/android/import-contacts

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/am/ed/importcontacts/VcardImporter.java

  • Committer: edam
  • Date: 2012-12-23 09:57:22 UTC
  • Revision ID: tim@ed.am-20121223095722-ep12gv4433ahh6vz
Tags: 1.3
fixed version no. in news

Show diffs side-by-side

added added

removed removed

1
1
/*
2
2
 * VCFImporter.java
3
3
 *
4
 
 * Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
 
4
 * Copyright (C) 2009 to 2011 Tim Marston <tim@ed.am>
5
5
 *
6
6
 * This file is part of the Import Contacts program (hereafter referred
7
7
 * to as "this program"). For more information, see
8
 
 * http://www.waxworlds.org/edam/software/android/import-contacts
 
8
 * http://ed.am/dev/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 org.waxworlds.edam.importcontacts;
 
24
package am.ed.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;
41
42
import java.util.NoSuchElementException;
42
43
import java.util.Set;
43
44
import java.util.Vector;
44
45
import java.util.regex.Matcher;
45
46
import java.util.regex.Pattern;
46
47
 
 
48
import android.annotation.SuppressLint;
47
49
import android.content.SharedPreferences;
48
 
import android.provider.Contacts;
49
 
import android.provider.Contacts.PhonesColumns;
50
50
 
51
 
public class VCFImporter extends Importer
 
51
public class VcardImporter extends Importer
52
52
{
53
53
        private int _vcard_count = 0;
54
54
        private int _progress = 0;
55
55
 
56
 
        public VCFImporter( Doit doit )
 
56
        public VcardImporter( Doit doit )
57
57
        {
58
58
                super( doit );
59
59
        }
60
60
 
 
61
        @SuppressLint( "SdCardPath" )
61
62
        @Override
62
63
        protected void onImport() throws AbortImportException
63
64
        {
82
83
                                // get files
83
84
                                class VCardFilter implements FilenameFilter {
84
85
                                        public boolean accept( File dir, String name ) {
85
 
                                                return name.toLowerCase().endsWith( ".vcf" );
 
86
                                                return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
86
87
                                        }
87
88
                                }
88
89
                                files = file.listFiles( new VCardFilter() );
116
117
                setProgress( 0 );
117
118
                for( int i = 0; i < files.length; i++ )
118
119
                        importVCardFile( files[ i ] );
 
120
                setProgress( _vcard_count );
119
121
        }
120
122
 
121
123
        private void countVCardFile( File file ) throws AbortImportException
168
170
                        FileInputStream istream = new FileInputStream( file );
169
171
                        byte[] content = new byte[ (int)file.length() ];
170
172
                        istream.read( content );
 
173
                        istream = null;
171
174
 
172
175
                        // import
173
176
                        importVCardFileContent( content, file.getName() );
174
177
                }
 
178
                catch( OutOfMemoryError e ) {
 
179
                        showError( R.string.error_outofmemory );
 
180
                }
175
181
                catch( FileNotFoundException e ) {
176
182
                        showError( getText( R.string.error_filenotfound ) +
177
183
                                file.getName() );
185
191
                throws AbortImportException
186
192
        {
187
193
                // go through lines
188
 
                VCard vcard = null;
 
194
                Vcard vcard = null;
 
195
                int vcard_start_line = 0;
189
196
                ContentLineIterator cli = new ContentLineIterator( content );
190
197
                while( cli.hasNext() )
191
198
                {
192
 
                        ByteBuffer buffer = cli.next();
 
199
                        ContentLine content_line = cli.next();
193
200
 
194
 
                        // get a US-ASCII version of the line for processing
195
 
                        String line;
196
 
                        try {
197
 
                                line = new String( buffer.array(), buffer.position(),
198
 
                                        buffer.limit() - buffer.position(), "US-ASCII" );
199
 
                        }
200
 
                        catch( UnsupportedEncodingException e ) {
201
 
                                // we know US-ASCII is supported, so appease the compiler...
202
 
                                line = "";
203
 
                        }
 
201
                        // get a US-ASCII version of the string, for processing
 
202
                        String line = content_line.getUsAsciiLine();
204
203
 
205
204
                        if( vcard == null ) {
206
205
                                // look for vcard beginning
207
 
                                if( line.matches( "^BEGIN:VCARD" ) ) {
208
 
                                        setProgress( ++_progress );
209
 
                                        vcard = new VCard();
 
206
                                if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD" ) ) {
 
207
                                        setProgress( _progress++ );
 
208
                                        vcard = new Vcard();
 
209
                                        vcard_start_line = cli.getLineNumber();
210
210
                                }
211
211
                        }
212
212
                        else {
213
213
                                // look for vcard content or ending
214
 
                                if( line.matches( "^END:VCARD" ) )
 
214
                                if( line.matches( "^END[ \t]*:[ \t]*VCARD" ) )
215
215
                                {
216
 
                                        // store vcard and do away with it
 
216
                                        // finalise the vcard/contact
217
217
                                        try {
218
 
                                                vcard.finaliseParsing();
 
218
                                                vcard.finaliseVcard();
 
219
 
 
220
                                                // pass the finalised contact to the importer
219
221
                                                importContact( vcard );
220
222
                                        }
221
 
