(root)/android/import-contacts
» Revision
36
(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 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:
- .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
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
}
275
250
276
// Although we're continuing, we still need to abort
277
// 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
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
// Abort this vCard. Further lines will be ignored until
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
{
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
474
404
// split line into name and value parts and check to make sure we
475
405
// only got 2 parts and that the first part is not zero in length
476
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
406
String[] parts = line.split( ":", 2 );
477
407
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,
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,
495
427
AbortImportException
496
428
{
497
429
// do we have a version yet?
499
431
{
500
432
// tentatively get name and params from line
501
433
String name_and_params =
502
extractNameAndParamsFromLine( content_line );
434
extractNameAndParamsFromLine( buffer, line );
503
435
504
436
// is it a version line?
505
437
if( name_and_params != null &&
506
name_and_params.equalsIgnoreCase( "VERSION" ) )
438
name_and_params.equals( "VERSION" ) )
507
439
{
508
440
// yes, get it!
509
String value = extractValueFromLine( content_line );
510
if( value == null || (
511
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
512
{
441
String value = extractValueFromLine( buffer, line );
442
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
443
throw new ParseException( R.string.error_vcf_version );
514
}
515
444
_version = value;
516
445
517
446
// parse any buffers we've been accumulating while we waited
518
447
// 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;
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;
523
456
}
524
457
else
525
458
{
526
459
// 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 );
460
if( _buffers == null )
461
_buffers = new Vector< ByteBuffer >();
462
_buffers.add( buffer );
530
463
}
531
464
}
532
465
else
533
466
{
534
467
// name and params and the position in the buffer where the
535
// "value" part of the line starts
468
// "value" part of the line start
536
469
String name_and_params;
537
470
int pos;
538
471
539
if( _parser_multiline_state != MULTILINE_NONE )
472
if( _parser_in_encoded_multiline ||
473
_parser_in_folded_multiline )
540
474
{
541
475
// if we're currently in a multi-line value, use the stored
542
476
// property name and parameters
543
477
name_and_params = _parser_current_name_and_params;
544
478
545
// skip some initial line characters, depending on the type
546
// of multi-line we're handling
547
pos = content_line.getBuffer().position();
548
switch( _parser_multiline_state )
549
{
550
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 )
551
484
pos++;
552
break;
553
case MULTILINE_ENCODED:
554
while( pos < content_line.getBuffer().limit() && (
555
content_line.getBuffer().get( pos ) == ' ' ||
556
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' ) )
557
492
{
558
493
pos++;
559
494
}
560
break;
561
default:
562
// do nothing
563
}
564
565
// take us out of multi-line so that we can re-detect that
566
// this line is a multi-line or not
567
_parser_multiline_state = MULTILINE_NONE;
568
495
}
569
496
else
570
497
{
571
// skip empty lines
572
if( content_line.getUsAsciiLine().trim().length() == 0 )
573
return;
574
575
498
// get name and params from line, and since we're not
576
499
// parsing a subsequent line in a multi-line, this should
577
500
// not fail, or it's an error
578
501
name_and_params =
579
extractNameAndParamsFromLine( content_line );
502
extractNameAndParamsFromLine( buffer, line );
580
503
if( name_and_params == null )
581
504
throw new ParseException(
582
505
R.string.error_vcf_malformed );
583
506
584
507
// calculate how many chars to skip from beginning of line
585
508
// so we skip the property "name:" part
586
pos = content_line.getBuffer().position() +
587
name_and_params.length() + 1;
509
pos = buffer.position() + name_and_params.length() + 1;
588
510
589
511
// reset the saved multi-line state
590
512
_parser_current_name_and_params = name_and_params;
593
515
594
516
// get value from buffer, as raw bytes
595
517
ByteBuffer value;
596
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
597
content_line.getBuffer().limit() - pos );
518
value = ByteBuffer.wrap( buffer.array(), pos,
519
buffer.limit() - pos );
598
520
599
521
// get parameter parts
600
522
String[] name_param_parts = name_and_params.split( ";", -1 );
601
523
