(root)/android/import-contacts
» Revision
36
(compared to revision 89)
: src/org/waxworlds/edam/importcontacts/VCFImporter.java
To get this branch, use:
bzr branch
http://bzr.ed.am/android/import-contacts
« back to all changes in this revision
Viewing changes to src/org/waxworlds/edam/importcontacts/VCFImporter.java
- browse files at revision 36
- compare with another revision
- download diff
- download tarball
- view history from revision 36
- Committer: edam
- Date: 2011-03-19 20:33:09 UTC
- Revision ID: edam@waxworlds.org-20110319203309-5dzfyqrxwk94jtin
- formatting: removed some double-indents on overrunning lines
- updated TODO and NEWS
- rewrote central logic of parser so it makes more sense, looks nicer and has a small optimisation (getting name and params from line only when necessary)
- optimised unnecessary mutliple converting of lines to US-ASCII
- re-wrote line extraction from vcards so that we can lookahead for v3 folded lines
- added support for v3 folded lines
- updated TODO and NEWS
- rewrote central logic of parser so it makes more sense, looks nicer and has a small optimisation (getting name and params from line only when necessary)
- optimised unnecessary mutliple converting of lines to US-ASCII
- re-wrote line extraction from vcards so that we can lookahead for v3 folded lines
- added support for v3 folded lines
- files added:
- default.properties
- gen/org
- gen/org/waxworlds
- gen/org/waxworlds/edam
- gen/org/waxworlds/edam/importcontacts
- src/org
- src/org/waxworlds
- src/org/waxworlds/edam
- files removed:
- gen/am
- gen/am/ed
- gen/am/ed/importcontacts
- src/am
- src/am/ed
- files renamed:
- src/am/ed/importcontacts/ => src/org/waxworlds/edam/importcontacts/
- src/am/ed/importcontacts/VcardImporter.java => src/org/waxworlds/edam/importcontacts/VCFImporter.java
- files modified:
- .classpath
added
removed
src/org/waxworlds/edam/importcontacts/VCFImporter.java
1
1
/*
2
2
* VCFImporter.java
3
3
*
4
* Copyright (C) 2009 to 2011 Tim Marston <tim@ed.am>
4
* Copyright (C) 2009 Tim Marston <edam@waxworlds.org>
5
5
*
6
6
* This file is part of the Import Contacts program (hereafter referred
7
7
* to as "this program"). For more information, see
8
* http://ed.am/dev/android/import-contacts
8
* http://www.waxworlds.org/edam/software/android/import-contacts
9
9
*
10
10
* This program is free software: you can redistribute it and/or modify
11
11
* it under the terms of the GNU General Public License as published by
21
21
* along with this program. If not, see <http://www.gnu.org/licenses/>.
22
22
*/
23
23
24
package am.ed.importcontacts;
24
package org.waxworlds.edam.importcontacts;
25
25
26
26
import java.io.BufferedReader;
27
27
import java.io.File;
32
32
import java.io.IOException;
33
33
import java.io.UnsupportedEncodingException;
34
34
import java.nio.ByteBuffer;
35
import java.util.ArrayList;
36
35
import java.util.Arrays;
37
import java.util.HashMap;
38
36
import java.util.HashSet;
39
37
import java.util.Iterator;
40
38
import java.util.List;
41
import java.util.Locale;
42
import java.util.NoSuchElementException;
43
39
import java.util.Set;
44
40
import java.util.Vector;
45
41
import java.util.regex.Matcher;
46
42
import java.util.regex.Pattern;
43
import java.util.NoSuchElementException;
44
import java.lang.UnsupportedOperationException;
47
45
48
import android.annotation.SuppressLint;
49
46
import android.content.SharedPreferences;
47
import android.provider.Contacts;
48
import android.provider.Contacts.PhonesColumns;
50
49
51
public class VcardImporter extends Importer
50
public class VCFImporter extends Importer
52
51
{
53
private int _vcard_count = 0;
52
private int _vCardCount = 0;
54
53
private int _progress = 0;
55
54
56
public VcardImporter( Doit doit )
55
public VCFImporter( Doit doit )
57
56
{
58
57
super( doit );
59
58
}
60
59
61
@SuppressLint( "SdCardPath" )
62
60
@Override
63
61
protected void onImport() throws AbortImportException
64
62
{
83
81
// get files
84
82
class VCardFilter implements FilenameFilter {
85
83
public boolean accept( File dir, String name ) {
86
return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
84
return name.toLowerCase().endsWith( ".vcf" );
87
85
}
88
86
}
89
87
files = file.listFiles( new VCardFilter() );
111
109
countVCardFile( files[ i ] );
112
110
setTmpProgress( i );
113
111
}
114
setProgressMax( _vcard_count ); // will also update tmp progress
112
setProgressMax( _vCardCount ); // will also update tmp progress
115
113
116
114
// import them
117
115
setProgress( 0 );
118
116
for( int i = 0; i < files.length; i++ )
119
117
importVCardFile( files[ i ] );
120
setProgress( _vcard_count );
121
118
}
122
119
123
120
private void countVCardFile( File file ) throws AbortImportException
130
127
131
128
// read
132
129
String line;
133
boolean in_vcard = false;
130
boolean inVCard = false;
134
131
