(root)/android/import-contacts
» Revision
40
(compared to revision 99)
: 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 40
- compare with another revision
- download diff
- download tarball
- view history from revision 40
- Committer: edam
- Date: 2011-05-02 18:28:24 UTC
- Revision ID: edam@waxworlds.org-20110502182824-acgdi3qfxfzqgely
- fixed logic for vcard field types (home, work, cell, etc) so it works
- updated NEWS and TODO
- rewrote most of ContactsCache, including a new ContactIdentifier class to identify contacts in the cache and new cache building code
- contacts now identified in the same way that Andoid displays them (by name, or organisation, or number, or email, in that order)
- propper handling and support for organisations and titles
- validation of imported contact now done by Importer, not VcfImporter
- separated sanitisation and normalisation (for cache lookups)
- generacised PhoneData, EmailData and AddressData classes
- ContactData is now aware of primary numbers, emails and organisations (defaults to the first prefrred one seen, or the first one seen where none is preferred)
- updated NEWS and TODO
- rewrote most of ContactsCache, including a new ContactIdentifier class to identify contacts in the cache and new cache building code
- contacts now identified in the same way that Andoid displays them (by name, or organisation, or number, or email, in that order)
- propper handling and support for organisations and titles
- validation of imported contact now done by Importer, not VcfImporter
- separated sanitisation and normalisation (for cache lookups)
- generacised PhoneData, EmailData and AddressData classes
- ContactData is now aware of primary numbers, emails and organisations (defaults to the first prefrred one seen, or the first one seen where none is preferred)
- 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:
- .bzrignore
added
removed
src/org/waxworlds/edam/importcontacts/VCFImporter.java
1
1
/*
2
2
* VCFImporter.java
3
3
*
4
* Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
4
* Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
5
5
*
6
6
* This file is part of the Import Contacts program (hereafter referred
7
* to as "this program"). For more information, see
8
* http://ed.am/dev/android/import-contacts
7
* to as "this program"). For more information, see
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;
38
38
import java.util.HashSet;
39
39
import java.util.Iterator;
40
40
import java.util.List;
41
import java.util.Locale;
42
41
import java.util.NoSuchElementException;
43
42
import java.util.Set;
44
43
import java.util.Vector;
45
44
import java.util.regex.Matcher;
46
45
import java.util.regex.Pattern;
47
46
48
import android.annotation.SuppressLint;
47
import org.waxworlds.edam.importcontacts.Importer.ContactData.ExtraDetail;
48
49
49
import android.content.SharedPreferences;
50
import android.provider.Contacts;
51
import android.provider.Contacts.PhonesColumns;
50
52
51
public class VcardImporter extends Importer
53
public class VCFImporter extends Importer
52
54
{
53
private int _vcard_count = 0;
55
private int _vCardCount = 0;
54
56
private int _progress = 0;
55
57
56
public VcardImporter( Doit doit )
58
public VCFImporter( Doit doit )
57
59
{
58
60
super( doit );
59
61
}
60
62
61
@SuppressLint( "SdCardPath" )
62
63
@Override
63
64
protected void onImport() throws AbortImportException
64
65
{
83
84
// get files
84
85
class VCardFilter implements FilenameFilter {
85
86
public boolean accept( File dir, String name ) {
86
return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
87
return name.toLowerCase().endsWith( ".vcf" );
87
88
}
88
89
}
89
90
files = file.listFiles( new VCardFilter() );
111
112
countVCardFile( files[ i ] );
112
113
setTmpProgress( i );
113
114
}
114
setProgressMax( _vcard_count ); // will also update tmp progress
115
setProgressMax( _vCardCount ); // will also update tmp progress
115
116
116
117
// import them
117
118
setProgress( 0 );
118
119
for( int i = 0; i < files.length; i++ )
119
120
importVCardFile( files[ i ] );
120
setProgress( _vcard_count );
121
121
}
122
122
123
123
private void countVCardFile( File file ) throws AbortImportException
130
130
131
131
// read
132
132
String line;
133
boolean in_vcard = false;
133
boolean inVCard = false;
134
134
while( ( line = reader.readLine() ) != null )
135
135
{
136
if( !in_vcard )
137
{
136
if( !inVCard ) {
138
137
// 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() );
138
if( line.matches( "^BEGIN:VCARD" ) ) {
139
inVCard = true;
140
_vCardCount++;
147
141
}
148
142
}
149
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
150
in_vcard = false;
143
else if( line.matches( "^END:VCARD" ) )
144
inVCard = false;
151
145
}
152
146
153
147
}
176
170
FileInputStream istream = new FileInputStream( file );
177
171
byte[] content = new byte[ (int)file.length() ];
178
172
istream.read( content );
179
istream = null;
180
173
181
174
// import
182
175
importVCardFileContent( content, file.getName() );
183
176
}
184
catch( OutOfMemoryError e ) {
185
showError( R.string.error_outofmemory );
186
}
187
177
catch( FileNotFoundException e ) {
188
178
showError( getText( R.string.error_filenotfound ) +
189
179
file.getName() );
197
187
throws AbortImportException
198
188
{
199
189
// go through lines
200
Vcard vcard = null;
201
int vcard_start_line = 0;
190
VCard vCard = null;
202
191
ContentLineIterator cli = new ContentLineIterator( content );
203
192
while( cli.hasNext() )
204
193
{
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 ) {
194
ByteBuffer buffer = cli.next();
195
196
// get a US-ASCII version of the line for processing
197
String line;
198
try {
199
line = new String( buffer.array(), buffer.position(),
200
buffer.limit() - buffer.position(), "US-ASCII" );
201
}
202
catch( UnsupportedEncodingException e ) {
203
// we know US-ASCII is supported, so appease the compiler...
