||
274 | buffer.position(buffer.position() + header.getRecordLength() - 1); //the 1 is for the deleted flag just read. |
||
275 | |||
276 | // add the row if it is not deleted.
|
||
277 | if (tempDeleted != '*') { |
||
278 | foundRecord = true;
|
||
279 | } |
||
280 | } |
||
281 | } |
||
282 | |||
283 | /** Copy the next record into the array starting at offset.
|
||
284 | * @param entry Th array to copy into.
|
||
285 | * @param offset The offset to start at
|
||
286 | * @throws IOException If an error occurs.
|
||
287 | * @return The same array passed in.
|
||
288 | */
|
||
289 | public Object[] readEntry(Object[] entry,final int offset) throws IOException { |
||
290 | if (entry.length - offset < header.getNumFields()) {
|
||
291 | throw new ArrayIndexOutOfBoundsException(); |
||
292 | } |
||
293 | |||
294 | read(); |
||
295 | |||
296 | // retrieve the record length
|
||
297 | final int numFields = header.getNumFields(); |
||
298 | |||
299 | int fieldOffset = 0; |
||
300 | for (int j=0; j < numFields; j++){ |
||
301 | entry[j + offset] = readObject(fieldOffset,j); |
||
302 | fieldOffset += fieldLengths[j]; |
||
303 | } |
||
304 | |||
305 | return entry;
|
||
306 | } |
||
307 | |||
308 | /**
|
||
309 | * Transfer, by bytes, the next record to the writer.
|
||
310 | */
|
||
311 | public void transferTo(DbaseFileWriterNIO writer) throws IOException { |
||
312 | bufferCheck(); |
||
313 | buffer.limit(buffer.position() + header.getRecordLength()); |
||
314 | writer.channel.write(buffer); |
||
315 | buffer.limit(buffer.capacity()); |
||
316 | |||
317 | cnt++; |
||
318 | } |
||
319 | |||
320 | private void read() throws IOException { |
||
321 | boolean foundRecord = false; |
||
322 | while (!foundRecord) {
|
||
323 | |||
324 | bufferCheck(); |
||
325 | |||
326 | // read the deleted flag
|
||
327 | char deleted = (char) buffer.get(); |
||
328 | if (deleted == '*') { |
||
329 | continue;
|
||
330 | } |
||
331 | |||
332 | charBuffer.position(0);
|
||
333 | buffer.limit(buffer.position() + header.getRecordLength() - 1);
|
||
334 | decoder.decode(buffer,charBuffer,true);
|
||
335 | buffer.limit(buffer.capacity()); |
||
336 | charBuffer.flip(); |
||
337 | |||
338 | foundRecord = true;
|
||
339 | } |
||
340 | |||
341 | cnt++; |
||
342 | } |
||
343 | |||
344 | /** Copy the next entry into the array.
|
||
345 | * @param entry The array to copy into.
|
||
346 | * @throws IOException If an error occurs.
|
||
347 | * @return The same array passed in.
|
||
348 | */
|
||
349 | public Object[] readEntry(Object[] entry) throws IOException { |
||
350 | return readEntry(entry,0); |
||
351 | } |
||
352 | |||
353 | private Object readObject(final int fieldOffset,final int fieldNum) throws IOException { |
||
354 | final char type = fieldTypes[fieldNum]; |
||
355 | final int fieldLen = fieldLengths[fieldNum]; |
||
356 | Object object = null; |
||
357 | |||
358 | //System.out.println( charBuffer.subSequence(fieldOffset,fieldOffset + fieldLen));
|
||
359 | |||
360 | if(fieldLen > 0) { |
||
361 | |||
362 | switch (type){
|
||
363 | // (L)logical (T,t,F,f,Y,y,N,n)
|
||
364 | case 'l': |
||
365 | case 'L': |
||
366 | switch (charBuffer.charAt(fieldOffset)) {
|
||
367 | |||
368 | case 't': case 'T': case 'Y': case 'y': |
||
369 | object = Boolean.TRUE;
|
||
370 | break;
|
||
371 | case 'f': case 'F': case 'N': case 'n': |
||
372 | object = Boolean.FALSE;
|
||
373 | break;
|
||
374 | default:
|
||
375 | |||
376 | throw new IOException("Unknown logical value : '" + charBuffer.charAt(fieldOffset) + "'"); |
||
377 | } |
||
378 | break;
|
||
379 | // (C)character (String)
|
||
380 | case 'c': |
||
381 | case 'C': |
||
382 | // oh, this seems like a lot of work to parse strings...but,
|
||
383 | // For some reason if zero characters ( (int) char == 0 ) are allowed
|
||
384 | // in these strings, they do not compare correctly later on down the
|
||
385 | // line....