while( ( line = reader.readLine() ) != null )
135
132
{
136
if( !in_vcard )
137
{
133
if( !inVCard ) {
138
134
// look for vcard beginning
139
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
140
in_vcard = true;
141
_vcard_count++;
142
}
143
// check for vMsg files
144
else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
145
showError( getText( R.string.error_vcf_vmsgfile )
146
+ file.getName() );
135
if( line.matches( "^BEGIN:VCARD" ) ) {
136
inVCard = true;
137
_vCardCount++;
147
138
}
148
139
}
149
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
150
in_vcard = false;
140
else if( line.matches( "^END:VCARD" ) )
141
inVCard = false;
151
142
}
152
143
153
144
}
176
167
FileInputStream istream = new FileInputStream( file );
177
168
byte[] content = new byte[ (int)file.length() ];
178
169
istream.read( content );
179
istream = null;
180
170
181
171
// import
182
172
importVCardFileContent( content, file.getName() );
183
173
}
184
catch( OutOfMemoryError e ) {
185
showError( R.string.error_outofmemory );
186
}
187
174
catch( FileNotFoundException e ) {
188
175
showError( getText( R.string.error_filenotfound ) +
189
176
file.getName() );
197
184
throws AbortImportException
198
185
{
199
186
// go through lines
200
Vcard vcard = null;
201
int vcard_start_line = 0;
187
VCard vCard = null;
202
188
ContentLineIterator cli = new ContentLineIterator( content );
203
189
while( cli.hasNext() )
204
190
{
205
ContentLine content_line = cli.next();
206
207
// get a US-ASCII version of the string, for processing
208
String line = content_line.getUsAsciiLine();
209
210
if( vcard == null ) {
191
ByteBuffer buffer = cli.next();
192
193
// get a US-ASCII version of the line for processing
194
String line;
195
try {
196
line = new String( buffer.array(), buffer.position(),
197
buffer.limit() - buffer.position(), "US-ASCII" );
198
}
199
catch( UnsupportedEncodingException e ) {
200
// we know US-ASCII is supported, so appease the compiler...
201
line = "";
202
}
203
204
if( vCard == null ) {
211
205
// look for vcard beginning
212
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
213
setProgress( _progress++ );
214
vcard = new Vcard();
215
vcard_start_line = cli.getLineNumber();
206
if( line.matches( "^BEGIN:VCARD" ) ) {
207
setProgress( ++_progress );
208
vCard = new VCard();
216
209
}
217
210
}
218
211
else {
219
212
// look for vcard content or ending
220
if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
213
if( line.matches( "^END:VCARD" ) )
221
214
{
222
// finalise the vcard/contact
215
// store vcard and do away with it
223
216
try {
224
vcard.finaliseVcard();
225
226
// pass the finalised contact to the importer
227
importContact( vcard );
228
}
229
catch( Vcard.ParseException e ) {
230
if( !showContinue(
231
getText( R.string.error_vcf_parse ).toString()
232
+ fileName +
233
getText( R.string.error_vcf_parse_line ).toString()
234
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
235
{
236
finish( ACTION_ABORT );
237
}
238
else
239
skipContact();
240
}
241
catch( ContactData.ContactNotIdentifiableException e ) {
242
if( !showContinue(
243
getText( R.string.error_vcf_parse ).toString()
244
+ fileName +
245
getText( R.string.error_vcf_parse_line ).toString()
246
+ vcard_start_line + ":\n" + getText(
247
R.string.error_vcf_notenoughinfo ).toString()
248
) )
249
{
250
finish( ACTION_ABORT );
251
}
252
else
253
skipContact();
254
}
255
256
// discard this vcard
257
vcard = null;
217
vCard.finaliseParsing();
218
importContact( vCard );
219
}
220
catch( VCard.ParseException e ) {
221
skipContact();
222
if( !showContinue(
223
getText( R.string.error_vcf_parse ).toString()
224
+ fileName + "\n" + e.getMessage() ) )
225
{
226
finish( ACTION_ABORT );
227
}
228
}
229
catch( VCard.SkipContactException e ) {
230
skipContact();
231
// do nothing
232
}
233
vCard = null;
258
234
}
259
235
else
260
236
{
261
237
// try giving the line to the vcard
262
238
try {
263
vcard.parseLine( content_line );
239
vCard.parseLine( buffer, line,
240
cli.doesNextLineLookFolded() );
264
241
}
265
catch( Vcard.ParseException e ) {
242
catch( VCard.ParseException e ) {
266
243
skipContact();
267
244
if( !showContinue(
268
245
getText( R.string.error_vcf_parse ).toString()
269
+ fileName +
270
getText( R.string.error_vcf_parse_line ).toString()
271
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
246
+ fileName + "\n" + e.getMessage() ) )
272
247
{
273
248
finish( ACTION_ABORT );
274
249
}
276
251
// although we're continuing, we still need to abort
277
252
// this vCard. Further lines will be ignored until we
278
253
// get to another BEGIN:VCARD line.
279
vcard = null;
254
vCard = null;
280
255
}
281
catch( Vcard.SkipImportException e ) {
256
catch( VCard.SkipContactException e ) {
282
257
skipContact();
283
258
// abort this vCard. Further lines will be ignored until
284
259
// we get to another BEGIN:VCARD line.