204
line = "";
205
}
206
207
if( vCard == null ) {
211
208
// 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();
209
if( line.matches( "^BEGIN:VCARD" ) ) {
210
setProgress( ++_progress );
211
vCard = new VCard();
216
212
}
217
213
}
218
214
else {
219
215
// look for vcard content or ending
220
if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
216
if( line.matches( "^END:VCARD" ) )
221
217
{
222
// finalise the vcard/contact
218
// store vcard and do away with it
223
219
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;
220
vCard.finaliseParsing();
221
importContact( vCard );
222
}
223
catch( VCard.ParseException e ) {
224
skipContact();
225
if( !showContinue(
226
getText( R.string.error_vcf_parse ).toString()
227
+ fileName + "\n" + e.getMessage() ) )
228
{
229
finish( ACTION_ABORT );
230
}
231
}
232
catch( VCard.SkipContactException e ) {
233
skipContact();
234
// do nothing
235
}
236
vCard = null;
258
237
}
259
238
else
260
239
{
261
240
// try giving the line to the vcard
262
241
try {
263
vcard.parseLine( content_line );
242
vCard.parseLine( buffer, line,
243
cli.doesNextLineLookFolded() );
264
244
}
265
catch( Vcard.ParseException e ) {
245
catch( VCard.ParseException e ) {
266
246
skipContact();
267
247
if( !showContinue(
268
248
getText( R.string.error_vcf_parse ).toString()
269
+ fileName +
270
getText( R.string.error_vcf_parse_line ).toString()
271
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
249
+ fileName + "\n" + e.getMessage() ) )
272
250
{
273
251
finish( ACTION_ABORT );
274
252
}
275
253
276
// Although we're continuing, we still need to abort
277
// this vCard. Further lines will be ignored until we
254
// although we're continuing, we still need to abort
255
// this vCard. Further lines will be ignored until we
278
256
// get to another BEGIN:VCARD line.
279
vcard = null;
257
vCard = null;
280
258
}
281
catch( Vcard.SkipImportException e ) {
259
catch( VCard.SkipContactException e ) {
282
260
skipContact();
283
// Abort this vCard. Further lines will be ignored until
261
// abort this vCard. Further lines will be ignored until
284
262
// we get to another BEGIN:VCARD line.
285
vcard = null;
263
vCard = null;
286
264
}
287
265
}
288
266
}
289
267
}
290
268
}
291
269
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 >
270
class ContentLineIterator implements Iterator< ByteBuffer >
335
271
{
336
272
protected byte[] _content = null;
337
273
protected int _pos = 0;
338
protected int _line = 0;
339
274
340
275
public ContentLineIterator( byte[] content )
341
276
{
349
284
}
350
285
351
286
@Override
352
public ContentLine next()
287
public ByteBuffer next()
353
288
{
354
289
int initial_pos = _pos;
355
290
361
296
int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
362
297
_pos > initial_pos )? _pos - 1 : _pos;
363
298
_pos++;
364
_line++;
365
return new ContentLine(
366
ByteBuffer.wrap( _content, initial_pos,
367
to - initial_pos ),
368
doesNextLineLookFolded() );
299
return ByteBuffer.wrap( _content, initial_pos,
300
to - initial_pos );
369
301
}
370
302
371
303
// we didn't find one, but were there bytes left?