for( int i = 0; i < name_param_parts.length; i++ )
602
524
name_param_parts[ i ] = name_param_parts[ i ].trim();
603
525
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
526
// parse encoding parameter
614
527
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" ) )
528
if( encoding != null ) encoding = encoding.toUpperCase();
529
if( encoding != null && !encoding.equals( "8BIT" ) &&
530
!encoding.equals( "QUOTED-PRINTABLE" ) )
531
//&& !encoding.equals( "BASE64" ) )
621
532
{
622
533
throw new ParseException( R.string.error_vcf_encoding );
623
534
}
624
535
625
536
// parse charset parameter
626
537
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" ) )
538
if( charset != null ) charset = charset.toUpperCase();
539
if( charset != null && !charset.equals( "US-ASCII" ) &&
540
!charset.equals( "ASCII" ) &&
541
!charset.equals( "UTF-8" ) )
633
542
{
634
543
throw new ParseException( R.string.error_vcf_charset );
635
544
}
637
546
// do unencoding (or default to a fake unencoding result with
638
547
// the raw string)
639
548
UnencodeResult unencoding_result = null;
640
if( encoding != null &&
641
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
642
{
549
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
643
550
unencoding_result = unencodeQuotedPrintable( value );
644
}
645
// else if( encoding != null &&
646
// encoding.equalsIgnoreCase( "BASE64" ) )
647
// {
551
// else if( encoding != null && encoding.equals( "BASE64" ) )
648
552
// unencoding_result = unencodeBase64( props[ 1 ], charset );
649
// }
650
553
if( unencoding_result != null ) {
651
554
value = unencoding_result.getBuffer();
652
if( unencoding_result.isAnotherLineRequired() )
653
_parser_multiline_state = MULTILINE_ENCODED;
555
_parser_in_encoded_multiline =
556
unencoding_result.isAnotherLineRequired();
654
557
}
655
558
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
{
559
// convert 8-bit ASCII charset to US-ASCII
560
if( charset == null || charset.equals( "ASCII" ) ) {
663
561
value = transcodeAsciiToUtf8( value );
562
charset = "UTF-8";
664
563
}
665
564
666
// process charset (value is now in UTF-8)
565
// process charset
667
566
String string_value;
668
567
try {
669
568
string_value = new String( value.array(), value.position(),
670
value.limit() - value.position(), "UTF-8" );
569
value.limit() - value.position(), charset );
671
570
} catch( UnsupportedEncodingException e ) {
672
571
throw new ParseException( R.string.error_vcf_charset );
673
572
}
674
573
675
// for some entries that have semicolon-separated value parts,
676
// check to see if the value ends in an escape character, which
677
// 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" ) ) &&
681
doesStringEndInAnEscapeChar( string_value ) )
682
{
683
_parser_multiline_state = MULTILINE_ESCAPED;
684
string_value = string_value.substring( 0,
685
string_value.length() - 1 );
686
}
687
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
690
if( _parser_multiline_state == MULTILINE_NONE &&
691
content_line.doesNextLineLookFolded() )
692
{
693
_parser_multiline_state = MULTILINE_FOLDED;
694
}
695
696
// handle multi-lines by buffering them and parsing them when we
697
// are processing the last line in a multi-line sequence
698
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
{
699
583
_parser_buffered_value_so_far += string_value;
700
584
return;
701
585
}
586
587
// add on buffered multi-line content
702
588
String complete_value =
703
( _parser_buffered_value_so_far + string_value ).trim();
589
_parser_buffered_value_so_far + string_value;
704
590
705
591
// ignore empty values
706
592
if( complete_value.length() < 1 ) return;
707
593
708
594
// parse some properties
709
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
595
if( name_param_parts[ 0 ].equals( "N" ) )
710
596
parseN( name_param_parts, complete_value );
711
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
597
else if( name_param_parts[ 0 ].equals( "FN" ) )
712
598
parseFN( name_param_parts, complete_value );
713
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
599
else if( name_param_parts[ 0 ].equals( "ORG" ) )
714
600
parseORG( name_param_parts, complete_value );
715
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
716
parseTITLE( name_param_parts, complete_value );
717
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
601
else if( name_param_parts[ 0 ].equals( "TEL" ) )
718
602
parseTEL( name_param_parts, complete_value );
719
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
603
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
720
604
parseEMAIL( name_param_parts, complete_value );
721
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
722
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
}
730
}
731
732