                                        catch( VCard.ParseException e ) {
222
 
                                                skipContact();
223
 
                                                if( !showContinue(
224
 
                                                        getText( R.string.error_vcf_parse ).toString()
225
 
                                                        + fileName + "\n" + e.getMessage() ) )
226
 
                                                {
227
 
                                                        finish( ACTION_ABORT );
228
 
                                                }
229
 
                                        }
230
 
                                        catch( VCard.SkipContactException e ) {
231
 
                                                skipContact();
232
 
                                                // do nothing
233
 
                                        }
 
223
                                        catch( Vcard.ParseException e ) {
 
224
                                                if( !showContinue(
 
225
                                                        getText( R.string.error_vcf_parse ).toString()
 
226
                                                        + fileName +
 
227
                                                        getText( R.string.error_vcf_parse_line ).toString()
 
228
                                                        + cli.getLineNumber() + ":\n" + e.getMessage() ) )
 
229
                                                {
 
230
                                                        finish( ACTION_ABORT );
 
231
                                                }
 
232
                                                else
 
233
                                                        skipContact();
 
234
                                        }
 
235
                                        catch( ContactData.ContactNotIdentifiableException e ) {
 
236
                                                if( !showContinue(
 
237
                                                        getText( R.string.error_vcf_parse ).toString()
 
238
                                                        + fileName +
 
239
                                                        getText( R.string.error_vcf_parse_line ).toString()
 
240
                                                        + vcard_start_line + ":\n" + getText(
 
241
                                                                R.string.error_vcf_notenoughinfo ).toString()
 
242
                                                ) )
 
243
                                                {
 
244
                                                        finish( ACTION_ABORT );
 
245
                                                }
 
246
                                                else
 
247
                                                        skipContact();
 
248
                                        }
 
249
 
 
250
                                        // discard this vcard
234
251
                                        vcard = null;
235
252
                                }
236
253
                                else
237
254
                                {
238
255
                                        // try giving the line to the vcard
239
256
                                        try {
240
 
                                                vcard.parseLine( buffer, line,
241
 
                                                        cli.doesNextLineLookFolded() );
 
257
                                                vcard.parseLine( content_line );
242
258
                                        }
243
 
                                        catch( VCard.ParseException e ) {
 
259
                                        catch( Vcard.ParseException e ) {
244
260
                                                skipContact();
245
261
                                                if( !showContinue(
246
262
                                                        getText( R.string.error_vcf_parse ).toString()
247
 
                                                        + fileName + "\n" + e.getMessage() ) )
 
263
                                                        + fileName +
 
264
                                                        getText( R.string.error_vcf_parse_line ).toString()
 
265
                                                        + cli.getLineNumber() + "\n" + e.getMessage() ) )
248
266
                                                {
249
267
                                                        finish( ACTION_ABORT );
250
268
                                                }
254
272
                                                // get to another BEGIN:VCARD line.
255
273
                                                vcard = null;
256
274
                                        }
257
 
                                        catch( VCard.SkipContactException e ) {
 
275
                                        catch( Vcard.SkipImportException e ) {
258
276
                                                skipContact();
259
277
                                                // abort this vCard. Further lines will be ignored until
260
278
                                                // we get to another BEGIN:VCARD line.
265
283
                }
266
284
        }
267
285
 
268
 
        class ContentLineIterator implements Iterator< ByteBuffer >
 
286
        class ContentLine
 
287
        {
 
288
                private ByteBuffer _buffer;
 
289
                private boolean _folded_next;
 
290
                private String _line;
 
291
 
 
292
                public ContentLine( ByteBuffer buffer, boolean folded_next )
 
293
                {
 
294
                        _buffer = buffer;
 
295
                        _folded_next = folded_next;
 
296
                        _line = null;
 
297
                }
 
298
 
 
299
                public ByteBuffer getBuffer()
 
300
                {
 
301
                        return _buffer;
 
302
                }
 
303
 
 
304
                public boolean doesNextLineLookFolded()
 
305
                {
 
306
                        return _folded_next;
 
307
                }
 
308
 
 
309
                public String getUsAsciiLine()
 
310
                {
 
311
                        // generated line and cache it
 
312
                        if( _line == null ) {
 
313
                                try {
 
314
                                        _line = new String( _buffer.array(), _buffer.position(),
 
315
                                                _buffer.limit() - _buffer.position(), "US-ASCII" );
 
316
                                }
 
317
                                catch( UnsupportedEncodingException e ) {
 
318
                                        // we know US-ASCII *is* supported, so appease the
 
319
                                        // compiler...
 
320
                                }
 
321
                        }
 
322
 
 
323
                        // return cached line
 
324
                        return _line;
 
325
                }
 
326
        }
 
327
 
 
328
        class ContentLineIterator implements Iterator< ContentLine >
269
329
        {
270
330
                protected byte[] _content = null;
271
331
                protected int _pos = 0;
 
332
                protected int _line = 0;
272
333
 
273
334
                public ContentLineIterator( byte[] content )
274
335
                {
282
343
                }
283
344
 
284
345
                @Override
285
 
                public ByteBuffer next()
 
346
                public ContentLine next()
286
347
                {
287
348
                        int initial_pos = _pos;
288
349
 
294
355
                                        int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
295
356
                                                _pos > initial_pos )? _pos - 1 : _pos;
296
357
                                        _pos++;
297
 
                                        return ByteBuffer.wrap( _content, initial_pos,
298
 
                                                to - initial_pos );
 
358
                                        _line++;
 
359
                                        return new ContentLine(
 
360
                                                ByteBuffer.wrap( _content, initial_pos,
 
361
                                                        to - initial_pos ),
 
362
                                                doesNextLineLookFolded() );
299
363
                                }
300
364
 
301
365
                        // we didn't find one, but were there bytes left?
302
366
                        if( _pos != initial_pos ) {
303
367
                                int to = _pos;
304
368
                                _pos++;
305
 
