(root)/android/import-contacts
» Revision
36
(compared to revision 102)
: 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:
- TODO
- 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 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;
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
reader.close();
153
143
154
144
}
155
145
catch( FileNotFoundException e ) {
177
167
FileInputStream istream = new FileInputStream( file );
178
168
byte[] content = new byte[ (int)file.length() ];
179
169
istream.read( content );
180
istream.close();
181
170
182
171
// import
183
172
importVCardFileContent( content, file.getName() );
184
173
}
185
catch( OutOfMemoryError e ) {
186
showError( R.string.error_outofmemory );
187
}
188
174
catch( FileNotFoundException e ) {
189
175
showError( getText( R.string.error_filenotfound ) +
190
176
file.getName() );
198
184
throws AbortImportException
199
185
{
200
186
// go through lines
201
Vcard vcard = null;
202
int vcard_start_line = 0;
187
VCard vCard = null;
203
188
ContentLineIterator cli = new ContentLineIterator( content );
204
189
while( cli.hasNext() )
205
190
{
206
ContentLine content_line = cli.next();
207
208
// get a US-ASCII version of the string, for processing
209
String line = content_line.getUsAsciiLine();
210
211
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 ) {
212
205
// look for vcard beginning
213
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
214
setProgress( _progress++ );
215
vcard = new Vcard();
216
vcard_start_line = cli.getLineNumber();
206
if( line.matches( "^BEGIN:VCARD" ) ) {
207
setProgress( ++_progress );
208
vCard = new VCard();
217
209
}
218
210
}
219
211
else {
220
212
// look for vcard content or ending
221
if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
213
if( line.matches( "^END:VCARD" ) )
222
214
{
223
// finalise the vcard/contact
215
// store vcard and do away with it
224
216
try {
225
vcard.finaliseVcard();
226
227
// pass the finalised contact to the importer
228
importContact( vcard );
229
}
230
catch( Vcard.ParseException e ) {
231
if( !showContinue(
232
getText( R.string.error_vcf_parse ).toString()
233
+ fileName +
234
getText( R.string.error_vcf_parse_line ).toString()
235
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
236
{
237
finish( ACTION_ABORT );
238
}
239
else
240
skipContact();
241
}
242
catch( ContactData.ContactNotIdentifiableException e ) {
243
if( !showContinue(
244
getText( R.string.error_vcf_parse ).toString()
245
+ fileName +
246
getText( R.string.error_vcf_parse_line ).toString()
247
+ vcard_start_line + ":\n" + getText(
248
R.string.error_vcf_notenoughinfo ).toString()
249
) )
250
{
251
finish( ACTION_ABORT );
252
}
253
else
254
skipContact();
255
}
256
257
// discard this vcard
258
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;
259
234
}
260
235
else
261
236
{
262
237
// try giving the line to the vcard
263
238
try {
264
vcard.parseLine( content_line );
239
vCard.parseLine( buffer, line,
240
cli.doesNextLineLookFolded() );
265
241
}
266
catch( Vcard.ParseException e ) {
242
catch( VCard.ParseException e ) {
267
243
skipContact();
268
244
if( !showContinue(
269
245
getText( R.string.error_vcf_parse ).toString()
270
+ fileName +
271
getText( R.string.error_vcf_parse_line ).toString()
272
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
246
+ fileName + "\n" + e.getMessage() ) )
273
247
{
274
248
finish( ACTION_ABORT );
275
249
}
276
250
277
// Although we're continuing, we still need to abort
278
// this vCard. Further lines will be ignored until we
251
// although we're continuing, we still need to abort
252
// this vCard. Further lines will be ignored until we
279
253
// get to another BEGIN:VCARD line.
280
vcard = null;
254
vCard = null;
281
255
}
282
catch( Vcard.SkipImportException e ) {
256
catch( VCard.SkipContactException e ) {
283
257
skipContact();
284
// Abort this vCard. Further lines will be ignored until
258
// abort this vCard. Further lines will be ignored until
285
259
// we get to another BEGIN:VCARD line.