private boolean doesStringEndInAnEscapeChar( String string )
733
{
734
// count the number of backslashes at the end of the string
735
int count = 0;
736
for( int a = string.length() - 1; a >= 0; a-- )
737
if( string.charAt( a ) == '\\' )
738
count++;
739
else
740
break;
741
742
// if there are an even number of backslashes then the final one
743
// doesn't count
744
return ( count & 1 ) == 1;
745
}
746
747
private String[] splitValueByCharacter( String value, char character )
748
{
749
// split string in to parts by specified character
750
ArrayList< String > parts = new ArrayList< String >(
751
Arrays.asList( value.split( "" + character ) ) );
752
753
// go through parts
754
for( int a = 0; a < parts.size(); a++ )
755
{
756
String str = parts.get( a );
757
758
// Look for parts that end in an escape character, but ignore
759
// the final part. We've already detected escape chars at the
760
// end of the final part in parseLine() and handled multi-lines
761
// accordingly.
762
if( a < parts.size() - 1 &&
763
doesStringEndInAnEscapeChar( str ) )
764
{
765
// append the escaped character, join the next part to this
766
// part and remove the next part
767
parts.set( a, str.substring( 0, str.length() - 1 ) +
768
character + parts.get( a + 1 ) );
769
parts.remove( a + 1 );
770
771
// re-visit this part
772
a--;
773
continue;
774
}
775
776
// trim and replace string
777
str = str.trim();
778
parts.set( a, str );
779
}
780
781
String[] ret = new String[ parts.size() ];
782
return parts.toArray( ret );
783
}
784
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();
605
}
832
606
}
833
607
834
608
private void parseN( String[] params, String value )
609
throws ParseException, SkipContactException,
610
AbortImportException
835
611
{
836
612
// already got a better name?
837
613
if( _name_level >= NAMELEVEL_N ) return;
838
614
839
615
// get name parts
840
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();
841
619
842
620
// build name
843
621
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
}
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 ];
859
626
860
627
// set name
861
setName( unescapeValue( value ) );
628
setName( value );
862
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();
863
635
}
864
636
865
637
private void parseFN( String[] params, String value )
638
throws ParseException, SkipContactException
866
639
{
867
640
// already got a better name?
868
641
if( _name_level >= NAMELEVEL_FN ) return;
869
642
870
643
// set name
871
setName( unescapeValue( value ) );
644
setName( value );
872
645
_name_level = NAMELEVEL_FN;
873
646
}
874
647
875
648
private void parseORG( String[] params, String value )
649
throws ParseException, SkipContactException
876
650
{
651
// already got a better name?
652
if( _name_level >= NAMELEVEL_ORG ) return;
653
877
654
// get org parts
878
String[] org_parts = splitValueByCharacter( value, ';' );
879
if( org_parts == null || org_parts.length < 1 ) return;
880
881
// build organisation name
882
StringBuilder builder = new StringBuilder(
883
String.valueOf( org_parts[ 0 ] ) );
884
for( int a = 1; a < org_parts.length; a++ )
885
builder.append( ", " ).append( org_parts[ a ] );
886
String organisation = unescapeValue( builder.toString() );
887
888
// set organisation name (using a title we've previously found)
889
addOrganisation( organisation, _cached_title, true );
890
891
// if we've not previously found a title, store this organisation
892
// name (we'll need it when we find a title to update the
893
// organisation, by name), else if we *have* previously found a
894
// title, clear it (since we just used it)
895
if( _cached_title == null )
896
_cached_organisation = organisation;
897
else
898
_cached_title = null;
899
}
900
901
private void parseTITLE( String[] params, String value )
902
{
903
value = unescapeValue( value );
904
905
// if we previously had an organisation, look it up and append this
906
// title to it
907
if( _cached_organisation != null && hasOrganisations() ) {
908
HashMap< String, ExtraDetail > datas = getOrganisations();
909
ExtraDetail detail = datas.get( _cached_organisation );
910
if( detail != null )
911
detail.setExtra( value );
912
}
913
914
// same as when handling organisation, if we've not previously found
915
// an organisation we store this title, else we clear it (since we
916
// just appended this title to it)
917
if( _cached_organisation == null )
918
_cached_title = value;
919
else
920
_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;
921
668
}
922
669
923
670
private void parseTEL( String[] params, String value )
671
throws ParseException
924
672
{
925
673
if( value.length() == 0 ) return;
926
674
929
677
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
930
678
931
679
// here's the logic...