                                return ByteBuffer.wrap( _content, initial_pos,
306
 
                                        to - initial_pos );
 
369
                                _line++;
 
370
                                return new ContentLine(
 
371
                                        ByteBuffer.wrap( _content, initial_pos,
 
372
                                                to - initial_pos ),
 
373
                                        doesNextLineLookFolded() );
307
374
                        }
308
375
 
309
376
                        // no bytes left
321
388
                 * onto the end of this one?
322
389
                 * @return
323
390
                 */
324
 
                public boolean doesNextLineLookFolded()
 
391
                private boolean doesNextLineLookFolded()
325
392
                {
326
393
                        return _pos > 0 && _pos < _content.length &&
327
 
                                _content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
 
394
                                _content[ _pos - 1 ] == '\n' &&
 
395
                                ( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
 
396
                }
 
397
 
 
398
                public int getLineNumber()
 
399
                {
 
400
                        return _line;
328
401
                }
329
402
        }
330
403
 
331
 
        private class VCard extends ContactData
 
404
        private class Vcard extends ContactData
332
405
        {
333
406
                private final static int NAMELEVEL_NONE = 0;
334
 
                private final static int NAMELEVEL_FN = 1;
335
 
                private final static int NAMELEVEL_N = 2;
 
407
                private final static int NAMELEVEL_N = 1;
 
408
                private final static int NAMELEVEL_FN = 2;
336
409
 
337
410
                private final static int MULTILINE_NONE = 0;
338
411
                private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
339
412
                private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
340
 
                private final static int MULTILINE_FOLDED = 3;  // v3.0 folding
 
413
                private final static int MULTILINE_FOLDED = 3;  // MIME-DIR folding
341
414
 
342
415
                private String _version = null;
343
 
                private Vector< ByteBuffer > _buffers = null;
 
416
                private Vector< ContentLine > _content_lines = null;
344
417
                private int _name_level = NAMELEVEL_NONE;
345
418
                private int _parser_multiline_state = MULTILINE_NONE;
346
419
                private String _parser_current_name_and_params = null;
382
455
 
383
456
                        public ParseException( int res )
384
457
                        {
385
 
                                super( VCFImporter.this.getText( res ).toString() );
 
458
                                super( VcardImporter.this.getText( res ).toString() );
386
459
                        }
387
460
                }
388
461
 
389
462
                @SuppressWarnings("serial")
390
 
                protected class SkipContactException extends Exception { }
 
463
                protected class SkipImportException extends Exception { }
391
464
 
392
 
                private String extractCollonPartFromLine( ByteBuffer buffer,
393
 
                        String line, boolean former )
 
465
                private String extractCollonPartFromLine( ContentLine content_line,
 
466
                        boolean former )
394
467
                {
395
468
                        String ret = null;
396
469
 
397
 
                        // get a US-ASCII version of the line for processing, unless we were
398
 
                        // supplied with one
399
 
                        if( line == null ) {
400
 
                                try {
401
 
                                        line = new String( buffer.array(), buffer.position(),
402
 
                                                buffer.limit() - buffer.position(), "US-ASCII" );
403
 
                                }
404
 
                                catch( UnsupportedEncodingException e ) {
405
 
                                        // we know US-ASCII is supported, so appease the compiler...
406
 
                                        line = "";
407
 
                                }
408
 
                        }
409
 
 
410
470
                        // split line into name and value parts and check to make sure we
411
471
                        // only got 2 parts and that the first part is not zero in length
412
 
                        String[] parts = line.split( ":", 2 );
 
472
                        String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
413
473
                        if( parts.length == 2 && parts[ 0 ].length() > 0 )
414
474
                                ret = parts[ former? 0 : 1 ];
415
475
 
416
476
                        return ret;
417
477
                }
418
478
 
419
 
                private String extractNameAndParamsFromLine( ByteBuffer buffer,
420
 
                        String line )
421
 
                {
422
 
                        return extractCollonPartFromLine( buffer, line, true );
423
 
                }
424
 
 
425
 
                private String extractValueFromLine( ByteBuffer buffer, String line )
426
 
                {
427
 
                        return extractCollonPartFromLine( buffer, line, false );
428
 
                }
429
 
 
430
 
                public void parseLine( ByteBuffer buffer, String line,
431
 
                        boolean next_line_looks_folded )
432
 
                        throws ParseException, SkipContactException,
 
479
                private String extractNameAndParamsFromLine( ContentLine content_line )
 
480
                {
 
481
                        return extractCollonPartFromLine( content_line, true );
 
482
                }
 
483
 
 
484
                private String extractValueFromLine( ContentLine content_line )
 
485
                {
 
486
                        return extractCollonPartFromLine( content_line, false );
 
487
                }
 
488
 
 
489
                public void parseLine( ContentLine content_line )
 
490
                        throws ParseException, SkipImportException,
433
491
                        AbortImportException
434
492
                {
435
493
                        // do we have a version yet?
437
495
                        {
438
496
                                // tentatively get name and params from line
439
497
                                String name_and_params =
440
 
                                        extractNameAndParamsFromLine( buffer, line );
 
498
                                        extractNameAndParamsFromLine( content_line );
441
499
 
442
500
                                // is it a version line?
443
501
                                if( name_and_params != null &&
444
 
                                        name_and_params.equals( "VERSION" ) )
 
502
                                        name_and_params.equalsIgnoreCase( "VERSION" ) )
445
503
                                {
446
504
                                        // yes, get it!
447
 
                                        String value = extractValueFromLine( buffer, line );
 