286
vcard = null;
260
vCard = null;
287
261
}
288
262
}
289
263
}
290
264
}
291
265
}
292
266
293
class ContentLine
294
{
295
private ByteBuffer _buffer;
296
private boolean _folded_next;
297
private String _line;
298
299
public ContentLine( ByteBuffer buffer, boolean folded_next )
300
{
301
_buffer = buffer;
302
_folded_next = folded_next;
303
_line = null;
304
}
305
306
public ByteBuffer getBuffer()
307
{
308
return _buffer;
309
}
310
311
public boolean doesNextLineLookFolded()
312
{
313
return _folded_next;
314
}
315
316
public String getUsAsciiLine()
317
{
318
// generated line and cache it
319
if( _line == null ) {
320
try {
321
_line = new String( _buffer.array(), _buffer.position(),
322
_buffer.limit() - _buffer.position(), "US-ASCII" );
323
}
324
catch( UnsupportedEncodingException e ) {
325
// we know US-ASCII *is* supported, so appease the
326
// compiler...
327
}
328
}
329
330
// return cached line
331
return _line;
332
}
333
}
334
335
class ContentLineIterator implements Iterator< ContentLine >
267
class ContentLineIterator implements Iterator< ByteBuffer >
336
268
{
337
269
protected byte[] _content = null;
338
270
protected int _pos = 0;
339
protected int _line = 0;
340
271
341
272
public ContentLineIterator( byte[] content )
342
273
{
350
281
}
351
282
352
283
@Override
353
public ContentLine next()
284
public ByteBuffer next()
354
285
{
355
286
int initial_pos = _pos;
356
287
362
293
int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
363
294
_pos > initial_pos )? _pos - 1 : _pos;
364
295
_pos++;
365
_line++;
366
return new ContentLine(
367
ByteBuffer.wrap( _content, initial_pos,
368
to - initial_pos ),
369
doesNextLineLookFolded() );
296
return ByteBuffer.wrap( _content, initial_pos,
297
to - initial_pos );
370
298
}
371
299
372
300
// we didn't find one, but were there bytes left?
373
301
if( _pos != initial_pos ) {
374
302
int to = _pos;
375
303
_pos++;
376
_line++;
377
return new ContentLine(
378
ByteBuffer.wrap( _content, initial_pos,
379
to - initial_pos ),
380
doesNextLineLookFolded() );
304
return ByteBuffer.wrap( _content, initial_pos,
305
to - initial_pos );
381
306
}
382
307
383
308
// no bytes left
395
320
* onto the end of this one?
396
321
* @return
397
322
*/
398
private boolean doesNextLineLookFolded()
323
public boolean doesNextLineLookFolded()
399
324
{
400
325
return _pos > 0 && _pos < _content.length &&
401
_content[ _pos - 1 ] == '\n' &&
402
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
403
}
404
405
public int getLineNumber()
406
{
407
return _line;
326
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
408
327
}
409
328
}
410
329
411
private class Vcard extends ContactData
330
private class VCard extends ContactData
412
331
{
413
332
private final static int NAMELEVEL_NONE = 0;
414
private final static int NAMELEVEL_N = 1;
333
private final static int NAMELEVEL_ORG = 1;
415
334
private final static int NAMELEVEL_FN = 2;
416
417
private final static int MULTILINE_NONE = 0;
418
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
419
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
420
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
335
private final static int NAMELEVEL_N = 3;
421
336
422
337
private String _version = null;
423
private Vector< ContentLine > _content_lines = null;
338
private Vector< ByteBuffer > _buffers = null;
424
339
private int _name_level = NAMELEVEL_NONE;
425
private int _parser_multiline_state = MULTILINE_NONE;
340
private boolean _parser_in_encoded_multiline = false;
341
private boolean _parser_in_folded_multiline = false;
426
342
private String _parser_current_name_and_params = null;
427
343
private String _parser_buffered_value_so_far = "";
428
private String _cached_organisation = null;
429
private String _cached_title = null;
430
344
431
345
protected class UnencodeResult
432
346
{
462
376
463
377
public ParseException( int res )
464
378
{
465
super( VcardImporter.this.getText( res ).toString() );
379
super( VCFImporter.this.getText( res ).toString() );
466
380
}