932
boolean is_preferred = types.contains( "PREF" );
933
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;
934
689
if( types.contains( "FAX" ) )
935
690
if( types.contains( "HOME" ) )
936
type = TYPE_FAX_HOME;
691
type = PhonesColumns.TYPE_FAX_HOME;
937
692
else
938
type = TYPE_FAX_WORK;
939
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
940
type = TYPE_MOBILE;
941
else if( types.contains( "PAGER" ) )
942
type = TYPE_PAGER;
943
else if( types.contains( "WORK" ) )
944
type = TYPE_WORK;
945
else
946
type = TYPE_HOME;
693
type = PhonesColumns.TYPE_FAX_WORK;
694
if( types.contains( "PAGER" ) )
695
type = PhonesColumns.TYPE_PAGER;
947
696
948
697
// add phone number
949
addNumber( value, type, is_preferred );
698
addPhone( value, type, preferred );
950
699
}
951
700
952
701
public void parseEMAIL( String[] params, String value )
702
throws ParseException
953
703
{
954
704
if( value.length() == 0 ) return;
955
705
956
706
Set< String > types = extractTypes( params, Arrays.asList(
957
707
"PREF", "WORK", "HOME", "INTERNET" ) );
958
708
959
// add email address
960
boolean is_preferred = types.contains( "PREF" );
961
int type;
962
if( types.contains( "WORK" ) )
963
type = TYPE_WORK;
964
else
965
type = TYPE_HOME;
966
967
addEmail( unescapeValue( value ), type, is_preferred );
968
}
969
970
private void parseADR( String[] params, String value )
971
{
972
// get address parts
973
String[] adr_parts = splitValueByCharacter( value, ';' );
974
975
// build address
976
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
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
1042
720
{
1043
721
// missing version (and data is present)
1044
if( _version == null && _content_lines != null )
722
if( _version == null && _buffers != null )
1045
723
throw new ParseException( R.string.error_vcf_malformed );
1046
724
1047
// finalise the parent class
1048
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();
1049
734
}
1050
735
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
736
private String checkParam( String[] params, String name )
1060
737
{
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
738
Pattern p = Pattern.compile(
1077
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1078
Pattern.CASE_INSENSITIVE );
739
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1079
740
for( int i = 0; i < params.length; i++ ) {
1080
741
Matcher m = p.matcher( params[ i ] );
1081
742
if( m.matches() )
1082
ret.add( m.group( 2 ) );
743
return m.group( 2 );
1083
744
}
1084
1085
return (String[]) ret.toArray( new String[ ret.size() ] );
745
return null;
1086
746
}
1087
747
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
748
private Set< String > extractTypes( String[] params,
1098
749
List< String > valid_types )
1099
750
{
1100
751
HashSet< String > types = new HashSet< String >();
1101
752
1102
753
// 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
}
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 ] );
1114
760
}
1115
761
1116
762
// get 2.1-style type param
1117
763
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
}
764
for( int i = 1; i < params.length; i++ )
765
if( valid_types.contains( params[ i ] ) )
766
types.add( params[ i ] );
1123
767
}
1124
768
1125
769
return types;
1147
791
else if( ch == '=' && i == in.limit() - 1 )
1148
792
{
1149
793
// we found a '=' at the end of a line signifying a multi-
1150
// line string, so we don't add it
794
// line string, so we don't add it.
1151
795
another = true;
1152
796
continue;
1153
797
}
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480