505
                                        String value = extractValueFromLine( content_line );
448
506
                                        if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
449
507
                                                throw new ParseException( R.string.error_vcf_version );
450
508
                                        _version = value;
451
509
 
452
510
                                        // parse any buffers we've been accumulating while we waited
453
511
                                        // for a version
454
 
                                        if( _buffers != null )
455
 
                                                for( int i = 0; i < _buffers.size(); i++ )
456
 
                                                        parseLine( _buffers.get( i ), null,
457
 
                                                                i + 1 < _buffers.size() &&
458
 
                                                                _buffers.get( i + 1 ).hasRemaining() &&
459
 
                                                                _buffers.get( i + 1 ).get(
460
 
                                                                        _buffers.get( i + 1 ).position() ) == ' ' );
461
 
                                        _buffers = null;
 
512
                                        if( _content_lines != null )
 
513
                                                for( int i = 0; i < _content_lines.size(); i++ )
 
514
                                                        parseLine( _content_lines.get( i ) );
 
515
                                        _content_lines = null;
462
516
                                }
463
517
                                else
464
518
                                {
465
519
                                        // no, so stash this line till we get a version
466
 
                                        if( _buffers == null )
467
 
                                                _buffers = new Vector< ByteBuffer >();
468
 
                                        _buffers.add( buffer );
 
520
                                        if( _content_lines == null )
 
521
                                                _content_lines = new Vector< ContentLine >();
 
522
                                        _content_lines.add( content_line );
469
523
                                }
470
524
                        }
471
525
                        else
472
526
                        {
473
527
                                // name and params and the position in the buffer where the
474
 
                                // "value" part of the line start
 
528
                                // "value" part of the line starts
475
529
                                String name_and_params;
476
530
                                int pos;
477
531
 
483
537
 
484
538
                                        // skip some initial line characters, depending on the type
485
539
                                        // of multi-line we're handling
486
 
                                        pos = buffer.position();
 
540
                                        pos = content_line.getBuffer().position();
487
541
                                        switch( _parser_multiline_state )
488
542
                                        {
489
543
                                        case MULTILINE_FOLDED:
490
544
                                                pos++;
491
545
                                                break;
492
546
                                        case MULTILINE_ENCODED:
493
 
                                                while( pos < buffer.limit() && (
494
 
                                                        buffer.get( pos ) == ' ' ||
495
 
                                                        buffer.get( pos ) == '\t' ) )
 
547
                                                while( pos < content_line.getBuffer().limit() && (
 
548
                                                        content_line.getBuffer().get( pos ) == ' ' ||
 
549
                                                        content_line.getBuffer().get( pos ) == '\t' ) )
496
550
                                                {
497
551
                                                        pos++;
498
552
                                                }
507
561
                                }
508
562
                                else
509
563
                                {
 
564
                                        // skip empty lines
 
565
                                        if( content_line.getUsAsciiLine().trim().length() == 0 )
 
566
                                                return;
 
567
 
510
568
                                        // get name and params from line, and since we're not
511
569
                                        // parsing a subsequent line in a multi-line, this should
512
570
                                        // not fail, or it's an error
513
571
                                        name_and_params =
514
 
                                                extractNameAndParamsFromLine( buffer, line );
 
572
                                                extractNameAndParamsFromLine( content_line );
515
573
                                        if( name_and_params == null )
516
574
                                                throw new ParseException(
517
575
                                                        R.string.error_vcf_malformed );
518
576
 
519
577
                                        // calculate how many chars to skip from beginning of line
520
578
                                        // so we skip the property "name:" part
521
 
                                        pos = buffer.position() + name_and_params.length() + 1;
 
579
                                        pos = content_line.getBuffer().position() +
 
580
                                                name_and_params.length() + 1;
522
581
 
523
582
                                        // reset the saved multi-line state
524
583
                                        _parser_current_name_and_params = name_and_params;
527
586
 
528
587
                                // get value from buffer, as raw bytes
529
588
                                ByteBuffer value;
530
 
                                value = ByteBuffer.wrap( buffer.array(), pos,
531
 
                                        buffer.limit() - pos );
 
589
                                value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
 
590
                                        content_line.getBuffer().limit() - pos );
532
591
 
533
592
                                // get parameter parts
534
593
                                String[] name_param_parts = name_and_params.split( ";", -1 );
535
594
                                for( int i = 0; i < name_param_parts.length; i++ )
536
595
                                        name_param_parts[ i ] = name_param_parts[ i ].trim();
537
596
 
 
597
                                // determine whether we care about this entry
 
598
                                final HashSet< String > interesting_fields =
 
599
                                        new HashSet< String >( Arrays.asList( new String[] { "N",
 
600
                                                "FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
 
601
                                ) );
 
602
                                boolean is_interesting_field =
 
603
                                        interesting_fields.contains(
 
604
                                                name_param_parts[ 0 ].toUpperCase( Locale.US ) );
 
605
 
538
606
                                // parse encoding parameter
539
607
                                String encoding = checkParam( name_param_parts, "ENCODING" );
540
 
                                if( encoding != null ) encoding = encoding.toUpperCase();
541
 
                                if( encoding != null && !encoding.equals( "8BIT" ) &&
542
 
                                        !encoding.equals( "QUOTED-PRINTABLE" ) )
543
 
                                        //&& !encoding.equals( "BASE64" ) )
 
608
                                if( encoding != null )
 
609
                                        encoding = encoding.toUpperCase( Locale.US );
 
610
                                if( is_interesting_field && encoding != null &&
 
611
                                        !encoding.equalsIgnoreCase( "8BIT" ) &&
 