467
381
}
468
382
469
383
@SuppressWarnings("serial")
470
protected class SkipImportException extends Exception { }
384
protected class SkipContactException extends Exception { }
471
385
472
private String extractCollonPartFromLine( ContentLine content_line,
473
boolean former )
386
private String extractCollonPartFromLine( ByteBuffer buffer,
387
String line, boolean former )
474
388
{
389
String ret = null;
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
475
404
// split line into name and value parts and check to make sure we
476
405
// only got 2 parts and that the first part is not zero in length
477
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
406
String[] parts = line.split( ":", 2 );
478
407
if( parts.length == 2 && parts[ 0 ].length() > 0 )
479
return parts[ former? 0 : 1 ].trim();
480
481
return null;
482
}
483
484
private String extractNameAndParamsFromLine( ContentLine content_line )
485
{
486
return extractCollonPartFromLine( content_line, true );
487
}
488
489
private String extractValueFromLine( ContentLine content_line )
490
{
491
return extractCollonPartFromLine( content_line, false );
492
}
493
494
public void parseLine( ContentLine content_line )
495
throws ParseException, SkipImportException,
408
ret = parts[ former? 0 : 1 ];
409
410
return ret;
411
}
412
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,
496
427
AbortImportException
497
428
{
498
429
// do we have a version yet?
500
431
{
501
432
// tentatively get name and params from line
502
433
String name_and_params =
503
extractNameAndParamsFromLine( content_line );
434
extractNameAndParamsFromLine( buffer, line );
504
435
505
436
// is it a version line?
506
437
if( name_and_params != null &&
507
name_and_params.equalsIgnoreCase( "VERSION" ) )
438
name_and_params.equals( "VERSION" ) )
508
439
{
509
440
// yes, get it!
510
String value = extractValueFromLine( content_line );
511
if( value == null || (
512
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
513
{
441
String value = extractValueFromLine( buffer, line );
442
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
514
443
throw new ParseException( R.string.error_vcf_version );
515
}
516
444
_version = value;
517
445
518
446
// parse any buffers we've been accumulating while we waited
519
447
// for a version
520
if( _content_lines != null )
521
for( int i = 0; i < _content_lines.size(); i++ )
522
parseLine( _content_lines.get( i ) );
523
_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;
524
456
}
525
457
else
526
458
{
527
459
// no, so stash this line till we get a version
528
if( _content_lines == null )
529
_content_lines = new Vector< ContentLine >();
530
_content_lines.add( content_line );
460
if( _buffers == null )
461
_buffers = new Vector< ByteBuffer >();
462
_buffers.add( buffer );
531
463
}
532
464
}
533
465
else
534
466
{
535
467
// name and params and the position in the buffer where the
536
// "value" part of the line starts
468
// "value" part of the line start
537
469
String name_and_params;
538
470
int pos;
539
471
540
if( _parser_multiline_state != MULTILINE_NONE )
472
if( _parser_in_encoded_multiline ||
473
_parser_in_folded_multiline )
541
474
{
542
475
// if we're currently in a multi-line value, use the stored
543
476
// property name and parameters
544
477
name_and_params = _parser_current_name_and_params;
545
478
546
// skip some initial line characters, depending on the type
547
// of multi-line we're handling
548
pos = content_line.getBuffer().position();
549
switch( _parser_multiline_state )
550
{
551
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 )
552
484
pos++;
553
break;
554
case MULTILINE_ENCODED:
555
while( pos < content_line.getBuffer().limit() && (
556
content_line.getBuffer().get( pos ) == ' ' ||
557
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' ) )
558
492
{
559
493
pos++;
560
494
}
561
break;
562
default:
563
// do nothing
564
}
565
566
// take us out of multi-line so that we can re-detect that
567
// this line is a multi-line or not
568