285
vcard = null;
260
vCard = null;
286
261
}
287
262
}
288
263
}
289
264
}
290
265
}
291
266
292
class ContentLine
293
{
294
private ByteBuffer _buffer;
295
private boolean _folded_next;
296
private String _line;
297
298
public ContentLine( ByteBuffer buffer, boolean folded_next )
299
{
300
_buffer = buffer;
301
_folded_next = folded_next;
302
_line = null;
303
}
304
305
public ByteBuffer getBuffer()
306
{
307
return _buffer;
308
}
309
310
public boolean doesNextLineLookFolded()
311
{
312
return _folded_next;
313
}
314
315
public String getUsAsciiLine()
316
{
317
// generated line and cache it
318
if( _line == null ) {
319
try {
320
_line = new String( _buffer.array(), _buffer.position(),
321
_buffer.limit() - _buffer.position(), "US-ASCII" );
322
}
323
catch( UnsupportedEncodingException e ) {
324
// we know US-ASCII *is* supported, so appease the
325
// compiler...
326
}
327
}
328
329
// return cached line
330
return _line;
331
}
332
}
333
334
class ContentLineIterator implements Iterator< ContentLine >
267
class ContentLineIterator implements Iterator< ByteBuffer >
335
268
{
336
269
protected byte[] _content = null;
337
270
protected int _pos = 0;
338
protected int _line = 0;
339
271
340
272
public ContentLineIterator( byte[] content )
341
273
{
349
281
}
350
282
351
283
@Override
352
public ContentLine next()
284
public ByteBuffer next()
353
285
{
354
286
int initial_pos = _pos;
355
287
361
293
int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
362
294
_pos > initial_pos )? _pos - 1 : _pos;
363
295
_pos++;
364
_line++;
365
return new ContentLine(
366
ByteBuffer.wrap( _content, initial_pos,
367
to - initial_pos ),
368
doesNextLineLookFolded() );
296
return ByteBuffer.wrap( _content, initial_pos,
297
to - initial_pos );
369
298
}
370
299
371
300
// we didn't find one, but were there bytes left?
372
301
if( _pos != initial_pos ) {
373
302
int to = _pos;
374
303
_pos++;
375
_line++;
376
return new ContentLine(
377
ByteBuffer.wrap( _content, initial_pos,
378
to - initial_pos ),
379
doesNextLineLookFolded() );
304
return ByteBuffer.wrap( _content, initial_pos,
305
to - initial_pos );
380
306
}
381
307
382
308
// no bytes left
394
320
* onto the end of this one?
395
321
* @return
396
322
*/
397
private boolean doesNextLineLookFolded()
323
public boolean doesNextLineLookFolded()
398
324
{
399
325
return _pos > 0 && _pos < _content.length &&
400
_content[ _pos - 1 ] == '\n' &&
401
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
402
}
403
404
public int getLineNumber()
405
{
406
return _line;
326
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
407
327
}
408
328
}
409
329
410
private class Vcard extends ContactData
330
private class VCard extends ContactData
411
331
{
412
332
private final static int NAMELEVEL_NONE = 0;
413
private final static int NAMELEVEL_N = 1;
333
private final static int NAMELEVEL_ORG = 1;
414
334
private final static int NAMELEVEL_FN = 2;
415
416
private final static int MULTILINE_NONE = 0;
417
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
418
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
419
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
335
private final static int NAMELEVEL_N = 3;
420
336
421
337
private String _version = null;
422
private Vector< ContentLine > _content_lines = null;
338
private Vector< ByteBuffer > _buffers = null;
423
339
private int _name_level = NAMELEVEL_NONE;
424
private int _parser_multiline_state = MULTILINE_NONE;
340
private boolean _parser_in_encoded_multiline = false;
341
private boolean _parser_in_folded_multiline = false;
425
342
private String _parser_current_name_and_params = null;
426
343
private String _parser_buffered_value_so_far = "";
427
private String _cached_organisation = null;
428
private String _cached_title = null;
429
344
430
345
protected class UnencodeResult
431
346
{
461
376
462
377
public ParseException( int res )
463
378
{
464
super( VcardImporter.this.getText( res ).toString() );
379
super( VCFImporter.this.getText( res ).toString() );
465
380
}
466
381
}
467
382
468
383
@SuppressWarnings("serial")
469
protected class SkipImportException extends Exception { }
384
protected class SkipContactException extends Exception { }
470
385
471
private String extractCollonPartFromLine( ContentLine content_line,
472
boolean former )
386
private String extractCollonPartFromLine( ByteBuffer buffer,
387
String line, boolean former )
473
388
{
474
389
String ret = null;
475
390
391
// get a US-ASCII version of the line for processing, unless we were
392
// supplied with one
393
if( line == null ) {
394
try {
395
line = new String( buffer.array(), buffer.position(),
396
buffer.limit() - buffer.position(), "US-ASCII" );
397
}
398
catch( UnsupportedEncodingException e ) {
399
// we know US-ASCII is supported, so appease the compiler...