612
                                        !encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
 
613
                                        //&& !encoding.equalsIgnoreCase( "BASE64" ) )
544
614
                                {
545
615
                                        throw new ParseException( R.string.error_vcf_encoding );
546
616
                                }
547
617
 
548
618
                                // parse charset parameter
549
619
                                String charset = checkParam( name_param_parts, "CHARSET" );
550
 
                                if( charset != null ) charset = charset.toUpperCase();
551
 
                                if( charset != null && !charset.equals( "US-ASCII" ) &&
552
 
                                        !charset.equals( "ASCII" ) &&
553
 
                                        !charset.equals( "UTF-8" ) )
 
620
                                if( charset != null )
 
621
                                        charset = charset.toUpperCase( Locale.US );
 
622
                                if( charset != null &&
 
623
                                        !charset.equalsIgnoreCase( "US-ASCII" ) &&
 
624
                                        !charset.equalsIgnoreCase( "ASCII" ) &&
 
625
                                        !charset.equalsIgnoreCase( "UTF-8" ) )
554
626
                                {
555
627
                                        throw new ParseException( R.string.error_vcf_charset );
556
628
                                }
558
630
                                // do unencoding (or default to a fake unencoding result with
559
631
                                // the raw string)
560
632
                                UnencodeResult unencoding_result = null;
561
 
                                if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
 
633
                                if( encoding != null &&
 
634
                                        encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
 
635
                                {
562
636
                                        unencoding_result = unencodeQuotedPrintable( value );
563
 
//                              else if( encoding != null && encoding.equals( "BASE64" ) )
 
637
                                }
 
638
//                              else if( encoding != null &&
 
639
//                                      encoding.equalsIgnoreCase( "BASE64" ) )
 
640
//                              {
564
641
//                                      unencoding_result = unencodeBase64( props[ 1 ], charset );
 
642
//                              }
565
643
                                if( unencoding_result != null ) {
566
644
                                        value = unencoding_result.getBuffer();
567
645
                                        if( unencoding_result.isAnotherLineRequired() )
568
646
                                                _parser_multiline_state = MULTILINE_ENCODED;
569
647
                                }
570
648
 
571
 
                                // convert 8-bit ASCII charset to US-ASCII
572
 
                                if( charset == null || charset.equals( "ASCII" ) ) {
 
649
                                // convert 8-bit US-ASCII charset to UTF-8 (where no charset is
 
650
                                // specified for a v2.1 vcard entry, we assume it's US-ASCII)
 
651
                                if( ( charset == null && _version.equals( "2.1" ) ) ||
 
652
                                        ( charset != null && (
 
653
                                                charset.equalsIgnoreCase( "ASCII" ) ||
 
654
                                                charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
 
655
                                {
573
656
                                        value = transcodeAsciiToUtf8( value );
574
 
                                        charset = "UTF-8";
575
657
                                }
576
658
 
577
 
                                // process charset
 
659
                                // process charset (value is now in UTF-8)
578
660
                                String string_value;
579
661
                                try {
580
662
                                        string_value = new String( value.array(), value.position(),
581
 
                                                value.limit() - value.position(), charset );
 
663
                                                value.limit() - value.position(), "UTF-8" );
582
664
                                } catch( UnsupportedEncodingException e ) {
583
665
                                        throw new ParseException( R.string.error_vcf_charset );
584
666
                                }
586
668
                                // for some entries that have semicolon-separated value parts,
587
669
                                // check to see if the value ends in an escape character, which
588
670
                                // indicates that we have a multi-line value
589
 
                                if( ( name_param_parts[ 0 ].equals( "N" ) ||
590
 
                                        name_param_parts[ 0 ].equals( "ORG" ) ||
591
 
                                        name_param_parts[ 0 ].equals( "ADR" ) ) &&
 
671
                                if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
 
672
                                        name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
 
673
                                        name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
592
674
                                        doesStringEndInAnEscapeChar( string_value ) )
593
675
                                {
594
676
                                        _parser_multiline_state = MULTILINE_ESCAPED;
596
678
                                                string_value.length() - 1 );
597
679
                                }
598
680
 
599
 
                                // now we know whether we're in an encoding multi-line,
600
 
                                // determine if we're in a v3 folded multi-line or not
 
681
                                // if we know we're not in an encoding-based multi-line, check
 
682
                                // to see if we're in a folded multi-line
601
683
                                if( _parser_multiline_state == MULTILINE_NONE &&
602
 
                                        _version.equals( "3.0" ) && next_line_looks_folded )
 
684
                                        content_line.doesNextLineLookFolded() )
603
685
                                {
604
686
                                        _parser_multiline_state = MULTILINE_FOLDED;
605
687
                                }
617
699
                                if( complete_value.length() < 1 ) return;
618
700
 
619
701
                                // parse some properties
620
 
                                if( name_param_parts[ 0 ].equals( "N" ) )
 
702
                                if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
621
703
                                        parseN( name_param_parts, complete_value );
622
 
                                else if( name_param_parts[ 0 ].equals( "FN" ) )
 
704
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
623
705
                                        parseFN( name_param_parts, complete_value );
624
 
                                else if( name_param_parts[ 0 ].equals( "ORG" ) )
 
706
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
625
707
                                        parseORG( name_param_parts, complete_value );
626
 
                                else if( name_param_parts[ 0 ].equals( "TITLE" ) )
 
708
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
627
709
                                        parseTITLE( name_param_parts, complete_value );
628
 
                                else if( name_param_parts[ 0 ].equals( "TEL" ) )
 