_parser_multiline_state = MULTILINE_NONE;
569
495
}
570
496
else
571
497
{
572
// skip empty lines
573
if( content_line.getUsAsciiLine().trim().length() == 0 )
574
return;
575
576
498
// get name and params from line, and since we're not
577
499
// parsing a subsequent line in a multi-line, this should
578
500
// not fail, or it's an error
579
501
name_and_params =
580
extractNameAndParamsFromLine( content_line );
502
extractNameAndParamsFromLine( buffer, line );
581
503
if( name_and_params == null )
582
504
throw new ParseException(
583
505
R.string.error_vcf_malformed );
584
506
585
507
// calculate how many chars to skip from beginning of line
586
508
// so we skip the property "name:" part
587
pos = content_line.getBuffer().position() +
588
name_and_params.length() + 1;
509
pos = buffer.position() + name_and_params.length() + 1;
589
510
590
511
// reset the saved multi-line state
591
512
_parser_current_name_and_params = name_and_params;
594
515
595
516
// get value from buffer, as raw bytes
596
517
ByteBuffer value;
597
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
598
content_line.getBuffer().limit() - pos );
518
value = ByteBuffer.wrap( buffer.array(), pos,
519
buffer.limit() - pos );
599
520
600
521
// get parameter parts
601
522
String[] name_param_parts = name_and_params.split( ";", -1 );
602
523
for( int i = 0; i < name_param_parts.length; i++ )
603
524
name_param_parts[ i ] = name_param_parts[ i ].trim();
604
525
605
// determine whether we care about this entry
606
final HashSet< String > interesting_fields =
607
new HashSet< String >( Arrays.asList( new String[] { "N",
608
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
609
) );
610
boolean is_interesting_field =
611
interesting_fields.contains(
612
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
613
614
526
// parse encoding parameter
615
527
String encoding = checkParam( name_param_parts, "ENCODING" );
616
if( encoding != null )
617
encoding = encoding.toUpperCase( Locale.US );
618
if( is_interesting_field && encoding != null &&
619
!encoding.equalsIgnoreCase( "8BIT" ) &&
620
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
621
//&& !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" ) )
622
532
{
623
533
throw new ParseException( R.string.error_vcf_encoding );
624
534
}
625
535
626
536
// parse charset parameter
627
537
String charset = checkParam( name_param_parts, "CHARSET" );
628
if( charset != null )
629
charset = charset.toUpperCase( Locale.US );
630
if( charset != null &&
631
!charset.equalsIgnoreCase( "US-ASCII" ) &&
632
!charset.equalsIgnoreCase( "ASCII" ) &&
633
!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" ) )
634
542
{
635
543
throw new ParseException( R.string.error_vcf_charset );
636
544
}
638
546
// do unencoding (or default to a fake unencoding result with
639
547
// the raw string)
640
548
UnencodeResult unencoding_result = null;
641
if( encoding != null &&
642
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
643
{
549
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
644
550
unencoding_result = unencodeQuotedPrintable( value );
645
}
646
// else if( encoding != null &&
647
// encoding.equalsIgnoreCase( "BASE64" ) )
648
// {
551
// else if( encoding != null && encoding.equals( "BASE64" ) )
649
552
// unencoding_result = unencodeBase64( props[ 1 ], charset );
650
// }
651
553
if( unencoding_result != null ) {
652
554
value = unencoding_result.getBuffer();
653
if( unencoding_result.isAnotherLineRequired() )
654
_parser_multiline_state = MULTILINE_ENCODED;
555
_parser_in_encoded_multiline =
556
unencoding_result.isAnotherLineRequired();
655
557
}
656
558
657
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
658
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
659
if( ( charset == null && _version.equals( "2.1" ) ) ||
660
( charset != null && (
661
charset.equalsIgnoreCase( "ASCII" ) ||
662
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
663
{
559
// convert 8-bit ASCII charset to US-ASCII
560
if( charset == null || charset.equals( "ASCII" ) ) {