400
line = "";
401
}
402
}
403
476
404
// split line into name and value parts and check to make sure we
477
405
// only got 2 parts and that the first part is not zero in length
478
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
406
String[] parts = line.split( ":", 2 );
479
407
if( parts.length == 2 && parts[ 0 ].length() > 0 )
480
408
ret = parts[ former? 0 : 1 ];
481
409
482
410
return ret;
483
411
}
484
412
485
private String extractNameAndParamsFromLine( ContentLine content_line )
486
{
487
return extractCollonPartFromLine( content_line, true ).trim();
488
}
489
490
private String extractValueFromLine( ContentLine content_line )
491
{
492
return extractCollonPartFromLine( content_line, false );
493
}
494
495
public void parseLine( ContentLine content_line )
496
throws ParseException, SkipImportException,
413
private String extractNameAndParamsFromLine( ByteBuffer buffer,
414
String line )
415
{
416
return extractCollonPartFromLine( buffer, line, true );
417
}
418
419
private String extractValueFromLine( ByteBuffer buffer, String line )
420
{
421
return extractCollonPartFromLine( buffer, line, false );
422
}
423
424
public void parseLine( ByteBuffer buffer, String line,
425
boolean next_line_looks_folded )
426
throws ParseException, SkipContactException,
497
427
AbortImportException
498
428
{
499
429
// do we have a version yet?
501
431
{
502
432
// tentatively get name and params from line
503
433
String name_and_params =
504
extractNameAndParamsFromLine( content_line );
434
extractNameAndParamsFromLine( buffer, line );
505
435
506
436
// is it a version line?
507
437
if( name_and_params != null &&
508
name_and_params.equalsIgnoreCase( "VERSION" ) )
438
name_and_params.equals( "VERSION" ) )
509
439
{
510
440
// yes, get it!
511
String value = extractValueFromLine( content_line ).trim();
441
String value = extractValueFromLine( buffer, line );
512
442
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
443
throw new ParseException( R.string.error_vcf_version );
514
444
_version = value;
515
445
516
446
// parse any buffers we've been accumulating while we waited
517
447
// for a version
518
if( _content_lines != null )
519
for( int i = 0; i < _content_lines.size(); i++ )
520
parseLine( _content_lines.get( i ) );
521
_content_lines = null;
448
if( _buffers != null )
449
for( int i = 0; i < _buffers.size(); i++ )
450
parseLine( _buffers.get( i ), null,
451
i + 1 < _buffers.size() &&
452
_buffers.get( i + 1 ).hasRemaining() &&
453
_buffers.get( i + 1 ).get(
454
_buffers.get( i + 1 ).position() ) == ' ' );
455
_buffers = null;
522
456
}
523
457
else
524
458
{
525
459
// no, so stash this line till we get a version
526
if( _content_lines == null )
527
_content_lines = new Vector< ContentLine >();
528
_content_lines.add( content_line );
460
if( _buffers == null )
461
_buffers = new Vector< ByteBuffer >();
462
_buffers.add( buffer );
529
463
}
530
464
}
531
465
else
532
466
{
533
467
// name and params and the position in the buffer where the
534
// "value" part of the line starts
468
// "value" part of the line start
535
469
String name_and_params;
536
470
int pos;
537
471
538
if( _parser_multiline_state != MULTILINE_NONE )
472
if( _parser_in_encoded_multiline ||
473
_parser_in_folded_multiline )
539
474
{
540
475
// if we're currently in a multi-line value, use the stored
541
476
// property name and parameters
542
477
name_and_params = _parser_current_name_and_params;
543
478
544
// skip some initial line characters, depending on the type
545
// of multi-line we're handling
546
pos = content_line.getBuffer().position();
547
switch( _parser_multiline_state )
548
{
549
case MULTILINE_FOLDED:
479
pos = buffer.position();
480
481
// for folded multi-lines, skip the single space at the
482
// start of the next line
483
if( _parser_in_folded_multiline )
550
484
pos++;
551
break;
552
case MULTILINE_ENCODED:
553
while( pos < content_line.getBuffer().limit() && (
554
content_line.getBuffer().get( pos ) == ' ' ||
555
content_line.getBuffer().get( pos ) == '\t' ) )
485
486
// else, this must be an encoded multi-line, so skip any
487
// whitespace we find at the start of the next line
488
else
489
while( pos < buffer.limit() && (
490
buffer.get( pos ) == ' ' ||
491
buffer.get( pos ) == '\t' ) )
556
492
{
557
493
pos++;
558
494
}
559
break;
560
default:
561
// do nothing
562
}
563
564
// take us out of multi-line so that we can re-detect that
565
// this line is a multi-line or not
566
_parser_multiline_state = MULTILINE_NONE;
567
495
}
568
496
else
569
497
{
570
// skip empty lines
571
if( content_line.getUsAsciiLine().trim().length() == 0 )
572
return;
573
574
498
// get name and params from line, and since we're not
575
499
// parsing a subsequent line in a multi-line, this should
576
500
// not fail, or it's an error
577
501
name_and_params =
578
extractNameAndParamsFromLine( content_line );
502
extractNameAndParamsFromLine( buffer, line );
579
503
if( name_and_params == null )
580
504
throw new ParseException(
581
505
R.string.error_vcf_malformed );
582
506
583
507
// calculate how many chars to skip from beginning of line
584
508
// so we skip the property "name:" part
585
pos = content_line.getBuffer().position() +
586
name_and_params.length() + 1;
509
pos = buffer.position() + name_and_params.length() + 1;
587
510
588
511
// reset the saved multi-line state
589
512
_parser_current_name_and_params = name_and_params;
592
515
593
516
// get value from buffer, as raw bytes
594
517
ByteBuffer value;