710
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
629
711
                                        parseTEL( name_param_parts, complete_value );
630
 
                                else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
 
712
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
631
713
                                        parseEMAIL( name_param_parts, complete_value );
632
 
                                else if( name_param_parts[ 0 ].equals( "ADR" ) )
 
714
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
633
715
                                        parseADR( name_param_parts, complete_value );
 
716
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
 
717
                                        parseLABEL( name_param_parts, complete_value );
 
718
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
 
719
                                        parseNOTE( name_param_parts, complete_value );
634
720
                        }
635
721
                }
636
722
 
649
735
                        return ( count & 1 ) == 1;
650
736
                }
651
737
 
652
 
                private String[] splitValueBySemicolon( String value )
 
738
                private String[] splitValueByCharacter( String value, char character )
653
739
                {
654
 
                        // split string in to parts by semicolon
 
740
                        // split string in to parts by specified character
655
741
                        ArrayList< String > parts = new ArrayList< String >(
656
 
                                Arrays.asList( value.split(  ";" ) ) );
 
742
                                Arrays.asList( value.split( "" + character ) ) );
657
743
 
658
744
                        // go through parts
659
745
                        for( int a = 0; a < parts.size(); a++ )
667
753
                                if( a < parts.size() - 1 &&
668
754
                                        doesStringEndInAnEscapeChar( str ) )
669
755
                                {
670
 
                                        // join the next part to this part and remove the next part
 
756
                                        // append the escaped character, join the next part to this
 
757
                                        // part and remove the next part
671
758
                                        parts.set( a, str.substring( 0, str.length() - 1 ) +
672
 
                                                ';' + parts.get( a + 1 ) );
 
759
                                                character + parts.get( a + 1 ) );
673
760
                                        parts.remove( a + 1 );
674
761
 
675
762
                                        // re-visit this part
686
773
                        return parts.toArray( ret );
687
774
                }
688
775
 
 
776
                private String unescapeValue( String value )
 
777
                {
 
778
                        StringBuilder ret = new StringBuilder( value.length() );
 
779
                        boolean in_escape = false;
 
780
                        for( int a = 0; a < value.length(); a++ )
 
781
                        {
 
782
                                int c = value.codePointAt( a );
 
783
 
 
784
                                // process a normal character
 
785
                                if( !in_escape ) {
 
786
                                        if( c == '\\' )
 
787
                                                in_escape = true;
 
788
                                        else
 
789
                                                ret.append( Character.toChars( c ) );
 
790
                                        continue;
 
791
                                }
 
792
 
 
793
                                // process an escape sequence
 
794
                                in_escape = false;
 
795
                                switch( c )
 
796
                                {
 
797
                                case 'T':
 
798
                                case 't':
 
799
                                        // add tab (invalid/non-standard, but accepted)
 
800
                                        ret.append( '\t' );
 
801
                                        break;
 
802
                                case 'N':
 
803
                                case 'n':
 
804
                                        // add newline
 
805
                                        ret.append( '\n' );
 
806
                                        break;
 
807
                                case '\\':
 
808
                                case ',':
 
809
                                case ';':
 
810
                                        // add escaped character
 
811
                                        ret.append( Character.toChars( c ) );
 
812
                                        break;
 
813
                                default:
 
814
                                        // unknown escape sequence, so add it unescaped
 
815
                                        // (invalid/non-standard, but accepted)
 
816
                                        ret.append( "\\" );
 
817
                                        ret.append( Character.toChars( c ) );
 
818
                                        break;
 
819
                                }
 
820
                        }
 
821
 
 
822
                        return ret.toString();
 
823
                }
 
824
 
689
825
                private void parseN( String[] params, String value )
690
826
                {
691
827
                        // already got a better name?
692
828
                        if( _name_level >= NAMELEVEL_N ) return;
693
829
 
694
830
                        // get name parts
695
 
                        String[] name_parts = splitValueBySemicolon( value );
 
831
                        String[] name_parts = splitValueByCharacter( value, ';' );
696
832
 
697
833
                        // build name
698
834
                        value = "";
699
 
                        if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
700
 
                                value += name_parts[ 1 ];
701
 
                        if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
702
 
                                value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
 
835
                        final int[] part_order = { 3, 1, 2, 0, 4 };
 
836
                        for( int a = 0; a < part_order.length; a++ )
 
837
                                if( name_parts.length > part_order[ a ] &&
 
838
                                        name_parts[ part_order[ a ] ].length() > 0 )
 
839
                                {
 
840
                                        // split this part in to it's comma-separated bits
 
841
                                        String[] name_part_parts = splitValueByCharacter(
 
842
                                                name_parts[ part_order[ a ] ], ',' );
 
843
                                        for( int b = 0; b < name_part_parts.length; b++ )
 
844
                                                if( name_part_parts[ b ].length() > 0 )
 
845
                                                {
 
846
                                                        if( value.length() > 0 ) value += " ";
 
847
                                                        value += name_part_parts[ b ];
 
848
                                                }
 
849
                                }
703
850
 
704
851
                        // set name
705
 
                        setName( value );
 
852
                        setName( unescapeValue( value ) );
706
853
                        _name_level = NAMELEVEL_N;
707
854
                }
708
855
 
712
859
                        if( _name_level >= NAMELEVEL_FN ) return;
713
860
 
714
861
                        // set name
715
 
                        setName( value );
 
862
                        setName( unescapeValue( value ) );
716
863
                        _name_level = NAMELEVEL_FN;
717
864
                }
718
865
 