664
561
value = transcodeAsciiToUtf8( value );
562
charset = "UTF-8";
665
563
}
666
564
667
// process charset (value is now in UTF-8)
565
// process charset
668
566
String string_value;
669
567
try {
670
568
string_value = new String( value.array(), value.position(),
671
value.limit() - value.position(), "UTF-8" );
569
value.limit() - value.position(), charset );
672
570
} catch( UnsupportedEncodingException e ) {
673
571
throw new ParseException( R.string.error_vcf_charset );
674
572
}
675
573
676
// for some entries that have semicolon-separated value parts,
677
// check to see if the value ends in an escape character, which
678
// indicates that we have a multi-line value
679
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
680
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
681
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
682
doesStringEndInAnEscapeChar( string_value ) )
683
{
684
_parser_multiline_state = MULTILINE_ESCAPED;
685
string_value = string_value.substring( 0,
686
string_value.length() - 1 );
687
}
688
689
// if we know we're not in an encoding-based multi-line, check
690
// to see if we're in a folded multi-line
691
if( _parser_multiline_state == MULTILINE_NONE &&
692
content_line.doesNextLineLookFolded() )
693
{
694
_parser_multiline_state = MULTILINE_FOLDED;
695
}
696
697
// handle multi-lines by buffering them and parsing them when we
698
// are processing the last line in a multi-line sequence
699
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
{
700
583
_parser_buffered_value_so_far += string_value;
701
584
return;
702
585
}
586
587
// add on buffered multi-line content
703
588
String complete_value =
704
( _parser_buffered_value_so_far + string_value ).trim();
589
_parser_buffered_value_so_far + string_value;
705
590
706
591
// ignore empty values
707
592
if( complete_value.length() < 1 ) return;
708
593
709
594
// parse some properties
710
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
595
if( name_param_parts[ 0 ].equals( "N" ) )
711
596
parseN( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
597
else if( name_param_parts[ 0 ].equals( "FN" ) )
713
598
parseFN( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
599
else if( name_param_parts[ 0 ].equals( "ORG" ) )
715
600
parseORG( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
717
parseTITLE( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
601
else if( name_param_parts[ 0 ].equals( "TEL" ) )
719
602
parseTEL( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
603
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
721
604
parseEMAIL( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
723
parseADR( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
725
parseLABEL( name_param_parts, complete_value );
726
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
727
parseNOTE( name_param_parts, complete_value );
728
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
729
parseBDAY( name_param_parts, complete_value );
730
}
731
}
732
733
private boolean doesStringEndInAnEscapeChar( String string )
734
{
735
// count the number of backslashes at the end of the string
736
int count = 0;
737
for( int a = string.length() - 1; a >= 0; a-- )
738
if( string.charAt( a ) == '\\' )
739
count++;
740
else
741
break;
742
743
// if there are an even number of backslashes then the final one
744
// doesn't count
745
return ( count & 1 ) == 1;
746
}
747
748
private String[] splitValueByCharacter( String value, char character )
749
{
750
// split string in to parts by specified character
751
ArrayList< String > parts = new ArrayList< String >(
752
Arrays.asList( value.split( "" + character ) ) );
753
754
// go through parts
755
for( int a = 0; a < parts.size(); a++ )
756
{
757
String str = parts.get( a );
758
759
// Look for parts that end in an escape character, but ignore
760
// the final part. We've already detected escape chars at the
761
// end of the final part in parseLine() and handled multi-lines
762
// accordingly.
