/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: Tim Marston
  • Date: 2013-10-20 17:51:32 UTC
  • Revision ID: tim@ed.am-20131020175132-lvqrbal1ztz5jepl
updated .bzrignore

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 <tim@ed.am>
 
4
 * Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
5
5
 *
6
6
 * This file is part of the Import Contacts program (hereafter referred
7
 
 * to as "this program"). For more information, see
 
7
 * to as "this program").  For more information, see
8
8
 * http://ed.am/dev/android/import-contacts
9
9
 *
10
10
 * This program is free software: you can redistribute it and/or modify
133
133
                        boolean in_vcard = false;
134
134
                        while( ( line = reader.readLine() ) != null )
135
135
                        {
136
 
                                if( !in_vcard ) {
 
136
                                if( !in_vcard )
 
137
                                {
137
138
                                        // look for vcard beginning
138
 
                                        if( line.matches( "^BEGIN:VCARD" ) ) {
 
139
                                        if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
139
140
                                                in_vcard = true;
140
141
                                                _vcard_count++;
141
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() );
 
147
                                        }
142
148
                                }
143
 
                                else if( line.matches( "^END:VCARD" ) )
 
149
                                else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
144
150
                                        in_vcard = false;
145
151
                        }
146
152
 
175
181
                        // import
176
182
                        importVCardFileContent( content, file.getName() );
177
183
                }
 
184
                catch( OutOfMemoryError e ) {
 
185
                        showError( R.string.error_outofmemory );
 
186
                }
178
187
                catch( FileNotFoundException e ) {
179
188
                        showError( getText( R.string.error_filenotfound ) +
180
189
                                file.getName() );
193
202
                ContentLineIterator cli = new ContentLineIterator( content );
194
203
                while( cli.hasNext() )
195
204
                {
196
 
                        ByteBuffer buffer = cli.next();
 
205
                        ContentLine content_line = cli.next();
197
206
 
198
 
                        // get a US-ASCII version of the line for processing
199
 
                        String line;
200
 
                        try {
201
 
                                line = new String( buffer.array(), buffer.position(),
202
 
                                        buffer.limit() - buffer.position(), "US-ASCII" );
203
 
                        }
204
 
                        catch( UnsupportedEncodingException e ) {
205
 
                                // we know US-ASCII *is* supported, so appease the compiler...
206
 
                                line = "";
207
 
                        }
 
207
                        // get a US-ASCII version of the string, for processing
 
208
                        String line = content_line.getUsAsciiLine();
208
209
 
209
210
                        if( vcard == null ) {
210
211
                                // look for vcard beginning
211
 
                                if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD" ) ) {
 
212
                                if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
212
213
                                        setProgress( _progress++ );
213
214
                                        vcard = new Vcard();
214
215
                                        vcard_start_line = cli.getLineNumber();
216
217
                        }
217
218
                        else {
218
219
                                // look for vcard content or ending
219
 
                                if( line.matches( "^END[ \t]*:[ \t]*VCARD" ) )
 
220
                                if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
220
221
                                {
221
222
                                        // finalise the vcard/contact
222
223
                                        try {
259
260
                                {
260
261
                                        // try giving the line to the vcard
261
262
                                        try {
262
 
                                                vcard.parseLine( buffer, line,
263
 
                                                        cli.doesNextLineLookFolded() );
 
263
                                                vcard.parseLine( content_line );
264
264
                                        }
265
265
                                        catch( Vcard.ParseException e ) {
266
266
                                                skipContact();
273
273
                                                        finish( ACTION_ABORT );
274
274
                                                }
275
275
 
276
 
                                                // although we're continuing, we still need to abort
277
 
                                                // this vCard. Further lines will be ignored until we
 
276
                                                // Although we're continuing, we still need to abort
 
277
                                                // this vCard.  Further lines will be ignored until we
278
278
                                                // get to another BEGIN:VCARD line.
279
279
                                                vcard = null;
280
280
                                        }
281
281
                                        catch( Vcard.SkipImportException e ) {
282
282
                                                skipContact();
283
 
                                                // abort this vCard. Further lines will be ignored until
 
283
                                                // Abort this vCard.  Further lines will be ignored until
284
284
                                                // we get to another BEGIN:VCARD line.
285
285
                                                vcard = null;
286
286
                                        }
289
289
                }
290
290
        }
291
291
 