372
304
if( _pos != initial_pos ) {
373
305
int to = _pos;
374
306
_pos++;
375
_line++;
376
return new ContentLine(
377
ByteBuffer.wrap( _content, initial_pos,
378
to - initial_pos ),
379
doesNextLineLookFolded() );
307
return ByteBuffer.wrap( _content, initial_pos,
308
to - initial_pos );
380
309
}
381
310
382
311
// no bytes left
394
323
* onto the end of this one?
395
324
* @return
396
325
*/
397
private boolean doesNextLineLookFolded()
326
public boolean doesNextLineLookFolded()
398
327
{
399
328
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;
329
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
407
330
}
408
331
}
409
332
410
private class Vcard extends ContactData
333
private class VCard extends ContactData
411
334
{
412
335
private final static int NAMELEVEL_NONE = 0;
413
private final static int NAMELEVEL_N = 1;
414
private final static int NAMELEVEL_FN = 2;
336
private final static int NAMELEVEL_FN = 1;
337
private final static int NAMELEVEL_N = 2;
415
338
416
339
private final static int MULTILINE_NONE = 0;
417
340
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
418
341
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
419
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
342
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
420
343
421
344
private String _version = null;
422
private Vector< ContentLine > _content_lines = null;
345
private Vector< ByteBuffer > _buffers = null;
423
346
private int _name_level = NAMELEVEL_NONE;
424
347
private int _parser_multiline_state = MULTILINE_NONE;
425
348
private String _parser_current_name_and_params = null;
461
384
462
385
public ParseException( int res )
463
386
{
464
super( VcardImporter.this.getText( res ).toString() );
387
super( VCFImporter.this.getText( res ).toString() );
465
388
}
466
389
}
467
390
468
391
@SuppressWarnings("serial")
469
protected class SkipImportException extends Exception { }
392
protected class SkipContactException extends Exception { }
470
393
471
private String extractCollonPartFromLine( ContentLine content_line,
472
boolean former )
394
private String extractCollonPartFromLine( ByteBuffer buffer,
395
String line, boolean former )
473
396
{
397
String ret = null;
398
399
// get a US-ASCII version of the line for processing, unless we were
400
// supplied with one
401
if( line == null ) {
402
try {
403
line = new String( buffer.array(), buffer.position(),
404
buffer.limit() - buffer.position(), "US-ASCII" );
405
}
406
catch( UnsupportedEncodingException e ) {
407
// we know US-ASCII is supported, so appease the compiler...
408
line = "";
409
}
410
}
411
474
412
// split line into name and value parts and check to make sure we
475
413
// only got 2 parts and that the first part is not zero in length
476
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
414
String[] parts = line.split( ":", 2 );
477
415
if( parts.length == 2 && parts[ 0 ].length() > 0 )
478
return parts[ former? 0 : 1 ].trim();
479
480
return null;
481
}
482
483
private String extractNameAndParamsFromLine( ContentLine content_line )
484
{
485
return extractCollonPartFromLine( content_line, true );
486
}
487
488
private String extractValueFromLine( ContentLine content_line )
489
{
490
return extractCollonPartFromLine( content_line, false );
491
}
492
493
public void parseLine( ContentLine content_line )
494
throws ParseException, SkipImportException,
416
ret = parts[ former? 0 : 1 ];
417
418
return ret;
419
}
420
421
private String extractNameAndParamsFromLine( ByteBuffer buffer,
422
String line )
423
{
424
return extractCollonPartFromLine( buffer, line, true );
425
}
426
427
private String extractValueFromLine( ByteBuffer buffer, String line )
428
{
429
return extractCollonPartFromLine( buffer, line, false );
430
}
431
432
public void parseLine( ByteBuffer buffer, String line,
433
boolean next_line_looks_folded )
434
throws ParseException, SkipContactException,
495
435
AbortImportException
496
436
{
497
437
// do we have a version yet?
499
439
{
500
440
// tentatively get name and params from line
501
441
String name_and_params =
502
extractNameAndParamsFromLine( content_line );
442
extractNameAndParamsFromLine( buffer, line );
503
443
504
444
// is it a version line?
505
445
if( name_and_params != null &&
506
name_and_params.equalsIgnoreCase( "VERSION" ) )
446
name_and_params.equals( "VERSION" ) )
507
447
{
508
448
// yes, get it!