719
866
                private void parseORG( String[] params, String value )
720
867
                {
721
868
                        // get org parts
722
 
                        String[] org_parts = splitValueBySemicolon( value );
 
869
                        String[] org_parts = splitValueByCharacter( value, ';' );
723
870
                        if( org_parts == null || org_parts.length < 1 ) return;
724
871
 
725
872
                        // build organisation name
727
874
                                String.valueOf( org_parts[ 0 ] ) );
728
875
                        for( int a = 1; a < org_parts.length; a++ )
729
876
                                builder.append( ", " ).append( org_parts[ a ] );
730
 
                        String organisation = builder.toString();
 
877
                        String organisation = unescapeValue( builder.toString() );
731
878
 
732
879
                        // set organisation name (using a title we've previously found)
733
880
                        addOrganisation( organisation, _cached_title, true );
744
891
 
745
892
                private void parseTITLE( String[] params, String value )
746
893
                {
 
894
                        value = unescapeValue( value );
 
895
 
747
896
                        // if we previously had an organisation, look it up and append this
748
897
                        // title to it
749
898
                        if( _cached_organisation != null && hasOrganisations() ) {
775
924
                        int type;
776
925
                        if( types.contains( "FAX" ) )
777
926
                                if( types.contains( "HOME" ) )
778
 
                                        type = PhonesColumns.TYPE_FAX_HOME;
 
927
                                        type = TYPE_FAX_HOME;
779
928
                                else
780
 
                                        type = PhonesColumns.TYPE_FAX_WORK;
 
929
                                        type = TYPE_FAX_WORK;
781
930
                        else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
782
 
                                type = PhonesColumns.TYPE_MOBILE;
 
931
                                type = TYPE_MOBILE;
783
932
                        else if( types.contains( "PAGER" ) )
784
 
                                type = PhonesColumns.TYPE_PAGER;
 
933
                                type = TYPE_PAGER;
785
934
                        else if( types.contains( "WORK" ) )
786
 
                                type = PhonesColumns.TYPE_WORK;
 
935
                                type = TYPE_WORK;
787
936
                        else
788
 
                                type = PhonesColumns.TYPE_HOME;
 
937
                                type = TYPE_HOME;
789
938
 
790
939
                        // add phone number
791
940
                        addNumber( value, type, is_preferred );
802
951
                        boolean is_preferred = types.contains( "PREF" );
803
952
                        int type;
804
953
                        if( types.contains( "WORK" ) )
805
 
                                type = Contacts.ContactMethods.TYPE_WORK;
 
954
                                type = TYPE_WORK;
806
955
                        else
807
 
                                type = Contacts.ContactMethods.TYPE_HOME;
 
956
                                type = TYPE_HOME;
808
957
 
809
 
                        addEmail( value, type, is_preferred );
 
958
                        addEmail( unescapeValue( value ), type, is_preferred );
810
959
                }
811
960
 
812
961
                private void parseADR( String[] params, String value )
813
962
                {
814
963
                        // get address parts
815
 
                        String[] adr_parts = splitValueBySemicolon( value );
 
964
                        String[] adr_parts = splitValueByCharacter( value, ';' );
816
965
 
817
966
                        // build address
818
967
                        value = "";
819
 
                        for( int a = 0; a < adr_parts.length; a++ ) {
820
 
                                if( value.length() > 0 ) value += "\n";
821
 
                                value += adr_parts[ a ].trim();
822
 
                        }
823
 
 
824
 
                        Set< String > types = extractTypes( params, Arrays.asList(
825
 
                                "PREF", "WORK", "HOME", "INTERNET" ) );
826
 
 
827
 
                        // add address
828
 
                        int type;
829
 
                        if( types.contains( "WORK" ) )
830
 
                                type = Contacts.ContactMethods.TYPE_WORK;
831
 
                        else
832
 
                                type = Contacts.ContactMethods.TYPE_HOME;
833
 
 
834
 
                        addAddress( value, type );
835
 
                }
836
 
 
837
 
                public void finaliseParsing()
838
 
                        throws ParseException, SkipContactException,
839
 
                        AbortImportException
 
968
                        for( int a = 0; a < adr_parts.length; a++ )
 
969
                                if( adr_parts[ a ].length() > 0 )
 
970
                                {
 
971
                                        // version 3.0 vCards allow further splitting by comma
 
972
                                        if( _version.equals( "3.0" ) )
 
973
                                        {
 
974
                                                // split this part in to it's comma-separated bits and
 
975
                                                // add them on individual lines
 
976
                                                String[] adr_part_parts =
 
977
                                                        splitValueByCharacter( adr_parts[ a ], ',' );
 
978
                                                for( int b = 0; b < adr_part_parts.length; b++ )
 
979
                                                        if( adr_part_parts[ b ].length() > 0 )
 
980
                                                        {
 
981
                                                                if( value.length() > 0 ) value += "\n";
 
982
                                                                value += adr_part_parts[ b ];
 
983
                                                        }
 
984
                                        }
 
985
                                        else
 
986
                                        {
 
987
                                                // add this part on an individual line
 
988
                                                if( value.length() > 0 ) value += "\n";
 
989
                                                value += adr_parts[ a ];
 
990
                                        }
 
991
                                }
 
992
 
 
993
                        Set< String > types = extractTypes( params, Arrays.asList(
 
994
                                "PREF", "WORK", "HOME" ) );
 
995
 
 
996
                        // add address
 
997
                        int type;
 
998
                        if( types.contains( "WORK" ) )
 
999
                                type = TYPE_WORK;
 
1000
                        else
 
1001
                                type = TYPE_HOME;
 