292
 
        class ContentLineIterator implements Iterator< ByteBuffer >
 
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 >
293
335
        {
294
336
                protected byte[] _content = null;
295
337
                protected int _pos = 0;
307
349
                }
308
350
 
309
351
                @Override
310
 
                public ByteBuffer next()
 
352
                public ContentLine next()
311
353
                {
312
354
                        int initial_pos = _pos;
313
355
 
320
362
                                                _pos > initial_pos )? _pos - 1 : _pos;
321
363
                                        _pos++;
322
364
                                        _line++;
323
 
                                        return ByteBuffer.wrap( _content, initial_pos,
324
 
                                                to - initial_pos );
 
365
                                        return new ContentLine(
 
366
                                                ByteBuffer.wrap( _content, initial_pos,
 
367
                                                        to - initial_pos ),
 
368
                                                doesNextLineLookFolded() );
325
369
                                }
326
370
 
327
371
                        // we didn't find one, but were there bytes left?
329
373
                                int to = _pos;
330
374
                                _pos++;
331
375
                                _line++;
332
 
                                return ByteBuffer.wrap( _content, initial_pos,
333
 
                                        to - initial_pos );
 
376
                                return new ContentLine(
 
377
                                        ByteBuffer.wrap( _content, initial_pos,
 
378
                                                to - initial_pos ),
 
379
                                        doesNextLineLookFolded() );
334
380
                        }
335
381
 
336
382
                        // no bytes left
348
394
                 * onto the end of this one?
349
395
                 * @return
350
396
                 */
351
 
                public boolean doesNextLineLookFolded()
 
397
                private boolean doesNextLineLookFolded()
352
398
                {
353
399
                        return _pos > 0 && _pos < _content.length &&
354
400
                                _content[ _pos - 1 ] == '\n' &&
373
419
                private final static int MULTILINE_FOLDED = 3;  // MIME-DIR folding
374
420
 
375
421
                private String _version = null;
376
 
                private Vector< ByteBuffer > _buffers = null;
 
422
                private Vector< ContentLine > _content_lines = null;
377
423
                private int _name_level = NAMELEVEL_NONE;
378
424
                private int _parser_multiline_state = MULTILINE_NONE;
379
425
                private String _parser_current_name_and_params = null;
422
468
                @SuppressWarnings("serial")
423
469
                protected class SkipImportException extends Exception { }
424
470
 
425
 
                private String extractCollonPartFromLine( ByteBuffer buffer,
426
 
                        String line, boolean former )
 
471
                private String extractCollonPartFromLine( ContentLine content_line,
 
472
                        boolean former )
427
473
                {
428
 
                        String ret = null;
429
 
 
430
 
                        // get a US-ASCII version of the line for processing, unless we were
431
 
                        // supplied with one
432
 
                        if( line == null ) {
433
 
                                try {
434
 
                                        line = new String( buffer.array(), buffer.position(),
435
 
                                                buffer.limit() - buffer.position(), "US-ASCII" );
436
 
                                }
437
 
                                catch( UnsupportedEncodingException e ) {
438
 
                                        // we know US-ASCII is supported, so appease the compiler...
439
 
                                        line = "";
440
 
                                }
441
 
                        }
442
 
 
443
474
                        // split line into name and value parts and check to make sure we
444
475
                        // only got 2 parts and that the first part is not zero in length
445
 
                        String[] parts = line.split( ":", 2 );
 
476
                        String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
446
477
                        if( parts.length == 2 && parts[ 0 ].length() > 0 )
447
 
                                ret = parts[ former? 0 : 1 ];
448
 
 
449
 
                        return ret;
450
 
                }
451
 
 
452
 
                private String extractNameAndParamsFromLine( ByteBuffer buffer,
453
 
                        String line )
454
 
                {
455
 
                        return extractCollonPartFromLine( buffer, line, true );
456
 
                }
457
 
 
458
 
                private String extractValueFromLine( ByteBuffer buffer, String line )
459
 
                {
460
 
                        return extractCollonPartFromLine( buffer, line, false );
461
 
                }
462
 
 
463
 
                public void parseLine( ByteBuffer buffer, String line,
464
 
                        boolean next_line_looks_folded )
 