509
String value = extractValueFromLine( content_line );
510
if( value == null || (
511
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
512
{
449
String value = extractValueFromLine( buffer, line );
450
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
451
throw new ParseException( R.string.error_vcf_version );
514
}
515
452
_version = value;
516
453
517
454
// parse any buffers we've been accumulating while we waited
518
455
// for a version
519
if( _content_lines != null )
520
for( int i = 0; i < _content_lines.size(); i++ )
521
parseLine( _content_lines.get( i ) );
522
_content_lines = null;
456
if( _buffers != null )
457
for( int i = 0; i < _buffers.size(); i++ )
458
parseLine( _buffers.get( i ), null,
459
i + 1 < _buffers.size() &&
460
_buffers.get( i + 1 ).hasRemaining() &&
461
_buffers.get( i + 1 ).get(
462
_buffers.get( i + 1 ).position() ) == ' ' );
463
_buffers = null;
523
464
}
524
465
else
525
466
{
526
467
// no, so stash this line till we get a version
527
if( _content_lines == null )
528
_content_lines = new Vector< ContentLine >();
529
_content_lines.add( content_line );
468
if( _buffers == null )
469
_buffers = new Vector< ByteBuffer >();
470
_buffers.add( buffer );
530
471
}
531
472
}
532
473
else
533
474
{
534
475
// name and params and the position in the buffer where the
535
// "value" part of the line starts
476
// "value" part of the line start
536
477
String name_and_params;
537
478
int pos;
538
479
544
485
545
486
// skip some initial line characters, depending on the type
546
487
// of multi-line we're handling
547
pos = content_line.getBuffer().position();
488
pos = buffer.position();
548
489
switch( _parser_multiline_state )
549
490
{
550
491
case MULTILINE_FOLDED:
551
492
pos++;
552
493
break;
553
494
case MULTILINE_ENCODED:
554
while( pos < content_line.getBuffer().limit() && (
555
content_line.getBuffer().get( pos ) == ' ' ||
556
content_line.getBuffer().get( pos ) == '\t' ) )
495
while( pos < buffer.limit() && (
496
buffer.get( pos ) == ' ' ||
497
buffer.get( pos ) == '\t' ) )
557
498
{
558
499
pos++;
559
500
}
568
509
}
569
510
else
570
511
{
571
// skip empty lines
572
if( content_line.getUsAsciiLine().trim().length() == 0 )
573
return;
574
575
512
// get name and params from line, and since we're not
576
513
// parsing a subsequent line in a multi-line, this should
577
514
// not fail, or it's an error
578
515
name_and_params =
579
extractNameAndParamsFromLine( content_line );
516
extractNameAndParamsFromLine( buffer, line );
580
517
if( name_and_params == null )
581
518
throw new ParseException(
582
519
R.string.error_vcf_malformed );
583
520
584
521
// calculate how many chars to skip from beginning of line
585
522
// so we skip the property "name:" part
586
pos = content_line.getBuffer().position() +
587
name_and_params.length() + 1;
523
pos = buffer.position() + name_and_params.length() + 1;
588
524
589
525
// reset the saved multi-line state
590
526
_parser_current_name_and_params = name_and_params;
593
529
594
530
// get value from buffer, as raw bytes
595
531
ByteBuffer value;
596
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
597
content_line.getBuffer().limit() - pos );
532
value = ByteBuffer.wrap( buffer.array(), pos,
533
buffer.limit() - pos );
598
534
599
535
// get parameter parts
600
536
String[] name_param_parts = name_and_params.split( ";", -1 );
601
537
for( int i = 0; i < name_param_parts.length; i++ )
602
538
name_param_parts[ i ] = name_param_parts[ i ].trim();
603
539
604
// determine whether we care about this entry
605
final HashSet< String > interesting_fields =
606
new HashSet< String >( Arrays.asList( new String[] { "N",
607
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
608
) );
609
boolean is_interesting_field =
610
interesting_fields.contains(
611
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
612
613
540
// parse encoding parameter
614
541
String encoding = checkParam( name_param_parts, "ENCODING" );
615
if( encoding != null )
616
encoding = encoding.toUpperCase( Locale.US );
617
if( is_interesting_field && encoding != null &&
618
!encoding.equalsIgnoreCase( "8BIT" ) &&
619
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
620
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
542
if( encoding != null ) encoding = encoding.toUpperCase();
543
if( encoding != null && !encoding.equals( "8BIT" ) &&
544
!encoding.equals( "QUOTED-PRINTABLE" ) )
545
//&& !encoding.equals( "BASE64" ) )
621
546
{
622
547
throw new ParseException( R.string.error_vcf_encoding );
623
548
}
624
549
625
550
// parse charset parameter
626
551
String charset = checkParam( name_param_parts, "CHARSET" );
627
if( charset != null )
628
charset = charset.toUpperCase( Locale.US );
629
if( charset != null &&
630
!charset.equalsIgnoreCase( "US-ASCII" ) &&
631
!charset.equalsIgnoreCase( "ASCII" ) &&
632
!charset.equalsIgnoreCase( "UTF-8" ) )
552
if( charset != null ) charset = charset.toUpperCase();
553
if( charset != null && !charset.equals( "US-ASCII" ) &&
554
!charset.equals( "ASCII" ) &&
555
!charset.equals( "UTF-8" ) )
633
556
{
634
557
throw new ParseException( R.string.error_vcf_charset );
635
558
}
637
560
// do unencoding (or default to a fake unencoding result with
638
561
// the raw string)
639
562
UnencodeResult unencoding_result = null;