595
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
596
content_line.getBuffer().limit() - pos );
518
value = ByteBuffer.wrap( buffer.array(), pos,
519
buffer.limit() - pos );
597
520
598
521
// get parameter parts
599
522
String[] name_param_parts = name_and_params.split( ";", -1 );
600
523
for( int i = 0; i < name_param_parts.length; i++ )
601
524
name_param_parts[ i ] = name_param_parts[ i ].trim();
602
525
603
// determine whether we care about this entry
604
final HashSet< String > interesting_fields =
605
new HashSet< String >( Arrays.asList( new String[] { "N",
606
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
607
) );
608
boolean is_interesting_field =
609
interesting_fields.contains(
610
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
611
612
526
// parse encoding parameter
613
527
String encoding = checkParam( name_param_parts, "ENCODING" );
614
if( encoding != null )
615
encoding = encoding.toUpperCase( Locale.US );
616
if( is_interesting_field && encoding != null &&
617
!encoding.equalsIgnoreCase( "8BIT" ) &&
618
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
619
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
528
if( encoding != null ) encoding = encoding.toUpperCase();
529
if( encoding != null && !encoding.equals( "8BIT" ) &&
530
!encoding.equals( "QUOTED-PRINTABLE" ) )
531
//&& !encoding.equals( "BASE64" ) )
620
532
{
621
533
throw new ParseException( R.string.error_vcf_encoding );
622
534
}
623
535
624
536
// parse charset parameter
625
537
String charset = checkParam( name_param_parts, "CHARSET" );
626
if( charset != null )
627
charset = charset.toUpperCase( Locale.US );
628
if( charset != null &&
629
!charset.equalsIgnoreCase( "US-ASCII" ) &&
630
!charset.equalsIgnoreCase( "ASCII" ) &&
631
!charset.equalsIgnoreCase( "UTF-8" ) )
538
if( charset != null ) charset = charset.toUpperCase();
539
if( charset != null && !charset.equals( "US-ASCII" ) &&
540
!charset.equals( "ASCII" ) &&
541
!charset.equals( "UTF-8" ) )
632
542
{
633
543
throw new ParseException( R.string.error_vcf_charset );
634
544
}
636
546
// do unencoding (or default to a fake unencoding result with
637
547
// the raw string)
638
548
UnencodeResult unencoding_result = null;
639
if( encoding != null &&
640
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
641
{
549
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
642
550
unencoding_result = unencodeQuotedPrintable( value );
643
}
644
// else if( encoding != null &&
645
// encoding.equalsIgnoreCase( "BASE64" ) )
646
// {
551
// else if( encoding != null && encoding.equals( "BASE64" ) )
647
552
// unencoding_result = unencodeBase64( props[ 1 ], charset );
648
// }
649
553
if( unencoding_result != null ) {
650
554
value = unencoding_result.getBuffer();
651
if( unencoding_result.isAnotherLineRequired() )
652
_parser_multiline_state = MULTILINE_ENCODED;
555
_parser_in_encoded_multiline =
556
unencoding_result.isAnotherLineRequired();
653
557
}
654
558
655
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
656
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
657
if( ( charset == null && _version.equals( "2.1" ) ) ||
658
( charset != null && (
659
charset.equalsIgnoreCase( "ASCII" ) ||
660
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
661
{
559
// convert 8-bit ASCII charset to US-ASCII
560
if( charset == null || charset.equals( "ASCII" ) ) {
662
561
value = transcodeAsciiToUtf8( value );
562
charset = "UTF-8";
663
563
}
664
564
665
// process charset (value is now in UTF-8)
565
// process charset
666
566
String string_value;
667
567
try {
668
568
string_value = new String( value.array(), value.position(),
669
value.limit() - value.position(), "UTF-8" );
569
value.limit() - value.position(), charset );
670
570
} catch( UnsupportedEncodingException e ) {
671
571
throw new ParseException( R.string.error_vcf_charset );
672
572
}
673
573
674
// for some entries that have semicolon-separated value parts,
675
// check to see if the value ends in an escape character, which
676
// indicates that we have a multi-line value
677
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
678
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
679
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
680
doesStringEndInAnEscapeChar( string_value ) )
681
{
682
_parser_multiline_state = MULTILINE_ESCAPED;
683
string_value = string_value.substring( 0,
684
string_value.length() - 1 );
685
}
686
687
// if we know we're not in an encoding-based multi-line, check
688
// to see if we're in a folded multi-line
689
if( _parser_multiline_state == MULTILINE_NONE &&
690
content_line.doesNextLineLookFolded() )
691
{
692
_parser_multiline_state = MULTILINE_FOLDED;
693
}
694
695
// handle multi-lines by buffering them and parsing them when we
696
// are processing the last line in a multi-line sequence
697
if( _parser_multiline_state != MULTILINE_NONE ) {
574
// now we know whether we're in an encoding multi-line,
575
// determine if we're in a v3 folded multi-line or not
576
_parser_in_folded_multiline = !_parser_in_encoded_multiline &&
577
_version.equals( "3.0" ) && next_line_looks_folded;
578
579
// handle multi-line requests
580
if( _parser_in_encoded_multiline ||
581
_parser_in_folded_multiline )
582
{
698
583
_parser_buffered_value_so_far += string_value;
699
584
return;
700
585
}
586
587
// add on buffered multi-line content
701
588
String complete_value =
702
( _parser_buffered_value_so_far + string_value ).trim();
589
_parser_buffered_value_so_far + string_value;