|
||
386 | int start = fieldOffset;
|
||
387 | int end = fieldOffset + fieldLen - 1; |
||
388 | // trim off whitespace and 'zero' chars
|
||
389 | while (start < end) {
|
||
390 | char c = charBuffer.get(start);
|
||
391 | if (c== 0 || Character.isWhitespace(c)) { |
||
392 | start++; |
||
393 | } |
||
394 | else break; |
||
395 | } |
||
396 | while (end > start) {
|
||
397 | char c = charBuffer.get(end);
|
||
398 | if (c == 0 || Character.isWhitespace(c)) { |
||
399 | end--; |
||
400 | } |
||
401 | else break; |
||
402 | } |
||
403 | // set up the new indexes for start and end
|
||
404 | charBuffer.position(start).limit(end + 1);
|
||
405 | String s = charBuffer.toString();
|
||
406 | // this resets the limit...
|
||
407 | charBuffer.clear(); |
||
408 | object = s; |
||
409 | break;
|
||
410 | // (D)date (Date)
|
||
411 | case 'd': |
||
412 | case 'D': |
||
413 | try{
|
||
414 | String tempString = charBuffer.subSequence(fieldOffset,fieldOffset + 4).toString(); |
||
415 | int tempYear = Integer.parseInt(tempString); |
||
416 | tempString = charBuffer.subSequence(fieldOffset + 4,fieldOffset + 6).toString(); |
||
417 | int tempMonth = Integer.parseInt(tempString) - 1; |
||
418 | tempString = charBuffer.subSequence(fieldOffset + 6,fieldOffset + 8).toString(); |
||
419 | int tempDay = Integer.parseInt(tempString); |
||
420 | Calendar cal = Calendar.getInstance(); |
||
421 | cal.clear(); |
||
422 | cal.set(cal.YEAR,tempYear); |
||
423 | cal.set(cal.MONTH, tempMonth); |
||
424 | cal.set(cal.DAY_OF_MONTH, tempDay); |
||
425 | object = cal.getTime(); |
||
426 | } |
||
427 | catch(NumberFormatException nfe){ |
||
428 | //todo: use progresslistener, this isn't a grave error.
|
||
429 | } |
||
430 | break;
|
||
431 | |||
432 | // (F)floating (Double)
|
||
433 | case 'n': |
||
434 | case 'N': |
||
435 | try {
|
||
436 | if (header.getFieldDecimalCount(fieldNum) == 0) { |
||
437 | object = new Integer(numberParser.parseInt(charBuffer, fieldOffset, fieldOffset + fieldLen - 1)); |
||
438 | break;
|
||
439 | } |
||
440 | // else will fall through to the floating point number
|
||
441 | } catch (NumberFormatException e) { |
||
442 | |||
443 | // todo: use progresslistener, this isn't a grave error.
|
||
444 | |||
445 | // don't do this!!! the Double parse will be attemted as we fall
|
||
446 | // through, so no need to create a new Object. -IanS
|
||
447 | //object = new Integer(0);
|
||
448 | |||
449 | // Lets try parsing a long instead...
|
||
450 | try {
|
||
451 | object = new Long(numberParser.parseLong(charBuffer,fieldOffset,fieldOffset + fieldLen - 1)); |
||
452 | break;
|
||
453 | } catch (NumberFormatException e2) { |
||
454 | |||
455 | } |
||
456 | } |
||
457 | |||
458 | case 'f': |
||
459 | case 'F': // floating point number |
||
460 | try {
|
||
461 | |||
462 | |||
463 | object = new Double(numberParser.parseDouble(charBuffer,fieldOffset, fieldOffset + fieldLen - 1)); |
||
464 | } catch (NumberFormatException e) { |
||
465 | // todo: use progresslistener, this isn't a grave error, though it
|
||
466 | // does indicate something is wrong
|
||
467 | |||
468 | // okay, now whatever we got was truly undigestable. Lets go with
|
||
469 | // a zero Double.
|
||
470 | object = new Double(0.0); |
||
471 | } |
||
472 | break;
|
||
473 | default:
|
||
474 | throw new IOException("Invalid field type : " + type); |
||
475 | } |
||
476 | |||
477 | } |
||
478 | return object;
|
||
479 | } |
||
480 | |||
481 | public static void main(String[] args) throws Exception { |
||
482 | /* FileChannel channel = new FileInputStream(args[0]).getChannel();
|
||
483 | DbaseFileReaderNIO reader = new DbaseFileReaderNIO(channel);
|
||
484 | System.out.println(reader.getHeader());
|
||
485 | int r = 0;
|
||
486 | while (reader.hasNext()) {
|
||
487 | System.out.println(++r + "," + java.util.Arrays.asList(reader.readEntry()));
|
||
488 | }
|
||
489 | */
|
||
490 | } |
||
491 | |||
492 | } |