640
if( encoding != null &&
641
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
642
{
563
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
643
564
unencoding_result = unencodeQuotedPrintable( value );
644
}
645
// else if( encoding != null &&
646
// encoding.equalsIgnoreCase( "BASE64" ) )
647
// {
565
// else if( encoding != null && encoding.equals( "BASE64" ) )
648
566
// unencoding_result = unencodeBase64( props[ 1 ], charset );
649
// }
650
567
if( unencoding_result != null ) {
651
568
value = unencoding_result.getBuffer();
652
569
if( unencoding_result.isAnotherLineRequired() )
653
570
_parser_multiline_state = MULTILINE_ENCODED;
654
571
}
655
572
656
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
657
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
658
if( ( charset == null && _version.equals( "2.1" ) ) ||
659
( charset != null && (
660
charset.equalsIgnoreCase( "ASCII" ) ||
661
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
662
{
573
// convert 8-bit ASCII charset to US-ASCII
574
if( charset == null || charset.equals( "ASCII" ) ) {
663
575
value = transcodeAsciiToUtf8( value );
576
charset = "UTF-8";
664
577
}
665
578
666
// process charset (value is now in UTF-8)
579
// process charset
667
580
String string_value;
668
581
try {
669
582
string_value = new String( value.array(), value.position(),
670
value.limit() - value.position(), "UTF-8" );
583
value.limit() - value.position(), charset );
671
584
} catch( UnsupportedEncodingException e ) {
672
585
throw new ParseException( R.string.error_vcf_charset );
673
586
}
675
588
// for some entries that have semicolon-separated value parts,
676
589
// check to see if the value ends in an escape character, which
677
590
// indicates that we have a multi-line value
678
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
679
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
680
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
591
if( ( name_param_parts[ 0 ].equals( "N" ) ||
592
name_param_parts[ 0 ].equals( "ORG" ) ||
593
name_param_parts[ 0 ].equals( "ADR" ) ) &&
681
594
doesStringEndInAnEscapeChar( string_value ) )
682
595
{
683
596
_parser_multiline_state = MULTILINE_ESCAPED;
685
598
string_value.length() - 1 );
686
599
}
687
600
688
// if we know we're not in an encoding-based multi-line, check
689
// to see if we're in a folded multi-line
601
// now we know whether we're in an encoding multi-line,
602
// determine if we're in a v3 folded multi-line or not
690
603
if( _parser_multiline_state == MULTILINE_NONE &&
691
content_line.doesNextLineLookFolded() )
604
_version.equals( "3.0" ) && next_line_looks_folded )
692
605
{
693
606
_parser_multiline_state = MULTILINE_FOLDED;
694
607
}
706
619
if( complete_value.length() < 1 ) return;
707
620
708
621
// parse some properties
709
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
622
if( name_param_parts[ 0 ].equals( "N" ) )
710
623
parseN( name_param_parts, complete_value );
711
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
624
else if( name_param_parts[ 0 ].equals( "FN" ) )
712
625
parseFN( name_param_parts, complete_value );
713
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
626
else if( name_param_parts[ 0 ].equals( "ORG" ) )
714
627
parseORG( name_param_parts, complete_value );
715
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
628
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
716
629
parseTITLE( name_param_parts, complete_value );
717
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
630
else if( name_param_parts[ 0 ].equals( "TEL" ) )
718
631
parseTEL( name_param_parts, complete_value );
719
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
632
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
720
633
parseEMAIL( name_param_parts, complete_value );
721
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
634
else if( name_param_parts[ 0 ].equals( "ADR" ) )
722
635
parseADR( name_param_parts, complete_value );
723
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
724
parseLABEL( name_param_parts, complete_value );
725
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
726
parseNOTE( name_param_parts, complete_value );
727
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
728
parseBDAY( name_param_parts, complete_value );
729
636
}
730
637
}
731
638
744
651
return ( count & 1 ) == 1;
745
652
}
746
653
747
private String[] splitValueByCharacter( String value, char character )
654
private String[] splitValueBySemicolon( String value )
748
655
{
749
// split string in to parts by specified character
656
// split string in to parts by semicolon
750
657
ArrayList< String > parts = new ArrayList< String >(
751
Arrays.asList( value.split( "" + character ) ) );
658
Arrays.asList( value.split( ";" ) ) );
752
659
753
660
// go through parts
754
661
for( int a = 0; a < parts.size(); a++ )
755
662
{
756
663
String str = parts.get( a );
757
664
758
// Look for parts that end in an escape character, but ignore
759
// the final part. We've already detected escape chars at the
665
// look for parts that end in an escape character, but ignore
666
// the final part. We've already detected escape chars at the
760
667
// end of the final part in parseLine() and handled multi-lines
761
668
// accordingly.