1002
 
 
1003
                        addAddress( unescapeValue( value ), type );
 
1004
                }
 
1005
 
 
1006
                private void parseLABEL( String[] params, String value )
 
1007
                {
 
1008
                        Set< String > types = extractTypes( params, Arrays.asList(
 
1009
                                "PREF", "WORK", "HOME" ) );
 
1010
 
 
1011
                        // add address
 
1012
                        int type;
 
1013
                        if( types.contains( "WORK" ) )
 
1014
                                type = TYPE_WORK;
 
1015
                        else
 
1016
                                type = TYPE_HOME;
 
1017
 
 
1018
                        addAddress( unescapeValue( value ), type );
 
1019
                }
 
1020
 
 
1021
                private void parseNOTE( String[] params, String value )
 
1022
                {
 
1023
                        addNote( unescapeValue( value ) );
 
1024
                }
 
1025
 
 
1026
                public void finaliseVcard()
 
1027
                        throws ParseException, ContactNotIdentifiableException
840
1028
                {
841
1029
                        // missing version (and data is present)
842
 
                        if( _version == null && _buffers != null )
 
1030
                        if( _version == null && _content_lines != null )
843
1031
                                throw new ParseException( R.string.error_vcf_malformed );
844
1032
 
845
 
                        // check if we should import this contact
846
 
                        try {
847
 
                                if( !isImportRequired( this ) )
848
 
                                        throw new SkipContactException();
849
 
                        }
850
 
                        catch( ContactNeedsMoreInfoException e ) {
851
 
                                throw new ParseException( R.string.error_vcf_notenoughinfo );
852
 
                        }
 
1033
                        // finalise the parent class
 
1034
                        finalise();
853
1035
                }
854
1036
 
 
1037
                /**
 
1038
                 * Amongst the params, find the value of the first, only, of any with
 
1039
                 * the specified name
 
1040
                 * @param params
 
1041
                 * @param name
 
1042
                 * @return a value, or null
 
1043
                 */
855
1044
                private String checkParam( String[] params, String name )
856
1045
                {
 
1046
                        String[] res = checkParams( params, name );
 
1047
                        return res.length > 0? res[ 0 ] : null;
 
1048
                }
 
1049
 
 
1050
                /**
 
1051
                 * Amongst the params, find the values of any with the specified name
 
1052
                 * @param params
 
1053
                 * @param name
 
1054
                 * @return an array of values, or null
 
1055
                 */
 
1056
                private String[] checkParams( String[] params, String name )
 
1057
                {
 
1058
                        HashSet< String > ret = new HashSet< String >();
 
1059
 
857
1060
                        Pattern p = Pattern.compile(
858
 
                                "^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
 
1061
                                "^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
 
1062
                                Pattern.CASE_INSENSITIVE );
859
1063
                        for( int i = 0; i < params.length; i++ ) {
860
1064
                                Matcher m = p.matcher( params[ i ] );
861
1065
                                if( m.matches() )
862
 
                                        return m.group( 2 );
 
1066
                                        ret.add( m.group( 2 ) );
863
1067
                        }
864
 
                        return null;
 
1068
 
 
1069
                        return (String[]) ret.toArray( new String[ ret.size() ] );
865
1070
                }
866
1071
 
 
1072
                /**
 
1073
                 * Amongst the params, return any type values present. For v2.1 vCards,
 
1074
                 * those types are just parameters. For v3.0, they are prefixed with
 
1075
                 * "TYPE=". There may also be multiple type parameters.
 
1076
                 * @param params an array of params to look for types in
 
1077
                 * @param valid_types an list of upper-case type values to look for
 
1078
                 * @return a set of present type values
 
1079
                 */
867
1080
                private Set< String > extractTypes( String[] params,
868
1081
                        List< String > valid_types )
869
1082
                {
870
1083
                        HashSet< String > types = new HashSet< String >();
871
1084
 
872
1085
                        // get 3.0-style TYPE= param
873
 
                        String type_param;
874
 
                        if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
875
 
                                String[] parts = type_param.split( "," );
876
 
                                for( int i = 0; i < parts.length; i++ )
877
 
                                        if( valid_types.contains( parts[ i ] ) )
878
 
                                                types.add( parts[ i ] );
 
1086
                        String type_params[] = checkParams( params, "TYPE" );
 
1087
                        for( int a = 0; a < type_params.length; a++ )
 
1088
                        {
 
1089
                                // check for a comma-separated list of types (why? I don't think
 
1090
                                // this is in the specs!)
 
1091
                                String[] parts = type_params[ a ].split( "," );
 
1092
                                for( int i = 0; i < parts.length; i++ ) {
 
1093
                                        String ucpart = parts[ i ].toUpperCase( Locale.US );
 
1094
                                        if( valid_types.contains( ucpart ) )
 
1095
                                                types.add( ucpart );
 
1096
                                }
879
1097
                        }
880
1098
 
881
1099
                        // get 2.1-style type param
882
1100
                        if( _version.equals( "2.1" ) ) {
883
 
                                for( int i = 1; i < params.length; i++ )
884
 
                                        if( valid_types.contains( params[ i ] ) )
885
 
                                                types.add( params[ i ] );
 
1101
                                for( int i = 1; i < params.length; i++ ) {
 
1102
                                        String ucparam = params[ i ].toUpperCase( Locale.US );
 
1103
                                        if( valid_types.contains( ucparam ) )
 
1104
                                                types.add( ucparam );
 
1105
                                }
886
1106
                        }
887
1107
 
888
1108
                        return types;