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 )
465
494
                        throws ParseException, SkipImportException,
466
495
                        AbortImportException
467
496
                {
470
499
                        {
471
500
                                // tentatively get name and params from line
472
501
                                String name_and_params =
473
 
                                        extractNameAndParamsFromLine( buffer, line );
 
502
                                        extractNameAndParamsFromLine( content_line );
474
503
 
475
504
                                // is it a version line?
476
505
                                if( name_and_params != null &&
477
506
                                        name_and_params.equalsIgnoreCase( "VERSION" ) )
478
507
                                {
479
508
                                        // yes, get it!
480
 
                                        String value = extractValueFromLine( buffer, line );
481
 
                                        if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
 
509
                                        String value = extractValueFromLine( content_line );
 
510
                                        if( value == null || (
 
511
                                                !value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
 
512
                                        {
482
513
                                                throw new ParseException( R.string.error_vcf_version );
 
514
                                        }
483
515
                                        _version = value;
484
516
 
485
517
                                        // parse any buffers we've been accumulating while we waited
486
518
                                        // for a version
487
 
                                        if( _buffers != null )
488
 
                                                for( int i = 0; i < _buffers.size(); i++ )
489
 
                                                        parseLine( _buffers.get( i ), null,
490
 
                                                                i + 1 < _buffers.size() &&
491
 
                                                                _buffers.get( i + 1 ).hasRemaining() &&
492
 
                                                                _buffers.get( i + 1 ).get(
493
 
                                                                        _buffers.get( i + 1 ).position() ) == ' ' );
494
 
                                        _buffers = null;
 
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;
495
523
                                }
496
524
                                else
497
525
                                {
498
526
                                        // no, so stash this line till we get a version
499
 
                                        if( _buffers == null )
500
 
                                                _buffers = new Vector< ByteBuffer >();
501
 
                                        _buffers.add( buffer );
 
527
                                        if( _content_lines == null )
 
528
                                                _content_lines = new Vector< ContentLine >();
 
529
                                        _content_lines.add( content_line );
502
530
                                }
503
531
                        }
504
532
                        else
516
544
 
517
545
                                        // skip some initial line characters, depending on the type
518
546
                                        // of multi-line we're handling
519
 
                                        pos = buffer.position();
 
547
                                        pos = content_line.getBuffer().position();
520
548
                                        switch( _parser_multiline_state )
521
549
                                        {
522
550
                                        case MULTILINE_FOLDED:
523
551
                                                pos++;
524
552
                                                break;
525
553
                                        case MULTILINE_ENCODED:
526
 
                                                while( pos < buffer.limit() && (
527
 
                                                        buffer.get( pos ) == ' ' ||
528
 
                                                        buffer.get( pos ) == '\t' ) )
 
554
                                                while( pos < content_line.getBuffer().limit() && (
 
555
                                                        content_line.getBuffer().get( pos ) == ' ' ||
 
556
                                                        content_line.getBuffer().get( pos ) == '\t' ) )
529
557
                                                {
530
558
                                                        pos++;
531
559
                                                }
541
569
                                else
542
570
                                {
543
571
                                        // skip empty lines
544
 
                                        if( line.trim().length() == 0 ) return;
 
572
                                        if( content_line.getUsAsciiLine().trim().length() == 0 )
 
573
                                                return;
545
574
 
546
575
                                        // get name and params from line, and since we're not
547
576
                                        // parsing a subsequent line in a multi-line, this should
548
577
                                        // not fail, or it's an error
549
578
                                        name_and_params =
550
 
                                                extractNameAndParamsFromLine( buffer, line );
 
579
                                                extractNameAndParamsFromLine( content_line );
551
580
                                        if( name_and_params == null )
552
581
                                                throw new ParseException(
553
582
                                                        R.string.error_vcf_malformed );
554
583
 
555
584
                                        // calculate how many chars to skip from beginning of line
556
585
                                        // so we skip the property "name:" part
557
 
                                        pos = buffer.position() + name_and_params.length() + 1;
 
586
                                        pos = content_line.getBuffer().position() +
 
587
                                                name_and_params.length() + 1;
558
588
 
559
589
                                        // reset the saved multi-line state
560
590
                                        _parser_current_name_and_params = name_and_params;
563
593
 