762
669
if( a < parts.size() - 1 &&
763
670
doesStringEndInAnEscapeChar( str ) )
764
671
{
765
// append the escaped character, join the next part to this
766
// part and remove the next part
672
// join the next part to this part and remove the next part
767
673
parts.set( a, str.substring( 0, str.length() - 1 ) +
768
character + parts.get( a + 1 ) );
674
';' + parts.get( a + 1 ) );
769
675
parts.remove( a + 1 );
770
676
771
677
// re-visit this part
782
688
return parts.toArray( ret );
783
689
}
784
690
785
private String unescapeValue( String value )
786
{
787
StringBuilder ret = new StringBuilder( value.length() );
788
boolean in_escape = false;
789
for( int a = 0; a < value.length(); a++ )
790
{
791
int c = value.codePointAt( a );
792
793
// process a normal character
794
if( !in_escape ) {
795
if( c == '\\' )
796
in_escape = true;
797
else
798
ret.append( Character.toChars( c ) );
799
continue;
800
}
801
802
// process an escape sequence
803
in_escape = false;
804
switch( c )
805
{
806
case 'T':
807
case 't':
808
// add tab (invalid/non-standard, but accepted)
809
ret.append( '\t' );
810
break;
811
case 'N':
812
case 'n':
813
// add newline
814
ret.append( '\n' );
815
break;
816
case '\\':
817
case ',':
818
case ';':
819
// add escaped character
820
ret.append( Character.toChars( c ) );
821
break;
822
default:
823
// unknown escape sequence, so add it unescaped
824
// (invalid/non-standard, but accepted)
825
ret.append( "\\" );
826
ret.append( Character.toChars( c ) );
827
break;
828
}
829
}
830
831
return ret.toString();
832
}
833
834
691
private void parseN( String[] params, String value )
835
692
{
836
693
// already got a better name?
837
694
if( _name_level >= NAMELEVEL_N ) return;
838
695
839
696
// get name parts
840
String[] name_parts = splitValueByCharacter( value, ';' );
697
String[] name_parts = splitValueBySemicolon( value );
841
698
842
699
// build name
843
700
value = "";
844
final int[] part_order = { 3, 1, 2, 0, 4 };
845
for( int a = 0; a < part_order.length; a++ )
846
if( name_parts.length > part_order[ a ] &&
847
name_parts[ part_order[ a ] ].length() > 0 )
848
{
849
// split this part in to it's comma-separated bits
850
String[] name_part_parts = splitValueByCharacter(
851
name_parts[ part_order[ a ] ], ',' );
852
for( int b = 0; b < name_part_parts.length; b++ )
853
if( name_part_parts[ b ].length() > 0 )
854
{
855
if( value.length() > 0 ) value += " ";
856
value += name_part_parts[ b ];
857
}
858
}
701
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
702
value += name_parts[ 1 ];
703
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
704
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
859
705
860
706
// set name
861
setName( unescapeValue( value ) );
707
setName( value );
862
708
_name_level = NAMELEVEL_N;
863
709
}
864
710
868
714
if( _name_level >= NAMELEVEL_FN ) return;
869
715
870
716
// set name
871
setName( unescapeValue( value ) );
717
setName( value );
872
718
_name_level = NAMELEVEL_FN;
873
719
}
874
720
875
721
private void parseORG( String[] params, String value )
876
722
{
877
723
// get org parts
878
String[] org_parts = splitValueByCharacter( value, ';' );
724
String[] org_parts = splitValueBySemicolon( value );
879
725
if( org_parts == null || org_parts.length < 1 ) return;
880
726
881
727
// build organisation name
883
729
String.valueOf( org_parts[ 0 ] ) );
884
730
for( int a = 1; a < org_parts.length; a++ )
885
731
builder.append( ", " ).append( org_parts[ a ] );
886
String organisation = unescapeValue( builder.toString() );
732
String organisation = builder.toString();
887
733
888
734
// set organisation name (using a title we've previously found)
889
735
addOrganisation( organisation, _cached_title, true );
900
746
901
747
private void parseTITLE( String[] params, String value )
902
748
{
903
value = unescapeValue( value );
904
905
749
// if we previously had an organisation, look it up and append this
906
750
// title to it
907
751
if( _cached_organisation != null && hasOrganisations() ) {
929
773
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
930
774
931
775
// here's the logic...