763
if( a < parts.size() - 1 &&
764
doesStringEndInAnEscapeChar( str ) )
765
{
766
// append the escaped character, join the next part to this
767
// part and remove the next part
768
parts.set( a, str.substring( 0, str.length() - 1 ) +
769
character + parts.get( a + 1 ) );
770
parts.remove( a + 1 );
771
772
// re-visit this part
773
a--;
774
continue;
775
}
776
777
// trim and replace string
778
str = str.trim();
779
parts.set( a, str );
780
}
781
782
String[] ret = new String[ parts.size() ];
783
return parts.toArray( ret );
784
}
785
786
private String unescapeValue( String value )
787
{
788
StringBuilder ret = new StringBuilder( value.length() );
789
boolean in_escape = false;
790
for( int a = 0; a < value.length(); a++ )
791
{
792
int c = value.codePointAt( a );
793
794
// process a normal character
795
if( !in_escape ) {
796
if( c == '\\' )
797
in_escape = true;
798
else
799
ret.append( Character.toChars( c ) );
800
continue;
801
}
802
803
// process an escape sequence
804
in_escape = false;
805
switch( c )
806
{
807
case 'T':
808
case 't':
809
// add tab (invalid/non-standard, but accepted)
810
ret.append( '\t' );
811
break;
812
case 'N':
813
case 'n':
814
// add newline
815
ret.append( '\n' );
816
break;
817
case '\\':
818
case ',':
819
case ';':
820
// add escaped character
821
ret.append( Character.toChars( c ) );
822
break;
823
default:
824
// unknown escape sequence, so add it unescaped
825
// (invalid/non-standard, but accepted)
826
ret.append( "\\" );
827
ret.append( Character.toChars( c ) );
828
break;
829
}
830
}
831
832
return ret.toString();
605
}
833
606
}
834
607
835
608
private void parseN( String[] params, String value )
609
throws ParseException, SkipContactException,
610
AbortImportException
836
611
{
837
612
// already got a better name?
838
613
if( _name_level >= NAMELEVEL_N ) return;
839
614
840
615
// get name parts
841
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();
842
619
843
620
// build name
844
621
value = "";
845
final int[] part_order = { 3, 1, 2, 0, 4 };
846
for( int a = 0; a < part_order.length; a++ )
847
if( name_parts.length > part_order[ a ] &&
848
name_parts[ part_order[ a ] ].length() > 0 )
849
{
850
// split this part in to it's comma-separated bits
851
String[] name_part_parts = splitValueByCharacter(
852
name_parts[ part_order[ a ] ], ',' );
853
for( int b = 0; b < name_part_parts.length; b++ )
854
if( name_part_parts[ b ].length() > 0 )
855
{
856
if( value.length() > 0 ) value += " ";
857
value += name_part_parts[ b ];
858
}
859
}
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 ];
860
626
861
627
// set name
862
setName( unescapeValue( value ) );
628
setName( value );
863
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();
864
635
}
865
636
866
637
private void parseFN( String[] params, String value )
638
throws ParseException, SkipContactException
867
639
{
868
640
// already got a better name?
869
641
if( _name_level >= NAMELEVEL_FN ) return;
870
642
871
643
// set name
872
setName( unescapeValue( value ) );
644
setName( value );
873
645
_name_level = NAMELEVEL_FN;
874
646
}
875
647
876
648
private void parseORG( String[] params, String value )
649
throws ParseException, SkipContactException
877
650
{
651
// already got a better name?
652
if( _name_level >= NAMELEVEL_ORG ) return;
653
878
654
// get org parts
879
String[] org_parts = splitValueByCharacter( value, ';' );
880
if( org_parts == null || org_parts.length < 1 ) return;
881
882
// build organisation name
883
StringBuilder builder = new StringBuilder(
884
String.valueOf( org_parts[ 0 ] ) );
885
for( int a = 1; a < org_parts.length; a++ )
886
builder.append( ", " ).append( org_parts[ a ] );
887
String organisation = unescapeValue( builder.toString() );
888
889
// set organisation name (using a title we've previously found)
890
addOrganisation( organisation, _cached_title, true );
891
892
// if we've not previously found a title, store this organisation
893
// name (we'll need it when we find a title to update the
894
// organisation, by name), else if we *have* previously found a
895
// title, clear it (since we just used it)
896
if( _cached_title == null )
897
_cached_organisation = organisation;
898
else
899
_cached_title = null;
900
}
901
902
private void parseTITLE( String[] params, String value )
903
{
904
value = unescapeValue( value );
905
906
// if we previously had an organisation, look it up and append this
907
// title to it
908
if( _cached_organisation != null && hasOrganisations() ) {
909
HashMap< String, ExtraDetail > datas = getOrganisations();
910
ExtraDetail detail = datas.get( _cached_organisation );
911
if( detail != null )
912
detail.setExtra( value );
913
}
914
915
// same as when handling organisation, if we've not previously found
916
// an organisation we store this title, else we clear it (since we
917
// just appended this title to it)
918
if( _cached_organisation == null )
919
_cached_title = value;
920
else
921
_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;
922
668
}
923
669
924
670
private void parseTEL( String[] params, String value )
671
throws ParseException
925
672
{
926
673
if( value.length() == 0 ) return;
927
674
930
677
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
931
678
932
679
// here's the logic...