564
594
                                // get value from buffer, as raw bytes
565
595
                                ByteBuffer value;
566
 
                                value = ByteBuffer.wrap( buffer.array(), pos,
567
 
                                        buffer.limit() - pos );
 
596
                                value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
 
597
                                        content_line.getBuffer().limit() - pos );
568
598
 
569
599
                                // get parameter parts
570
600
                                String[] name_param_parts = name_and_params.split( ";", -1 );
658
688
                                // if we know we're not in an encoding-based multi-line, check
659
689
                                // to see if we're in a folded multi-line
660
690
                                if( _parser_multiline_state == MULTILINE_NONE &&
661
 
                                        next_line_looks_folded )
 
691
                                        content_line.doesNextLineLookFolded() )
662
692
                                {
663
693
                                        _parser_multiline_state = MULTILINE_FOLDED;
664
694
                                }
694
724
                                        parseLABEL( name_param_parts, complete_value );
695
725
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
696
726
                                        parseNOTE( name_param_parts, complete_value );
 
727
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
 
728
                                        parseBDAY( name_param_parts, complete_value );
697
729
                        }
698
730
                }
699
731
 
723
755
                        {
724
756
                                String str = parts.get( a );
725
757
 
726
 
                                // look for parts that end in an escape character, but ignore
727
 
                                // the final part. We've already detected escape chars at the
 
758
                                // Look for parts that end in an escape character, but ignore
 
759
                                // the final part.  We've already detected escape chars at the
728
760
                                // end of the final part in parseLine() and handled multi-lines
729
761
                                // accordingly.
730
762
                                if( a < parts.size() - 1 &&
820
852
                                        for( int b = 0; b < name_part_parts.length; b++ )
821
853
                                                if( name_part_parts[ b ].length() > 0 )
822
854
                                                {
823
 
                                                        if( value.length() == 0 ) value += " ";
 
855
                                                        if( value.length() > 0 ) value += " ";
824
856
                                                        value += name_part_parts[ b ];
825
857
                                                }
826
858
                                }
1000
1032
                        addNote( unescapeValue( value ) );
1001
1033
                }
1002
1034
 
 
1035
                private void parseBDAY( String[] params, String value )
 
1036
                {
 
1037
                        setBirthday( value );
 
1038
                }
 
1039
 
1003
1040
                public void finaliseVcard()
1004
1041
                        throws ParseException, ContactNotIdentifiableException
1005
1042
                {
1006
1043
                        // missing version (and data is present)
1007
 
                        if( _version == null && _buffers != null )
 
1044
                        if( _version == null && _content_lines != null )
1008
1045
                                throw new ParseException( R.string.error_vcf_malformed );
1009
1046
 
1010
1047
                        // finalise the parent class
1013
1050
 
1014
1051
                /**
1015
1052
                 * Amongst the params, find the value of the first, only, of any with
1016
 
                 * the specified name
 
1053
                 * the specified name.
 
1054
                 *
1017
1055
                 * @param params
1018
1056
                 * @param name
1019
1057
                 * @return a value, or null
1025
1063
                }
1026
1064
 
1027
1065
                /**
1028
 
                 * Amongst the params, find the values of any with the specified name
 
1066
                 * Amongst the params, find the values of any with the specified name.
 
1067
                 *
1029
1068
                 * @param params
1030
1069
                 * @param name
1031
1070
                 * @return an array of values, or null
1047
1086
                }
1048
1087
 
1049
1088
                /**
1050
 
                 * Amongst the params, return any type values present. For v2.1 vCards,
1051
 
                 * those types are just parameters. For v3.0, they are prefixed with
1052
 
                 * "TYPE=". There may also be multiple type parameters.
 
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
                 *
1053
1093
                 * @param params an array of params to look for types in
1054
1094
                 * @param valid_types an list of upper-case type values to look for
1055
1095
                 * @return a set of present type values
1107
1147
                                else if( ch == '=' && i == in.limit() - 1 )
1108
1148
                                {
1109
1149
                                        // we found a '=' at the end of a line signifying a multi-
1110
 
                                        // line string, so we don't add it.
 
1150
                                        // line string, so we don't add it
1111
1151
                                        another = true;
1112
1152
                                        continue;
1113
1153
                                }