932
boolean is_preferred = types.contains( "PREF" );
776
boolean preferred = types.contains( "PREF" );
933
777
int type;
934
778
if( types.contains( "FAX" ) )
935
779
if( types.contains( "HOME" ) )
936
type = TYPE_FAX_HOME;
780
type = PhonesColumns.TYPE_FAX_HOME;
937
781
else
938
type = TYPE_FAX_WORK;
782
type = PhonesColumns.TYPE_FAX_WORK;
939
783
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
940
type = TYPE_MOBILE;
784
type = PhonesColumns.TYPE_MOBILE;
941
785
else if( types.contains( "PAGER" ) )
942
type = TYPE_PAGER;
786
type = PhonesColumns.TYPE_PAGER;
943
787
else if( types.contains( "WORK" ) )
944
type = TYPE_WORK;
788
type = PhonesColumns.TYPE_WORK;
945
789
else
946
type = TYPE_HOME;
790
type = PhonesColumns.TYPE_HOME;
947
791
948
792
// add phone number
949
addNumber( value, type, is_preferred );
793
addNumber( value, type, preferred );
950
794
}
951
795
952
796
public void parseEMAIL( String[] params, String value )
957
801
"PREF", "WORK", "HOME", "INTERNET" ) );
958
802
959
803
// add email address
960
boolean is_preferred = types.contains( "PREF" );
804
boolean preferred = types.contains( "PREF" );
961
805
int type;
962
806
if( types.contains( "WORK" ) )
963
type = TYPE_WORK;
807
type = Contacts.ContactMethods.TYPE_WORK;
964
808
else
965
type = TYPE_HOME;
809
type = Contacts.ContactMethods.TYPE_HOME;
966
810
967
addEmail( unescapeValue( value ), type, is_preferred );
811
addEmail( value, type, preferred );
968
812
}
969
813
970
814
private void parseADR( String[] params, String value )
971
815
{
972
816
// get address parts
973
String[] adr_parts = splitValueByCharacter( value, ';' );
817
String[] adr_parts = splitValueBySemicolon( value );
974
818
975
819
// build address
976
820
value = "";
977
for( int a = 0; a < adr_parts.length; a++ )
978
if( adr_parts[ a ].length() > 0 )
979
{
980
// version 3.0 vCards allow further splitting by comma
981
if( _version.equals( "3.0" ) )
982
{
983
// split this part in to it's comma-separated bits and
984
// add them on individual lines
985
String[] adr_part_parts =
986
splitValueByCharacter( adr_parts[ a ], ',' );
987
for( int b = 0; b < adr_part_parts.length; b++ )
988
if( adr_part_parts[ b ].length() > 0 )
989
{
990
if( value.length() > 0 ) value += "\n";
991
value += adr_part_parts[ b ];
992
}
993
}
994
else
995
{
996
// add this part on an individual line
997
if( value.length() > 0 ) value += "\n";
998
value += adr_parts[ a ];
999
}
1000
}
1001
1002
Set< String > types = extractTypes( params, Arrays.asList(
1003
"PREF", "WORK", "HOME" ) );
1004
1005
// add address
1006
int type;
1007
if( types.contains( "WORK" ) )
1008
type = TYPE_WORK;
1009
else
1010
type = TYPE_HOME;
1011
1012
addAddress( unescapeValue( value ), type );
1013
}
1014
1015
private void parseLABEL( String[] params, String value )
1016
{
1017
Set< String > types = extractTypes( params, Arrays.asList(
1018
"PREF", "WORK", "HOME" ) );
1019
1020
// add address
1021
int type;
1022
if( types.contains( "WORK" ) )
1023
type = TYPE_WORK;
1024
else
1025
type = TYPE_HOME;
1026
1027
addAddress( unescapeValue( value ), type );
1028
}
1029
1030
private void parseNOTE( String[] params, String value )
1031
{
1032
addNote( unescapeValue( value ) );
1033
}
1034
1035
private void parseBDAY( String[] params, String value )
1036
{
1037
setBirthday( value );
1038
}
1039
1040
public void finaliseVcard()
1041
throws ParseException, ContactNotIdentifiableException
821
for( int a = 0; a < adr_parts.length; a++ ) {
822
if( value.length() > 0 ) value += "\n";
823
value += adr_parts[ a ].trim();
824
}
825
826
Set< String > types = extractTypes( params, Arrays.asList(
827
"PREF", "WORK", "HOME", "INTERNET" ) );
828
829
// add address
830
int type;
831
if( types.contains( "WORK" ) )
832
type = Contacts.ContactMethods.TYPE_WORK;
833
else
834
type = Contacts.ContactMethods.TYPE_HOME;
835
836
addAddress( value, type );
837
}
838
839
public void finaliseParsing()
840
throws ParseException, SkipContactException,
841
AbortImportException
1042
842
{
1043
843
// missing version (and data is present)
1044
if( _version == null && _content_lines != null )
844
if( _version == null && _buffers != null )
1045
845
throw new ParseException( R.string.error_vcf_malformed );
1046
846
1047
// finalise the parent class
1048
finalise();
847
// check if we should import this contact
848
try {
849
if( !isImportRequired( this ) )
850
throw new SkipContactException();
851
}
852
catch( ContactNeedsMoreInfoException e ) {
853
throw new ParseException( R.string.error_vcf_notenoughinfo );
854
}
1049
855
}
1050
856
1051
/**
1052
* Amongst the params, find the value of the first, only, of any with
1053
* the specified name.