703
590
704
591
// ignore empty values
705
592
if( complete_value.length() < 1 ) return;
706
593
707
594
// parse some properties
708
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
595
if( name_param_parts[ 0 ].equals( "N" ) )
709
596
parseN( name_param_parts, complete_value );
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
597
else if( name_param_parts[ 0 ].equals( "FN" ) )
711
598
parseFN( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
599
else if( name_param_parts[ 0 ].equals( "ORG" ) )
713
600
parseORG( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
715
parseTITLE( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
601
else if( name_param_parts[ 0 ].equals( "TEL" ) )
717
602
parseTEL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
603
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
719
604
parseEMAIL( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
721
parseADR( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
723
parseLABEL( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
725
parseNOTE( name_param_parts, complete_value );
726
}
727
}
728
729
private boolean doesStringEndInAnEscapeChar( String string )
730
{
731
// count the number of backslashes at the end of the string
732
int count = 0;
733
for( int a = string.length() - 1; a >= 0; a-- )
734
if( string.charAt( a ) == '\\' )
735
count++;
736
else
737
break;
738
739
// if there are an even number of backslashes then the final one
740
// doesn't count
741
return ( count & 1 ) == 1;
742
}
743
744
private String[] splitValueByCharacter( String value, char character )
745
{
746
// split string in to parts by specified character
747
ArrayList< String > parts = new ArrayList< String >(
748
Arrays.asList( value.split( "" + character ) ) );
749
750
// go through parts
751
for( int a = 0; a < parts.size(); a++ )
752
{
753
String str = parts.get( a );
754
755
// look for parts that end in an escape character, but ignore
756
// the final part. We've already detected escape chars at the
757
// end of the final part in parseLine() and handled multi-lines
758
// accordingly.
759
if( a < parts.size() - 1 &&
760
doesStringEndInAnEscapeChar( str ) )
761
{
762
// append the escaped character, join the next part to this
763
// part and remove the next part
764
parts.set( a, str.substring( 0, str.length() - 1 ) +
765
character + parts.get( a + 1 ) );
766
parts.remove( a + 1 );
767
768
// re-visit this part
769
a--;
770
continue;
771
}
772
773
// trim and replace string
774
str = str.trim();
775
parts.set( a, str );
776
}
777
778
String[] ret = new String[ parts.size() ];
779
return parts.toArray( ret );
780
}
781
782
private String unescapeValue( String value )
783
{
784
StringBuilder ret = new StringBuilder( value.length() );
785
boolean in_escape = false;
786
for( int a = 0; a < value.length(); a++ )
787
{
788
int c = value.codePointAt( a );
789
790
// process a normal character
791
if( !in_escape ) {
792
if( c == '\\' )
793
in_escape = true;
794
else
795
ret.append( Character.toChars( c ) );
796
continue;
797
}
798
799
// process an escape sequence
800
in_escape = false;
801
switch( c )
802
{
803
case 'T':
804
case 't':
805
// add tab (invalid/non-standard, but accepted)
806
ret.append( '\t' );
807
break;
808
case 'N':
809
case 'n':
810
// add newline
811
ret.append( '\n' );
812
break;
813
case '\\':
814
case ',':
815
case ';':
816
// add escaped character
817
ret.append( Character.toChars( c ) );
818
break;
819
default:
820
// unknown escape sequence, so add it unescaped
821
// (invalid/non-standard, but accepted)
822
ret.append( "\\" );
823
ret.append( Character.toChars( c ) );
824
break;
825
}
826
}
827
828
return ret.toString();
605
}
829
606
}
830
607
831
608
private void parseN( String[] params, String value )
609
throws ParseException, SkipContactException,
610
AbortImportException
832
611
{
833
612
// already got a better name?
834
613
if( _name_level >= NAMELEVEL_N ) return;
835
614
836
615
// get name parts
837
String[] name_parts = splitValueByCharacter( value, ';' );
616
String[] name_parts = value.split( ";" );
617
for( int i = 0; i < name_parts.length; i++ )
618
name_parts[ i ] = name_parts[ i ].trim();
838
619
839
620
// build name
840
621
value = "";
841
final int[] part_order = { 3, 1, 2, 0, 4 };
842
for( int a = 0; a < part_order.length; a++ )
843
if( name_parts.length > part_order[ a ] &&
844
name_parts[ part_order[ a ] ].length() > 0 )
845
{
846
// split this part in to it's comma-separated bits
847
String[] name_part_parts = splitValueByCharacter(
848
name_parts[ part_order[ a ] ], ',' );
849
for( int b = 0; b < name_part_parts.length; b++ )
850
if( name_part_parts[ b ].length() > 0 )
851
{
852
if( value.length() > 0 ) value += " ";
853
value += name_part_parts[ b ];
854
}
855
}
622
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
623
value += name_parts[ 1 ];
624
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
625
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
856
626
857
627
// set name
858
setName( unescapeValue( value ) );
628
setName( value );
859
629
_name_level = NAMELEVEL_N;
630
631
// check now to see if we need to import this contact (to avoid
632
// parsing the rest of the vCard unnecessarily)
633
if( !isImportRequired( getName() ) )
634
throw new SkipContactException();
860
635
}
861
636
862
637
private void parseFN( String[] params, String value )
638
throws ParseException, SkipContactException
863
639
{
864
640
// already got a better name?