933
boolean is_preferred = types.contains( "PREF" );
934
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;
935
689
if( types.contains( "FAX" ) )
936
690
if( types.contains( "HOME" ) )
937
type = TYPE_FAX_HOME;
691
type = PhonesColumns.TYPE_FAX_HOME;
938
692
else
939
type = TYPE_FAX_WORK;
940
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
941
type = TYPE_MOBILE;
942
else if( types.contains( "PAGER" ) )
943
type = TYPE_PAGER;
944
else if( types.contains( "WORK" ) )
945
type = TYPE_WORK;
946
else
947
type = TYPE_HOME;
693
type = PhonesColumns.TYPE_FAX_WORK;
694
if( types.contains( "PAGER" ) )
695
type = PhonesColumns.TYPE_PAGER;
948
696
949
697
// add phone number
950
addNumber( value, type, is_preferred );
698
addPhone( value, type, preferred );
951
699
}
952
700
953
701
public void parseEMAIL( String[] params, String value )
702
throws ParseException
954
703
{
955
704
if( value.length() == 0 ) return;
956
705
957
706
Set< String > types = extractTypes( params, Arrays.asList(
958
707
"PREF", "WORK", "HOME", "INTERNET" ) );
959
708
960
// add email address
961
boolean is_preferred = types.contains( "PREF" );
962
int type;
963
if( types.contains( "WORK" ) )
964
type = TYPE_WORK;
965
else
966
type = TYPE_HOME;
967
968
addEmail( unescapeValue( value ), type, is_preferred );
969
}
970
971
private void parseADR( String[] params, String value )
972
{
973
// get address parts
974
String[] adr_parts = splitValueByCharacter( value, ';' );
975
976
// build address
977
value = "";
978
for( int a = 0; a < adr_parts.length; a++ )
979
if( adr_parts[ a ].length() > 0 )
980
{
981
// version 3.0 vCards allow further splitting by comma
982
if( _version.equals( "3.0" ) )
983
{
984
// split this part in to it's comma-separated bits and
985
// add them on individual lines
986
String[] adr_part_parts =
987
splitValueByCharacter( adr_parts[ a ], ',' );
988
for( int b = 0; b < adr_part_parts.length; b++ )
989
if( adr_part_parts[ b ].length() > 0 )
990
{
991
if( value.length() > 0 ) value += "\n";
992
value += adr_part_parts[ b ];
993
}
994
}
995
else
996
{
997
// add this part on an individual line
998
if( value.length() > 0 ) value += "\n";
999
value += adr_parts[ a ];
1000
}
1001
}
1002
1003
Set< String > types = extractTypes( params, Arrays.asList(
1004
"PREF", "WORK", "HOME" ) );
1005
1006
// add address
1007
int type;
1008
if( types.contains( "WORK" ) )
1009
type = TYPE_WORK;
1010
else
1011
type = TYPE_HOME;
1012
1013
addAddress( unescapeValue( value ), type );
1014
}
1015
1016
private void parseLABEL( String[] params, String value )
1017
{
1018
Set< String > types = extractTypes( params, Arrays.asList(
1019
"PREF", "WORK", "HOME" ) );
1020
1021
// add address
1022
int type;
1023
if( types.contains( "WORK" ) )
1024
type = TYPE_WORK;
1025
else
1026
type = TYPE_HOME;
1027
1028
addAddress( unescapeValue( value ), type );
1029
}
1030
1031
private void parseNOTE( String[] params, String value )
1032
{
1033
addNote( unescapeValue( value ) );
1034
}
1035
1036
private void parseBDAY( String[] params, String value )
1037
{
1038
setBirthday( value );
1039
}
1040
1041
public void finaliseVcard()
1042
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
1043
720
{
1044
721
// missing version (and data is present)
1045
if( _version == null && _content_lines != null )
722
if( _version == null && _buffers != null )
1046
723
throw new ParseException( R.string.error_vcf_malformed );
1047
724
1048
// finalise the parent class
1049
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();
1050
734
}
1051
735
1052
/**
1053
* Amongst the params, find the value of the first, only, of any with
1054
* the specified name.