1054
*
1055
* @param params
1056
* @param name
1057
* @return a value, or null
1058
*/
1059
857
private String checkParam( String[] params, String name )
1060
858
{
1061
String[] res = checkParams( params, name );
1062
return res.length > 0? res[ 0 ] : null;
1063
}
1064
1065
/**
1066
* Amongst the params, find the values of any with the specified name.
1067
*
1068
* @param params
1069
* @param name
1070
* @return an array of values, or null
1071
*/
1072
private String[] checkParams( String[] params, String name )
1073
{
1074
HashSet< String > ret = new HashSet< String >();
1075
1076
859
Pattern p = Pattern.compile(
1077
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1078
Pattern.CASE_INSENSITIVE );
860
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1079
861
for( int i = 0; i < params.length; i++ ) {
1080
862
Matcher m = p.matcher( params[ i ] );
1081
863
if( m.matches() )
1082
ret.add( m.group( 2 ) );
864
return m.group( 2 );
1083
865
}
1084
1085
return (String[]) ret.toArray( new String[ ret.size() ] );
866
return null;
1086
867
}
1087
868
1088
/**
1089
* Amongst the params, return any type values present. For v2.1 vCards,
1090
* those types are just parameters. For v3.0, they are prefixed with
1091
* "TYPE=". There may also be multiple type parameters.
1092
*
1093
* @param params an array of params to look for types in
1094
* @param valid_types an list of upper-case type values to look for
1095
* @return a set of present type values
1096
*/
1097
869
private Set< String > extractTypes( String[] params,
1098
870
List< String > valid_types )
1099
871
{
1100
872
HashSet< String > types = new HashSet< String >();
1101
873
1102
874
// get 3.0-style TYPE= param
1103
String type_params[] = checkParams( params, "TYPE" );
1104
for( int a = 0; a < type_params.length; a++ )
1105
{
1106
// check for a comma-separated list of types (why? I don't think
1107
// this is in the specs!)
1108
String[] parts = type_params[ a ].split( "," );
1109
for( int i = 0; i < parts.length; i++ ) {
1110
String ucpart = parts[ i ].toUpperCase( Locale.US );
1111
if( valid_types.contains( ucpart ) )
1112
types.add( ucpart );
1113
}
875
String type_param;
876
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
877
String[] parts = type_param.split( "," );
878
for( int i = 0; i < parts.length; i++ )
879
if( valid_types.contains( parts[ i ] ) )
880
types.add( parts[ i ] );
1114
881
}
1115
882
1116
883
// get 2.1-style type param
1117
884
if( _version.equals( "2.1" ) ) {
1118
for( int i = 1; i < params.length; i++ ) {
1119
String ucparam = params[ i ].toUpperCase( Locale.US );
1120
if( valid_types.contains( ucparam ) )
1121
types.add( ucparam );
1122
}
885
for( int i = 1; i < params.length; i++ )
886
if( valid_types.contains( params[ i ] ) )
887
types.add( params[ i ] );
1123
888
}
1124
889
1125
890
return types;
1147
912
else if( ch == '=' && i == in.limit() - 1 )
1148
913
{
1149
914
// we found a '=' at the end of a line signifying a multi-
1150
// line string, so we don't add it
915
// line string, so we don't add it.
1151
916
another = true;
1152
917
continue;
1153
918
}
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480