865
641
if( _name_level >= NAMELEVEL_FN ) return;
866
642
867
643
// set name
868
setName( unescapeValue( value ) );
644
setName( value );
869
645
_name_level = NAMELEVEL_FN;
870
646
}
871
647
872
648
private void parseORG( String[] params, String value )
649
throws ParseException, SkipContactException
873
650
{
651
// already got a better name?
652
if( _name_level >= NAMELEVEL_ORG ) return;
653
874
654
// get org parts
875
String[] org_parts = splitValueByCharacter( value, ';' );
876
if( org_parts == null || org_parts.length < 1 ) return;
877
878
// build organisation name
879
StringBuilder builder = new StringBuilder(
880
String.valueOf( org_parts[ 0 ] ) );
881
for( int a = 1; a < org_parts.length; a++ )
882
builder.append( ", " ).append( org_parts[ a ] );
883
String organisation = unescapeValue( builder.toString() );
884
885
// set organisation name (using a title we've previously found)
886
addOrganisation( organisation, _cached_title, true );
887
888
// if we've not previously found a title, store this organisation
889
// name (we'll need it when we find a title to update the
890
// organisation, by name), else if we *have* previously found a
891
// title, clear it (since we just used it)
892
if( _cached_title == null )
893
_cached_organisation = organisation;
894
else
895
_cached_title = null;
896
}
897
898
private void parseTITLE( String[] params, String value )
899
{
900
value = unescapeValue( value );
901
902
// if we previously had an organisation, look it up and append this
903
// title to it
904
if( _cached_organisation != null && hasOrganisations() ) {
905
HashMap< String, ExtraDetail > datas = getOrganisations();
906
ExtraDetail detail = datas.get( _cached_organisation );
907
if( detail != null )
908
detail.setExtra( value );
909
}
910
911
// same as when handling organisation, if we've not previously found
912
// an organisation we store this title, else we clear it (since we
913
// just appended this title to it)
914
if( _cached_organisation == null )
915
_cached_title = value;
916
else
917
_cached_organisation = null;
655
String[] org_parts = value.split( ";" );
656
for( int i = 0; i < org_parts.length; i++ )
657
org_parts[ i ] = org_parts[ i ].trim();
658
659
// build name
660
if( org_parts.length > 1 && org_parts[ 0 ].length() == 0 )
661
value = org_parts[ 1 ];
662
else
663
value = org_parts[ 0 ];
664
665
// set name
666
setName( value );
667
_name_level = NAMELEVEL_ORG;
918
668
}
919
669
920
670
private void parseTEL( String[] params, String value )
671
throws ParseException
921
672
{
922
673
if( value.length() == 0 ) return;
923
674
926
677
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
927
678
928
679
// here's the logic...
929
boolean is_preferred = types.contains( "PREF" );
930
int type;
680
boolean preferred = types.contains( "PREF" );
681
int type = PhonesColumns.TYPE_MOBILE;
682
if( types.contains( "VOICE" ) )
683
if( types.contains( "WORK" ) )
684
type = PhonesColumns.TYPE_WORK;
685
else
686
type = PhonesColumns.TYPE_HOME;
687
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
688
type = PhonesColumns.TYPE_MOBILE;
931
689
if( types.contains( "FAX" ) )
932
690
if( types.contains( "HOME" ) )
933
type = TYPE_FAX_HOME;
691
type = PhonesColumns.TYPE_FAX_HOME;
934
692
else
935
type = TYPE_FAX_WORK;
936
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
937
type = TYPE_MOBILE;
938
else if( types.contains( "PAGER" ) )
939
type = TYPE_PAGER;
940
else if( types.contains( "WORK" ) )
941
type = TYPE_WORK;
942
else
943
type = TYPE_HOME;
693
type = PhonesColumns.TYPE_FAX_WORK;
694
if( types.contains( "PAGER" ) )
695
type = PhonesColumns.TYPE_PAGER;
944
696
945
697
// add phone number
946
addNumber( value, type, is_preferred );
698
addPhone( value, type, preferred );
947
699
}
948
700
949
701
public void parseEMAIL( String[] params, String value )
702
throws ParseException
950
703
{
951
704
if( value.length() == 0 ) return;
952
705
953
706
Set< String > types = extractTypes( params, Arrays.asList(
954
707
"PREF", "WORK", "HOME", "INTERNET" ) );
955
708
956
// add email address
957
boolean is_preferred = types.contains( "PREF" );
958
int type;
959
if( types.contains( "WORK" ) )
960
type = TYPE_WORK;
961
else
962
type = TYPE_HOME;
963
964
addEmail( unescapeValue( value ), type, is_preferred );
965
}
966
967
private void parseADR( String[] params, String value )
968
{
969
// get address parts
970
String[] adr_parts = splitValueByCharacter( value, ';' );
971
972
// build address
973
value = "";
974
for( int a = 0; a < adr_parts.length; a++ )
975
if( adr_parts[ a ].length() > 0 )
976
{
977
// version 3.0 vCards allow further splitting by comma
978
if( _version.equals( "3.0" ) )
979
{
980
// split this part in to it's comma-separated bits and
981
// add them on individual lines
982
String[] adr_part_parts =
983
splitValueByCharacter( adr_parts[ a ], ',' );
984
for( int b = 0; b < adr_part_parts.length; b++ )
985
if( adr_part_parts[ b ].length() > 0 )
986
{
987
if( value.length() > 0 ) value += "\n";
988
value += adr_part_parts[ b ];
989
}
990
}
991
else
992
{
993
// add this part on an individual line
994
if( value.length() > 0 ) value += "\n";
995
value += adr_parts[ a ];
996
}
997
}
998
999
Set< String > types = extractTypes( params, Arrays.asList(
1000
"PREF", "WORK", "HOME" ) );
1001
1002
// add address
1003
int type;
1004
if( types.contains( "WORK" ) )
1005
type = TYPE_WORK;
1006
else
1007
type = TYPE_HOME;
1008
1009
addAddress( unescapeValue( value ), type );
1010
}
1011
1012
private void parseLABEL( String[] params, String value )
1013
{
1014
Set< String > types = extractTypes( params, Arrays.asList(
1015
"PREF", "WORK", "HOME" ) );
1016
1017
// add address
1018
int type;
1019
if( types.contains( "WORK" ) )
1020
type = TYPE_WORK;
1021
else
1022
type = TYPE_HOME;
1023
1024
addAddress( unescapeValue( value ), type );
1025
}
1026
1027
private void parseNOTE( String[] params, String value )
1028
{
1029
addNote( unescapeValue( value ) );
1030
}
1031
1032
public void finaliseVcard()
1033
throws ParseException, ContactNotIdentifiableException
709
// here's the logic...