1055
*
1056
* @param params
1057
* @param name
1058
* @return a value, or null
1059
*/
1060
736
private String checkParam( String[] params, String name )
1061
737
{
1062
String[] res = checkParams( params, name );
1063
return res.length > 0? res[ 0 ] : null;
1064
}
1065
1066
/**
1067
* Amongst the params, find the values of any with the specified name.
1068
*
1069
* @param params
1070
* @param name
1071
* @return an array of values, or null
1072
*/
1073
private String[] checkParams( String[] params, String name )
1074
{
1075
HashSet< String > ret = new HashSet< String >();
1076
1077
738
Pattern p = Pattern.compile(
1078
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1079
Pattern.CASE_INSENSITIVE );
739
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1080
740
for( int i = 0; i < params.length; i++ ) {
1081
741
Matcher m = p.matcher( params[ i ] );
1082
742
if( m.matches() )
1083
ret.add( m.group( 2 ) );
743
return m.group( 2 );
1084
744
}
1085
1086
return (String[]) ret.toArray( new String[ ret.size() ] );
745
return null;
1087
746
}
1088
747
1089
/**
1090
* Amongst the params, return any type values present. For v2.1 vCards,
1091
* those types are just parameters. For v3.0, they are prefixed with
1092
* "TYPE=". There may also be multiple type parameters.
1093
*
1094
* @param params an array of params to look for types in
1095
* @param valid_types an list of upper-case type values to look for
1096
* @return a set of present type values
1097
*/
1098
748
private Set< String > extractTypes( String[] params,
1099
749
List< String > valid_types )
1100
750
{
1101
751
HashSet< String > types = new HashSet< String >();
1102
752
1103
753
// get 3.0-style TYPE= param
1104
String type_params[] = checkParams( params, "TYPE" );
1105
for( int a = 0; a < type_params.length; a++ )
1106
{
1107
// check for a comma-separated list of types (why? I don't think
1108
// this is in the specs!)
1109
String[] parts = type_params[ a ].split( "," );
1110
for( int i = 0; i < parts.length; i++ ) {
1111
String ucpart = parts[ i ].toUpperCase( Locale.US );
1112
if( valid_types.contains( ucpart ) )
1113
types.add( ucpart );
1114
}
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 ] );
1115
760
}
1116
761
1117
762
// get 2.1-style type param
1118
763
if( _version.equals( "2.1" ) ) {
1119
for( int i = 1; i < params.length; i++ ) {
1120
String ucparam = params[ i ].toUpperCase( Locale.US );
1121
if( valid_types.contains( ucparam ) )
1122
types.add( ucparam );
1123
}
764
for( int i = 1; i < params.length; i++ )
765
if( valid_types.contains( params[ i ] ) )
766
types.add( params[ i ] );
1124
767
}
1125
768
1126
769
return types;
1148
791
else if( ch == '=' && i == in.limit() - 1 )
1149
792
{
1150
793
// we found a '=' at the end of a line signifying a multi-
1151
// line string, so we don't add it
794
// line string, so we don't add it.
1152
795
another = true;
1153
796
continue;
1154
797
}
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480