710
boolean preferred = types.contains( "PREF" );
711
if( types.contains( "WORK" ) )
712
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
713
else
714
addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
715
}
716
717
public void finaliseParsing()
718
throws ParseException, SkipContactException,
719
AbortImportException
1034
720
{
1035
721
// missing version (and data is present)
1036
if( _version == null && _content_lines != null )
722
if( _version == null && _buffers != null )
1037
723
throw new ParseException( R.string.error_vcf_malformed );
1038
724
1039
// finalise the parent class
1040
finalise();
725
// missing name properties?
726
if( _name_level == NAMELEVEL_NONE )
727
throw new ParseException( R.string.error_vcf_noname );
728
729
// check if we should import this one? If we've already got an 'N'-
730
// type name, this will already have been done by parseN() so we
731
// mustn't do this here (or it could prompt twice!)
732
if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
733
throw new SkipContactException();
1041
734
}
1042
735
1043
/**
1044
* Amongst the params, find the value of the first, only, of any with
1045
* the specified name
1046
* @param params
1047
* @param name
1048
* @return a value, or null
1049
*/
1050
736
private String checkParam( String[] params, String name )
1051
737
{
1052
String[] res = checkParams( params, name );
1053
return res.length > 0? res[ 0 ] : null;
1054
}
1055
1056
/**
1057
* Amongst the params, find the values of any with the specified name
1058
* @param params
1059
* @param name
1060
* @return an array of values, or null
1061
*/
1062
private String[] checkParams( String[] params, String name )
1063
{
1064
HashSet< String > ret = new HashSet< String >();
1065
1066
738
Pattern p = Pattern.compile(
1067
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1068
Pattern.CASE_INSENSITIVE );
739
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1069
740
for( int i = 0; i < params.length; i++ ) {
1070
741
Matcher m = p.matcher( params[ i ] );
1071
742
if( m.matches() )
1072
ret.add( m.group( 2 ) );
743
return m.group( 2 );
1073
744
}
1074
1075
return (String[]) ret.toArray( new String[ ret.size() ] );
745
return null;
1076
746
}
1077
747
1078
/**
1079
* Amongst the params, return any type values present. For v2.1 vCards,
1080
* those types are just parameters. For v3.0, they are prefixed with
1081
* "TYPE=". There may also be multiple type parameters.
1082
* @param params an array of params to look for types in
1083
* @param valid_types an list of upper-case type values to look for
1084
* @return a set of present type values
1085
*/
1086
748
private Set< String > extractTypes( String[] params,
1087
749
List< String > valid_types )
1088
750
{
1089
751
HashSet< String > types = new HashSet< String >();
1090
752
1091
753
// get 3.0-style TYPE= param
1092
String type_params[] = checkParams( params, "TYPE" );
1093
for( int a = 0; a < type_params.length; a++ )
1094
{
1095
// check for a comma-separated list of types (why? I don't think
1096
// this is in the specs!)
1097
String[] parts = type_params[ a ].split( "," );
1098
for( int i = 0; i < parts.length; i++ ) {
1099
String ucpart = parts[ i ].toUpperCase( Locale.US );
1100
if( valid_types.contains( ucpart ) )
1101
types.add( ucpart );
1102
}
754
String type_param;
755
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
756
String[] parts = type_param.split( "," );
757
for( int i = 0; i < parts.length; i++ )
758
if( valid_types.contains( parts[ i ] ) )
759
types.add( parts[ i ] );
1103
760
}
1104
761
1105
762
// get 2.1-style type param
1106
763
if( _version.equals( "2.1" ) ) {
1107
for( int i = 1; i < params.length; i++ ) {
1108
String ucparam = params[ i ].toUpperCase( Locale.US );
1109
if( valid_types.contains( ucparam ) )
1110
types.add( ucparam );
1111
}
764
for( int i = 1; i < params.length; i++ )
765
if( valid_types.contains( params[ i ] ) )
766
types.add( params[ i ] );
1112
767
}
1113
768
1114
769
return types;
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480