/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: 2012-12-21 14:47:57 UTC
  • Revision ID: tim@ed.am-20121221144757-5cb1lgsp7fdt7p2n
updated TODO

Show diffs side-by-side

added added

removed removed

1
1
/*
2
2
 * VCFImporter.java
3
3
 *
4
 
 * Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
 
4
 * Copyright (C) 2009 to 2011 Tim Marston <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 )
137
 
                                {
 
136
                                if( !in_vcard ) {
138
137
                                        // look for vcard beginning
139
 
                                        if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
 
138
                                        if( line.matches( "^BEGIN:VCARD" ) ) {
140
139
                                                in_vcard = true;
141
140
                                                _vcard_count++;
142
141
                                        }
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
 
                                        }
148
142
                                }
149
 
                                else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
 
143
                                else if( line.matches( "^END:VCARD" ) )
150
144
                                        in_vcard = false;
151
145
                        }
152
146
 
181
175
                        // import
182
176
                        importVCardFileContent( content, file.getName() );
183
177
                }
184
 
                catch( OutOfMemoryError e ) {
185
 
                        showError( R.string.error_outofmemory );
186
 
                }
187
178
                catch( FileNotFoundException e ) {
188
179
                        showError( getText( R.string.error_filenotfound ) +
189
180
                                file.getName() );
202
193
                ContentLineIterator cli = new ContentLineIterator( content );
203
194
                while( cli.hasNext() )
204
195
                {
205
 
                        ContentLine content_line = cli.next();
 
196
                        ByteBuffer buffer = cli.next();
206
197
 
207
 
                        // get a US-ASCII version of the string, for processing
208
 
                        String line = content_line.getUsAsciiLine();
 
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
                        }
209
208
 
210
209
                        if( vcard == null ) {
211
210
                                // look for vcard beginning
212
 
                                if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
 
211
                                if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD" ) ) {
213
212
                                        setProgress( _progress++ );
214
213
                                        vcard = new Vcard();
215
214
                                        vcard_start_line = cli.getLineNumber();
217
216
                        }
218
217
                        else {
219
218
                                // look for vcard content or ending
220
 
                                if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
 
219
                                if( line.matches( "^END[ \t]*:[ \t]*VCARD" ) )
221
220
                                {
222
221
                                        // finalise the vcard/contact
223
222
                                        try {
260
259
                                {
261
260
                                        // try giving the line to the vcard
262
261
                                        try {
263
 
                                                vcard.parseLine( content_line );
 
262
                                                vcard.parseLine( buffer, line,
 
263
                                                        cli.doesNextLineLookFolded() );
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 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 >
 
292
        class ContentLineIterator implements Iterator< ByteBuffer >
335
293
        {
336
294
                protected byte[] _content = null;
337
295
                protected int _pos = 0;
349
307
                }
350
308
 
351
309
                @Override
352
 
                public ContentLine next()
 
310
                public ByteBuffer next()
353
311
                {
354
312
                        int initial_pos = _pos;
355
313
 
362
320
                                                _pos > initial_pos )? _pos - 1 : _pos;
363
321
                                        _pos++;
364
322
                                        _line++;
365
 
                                        return new ContentLine(
366
 
                                                ByteBuffer.wrap( _content, initial_pos,
367
 
                                                        to - initial_pos ),
368
 
                                                doesNextLineLookFolded() );
 
323
                                        return ByteBuffer.wrap( _content, initial_pos,
 
324
                                                to - initial_pos );
369
325
                                }
370
326
 
371
327
                        // we didn't find one, but were there bytes left?
373
329
                                int to = _pos;
374
330
                                _pos++;
375
331
                                _line++;
376
 
                                return new ContentLine(
377
 
                                        ByteBuffer.wrap( _content, initial_pos,
378
 
                                                to - initial_pos ),
379
 
                                        doesNextLineLookFolded() );
 
332
                                return ByteBuffer.wrap( _content, initial_pos,
 
333
                                        to - initial_pos );
380
334
                        }
381
335
 
382
336
                        // no bytes left
394
348
                 * onto the end of this one?
395
349
                 * @return
396
350
                 */
397
 
                private boolean doesNextLineLookFolded()
 
351
                public boolean doesNextLineLookFolded()
398
352
                {
399
353
                        return _pos > 0 && _pos < _content.length &&
400
354
                                _content[ _pos - 1 ] == '\n' &&
419
373
                private final static int MULTILINE_FOLDED = 3;  // MIME-DIR folding
420
374
 
421
375
                private String _version = null;
422
 
                private Vector< ContentLine > _content_lines = null;
 
376
                private Vector< ByteBuffer > _buffers = null;
423
377
                private int _name_level = NAMELEVEL_NONE;
424
378
                private int _parser_multiline_state = MULTILINE_NONE;
425
379
                private String _parser_current_name_and_params = null;
468
422
                @SuppressWarnings("serial")
469
423
                protected class SkipImportException extends Exception { }
470
424
 
471
 
                private String extractCollonPartFromLine( ContentLine content_line,
472
 
                        boolean former )
 
425
                private String extractCollonPartFromLine( ByteBuffer buffer,
 
426
                        String line, boolean former )
473
427
                {
 
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
 
474
443
                        // split line into name and value parts and check to make sure we
475
444
                        // only got 2 parts and that the first part is not zero in length
476
 
                        String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
 
445
                        String[] parts = line.split( ":", 2 );
477
446
                        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 )
 
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 )
494
465
                        throws ParseException, SkipImportException,
495
466
                        AbortImportException
496
467
                {
499
470
                        {
500
471
                                // tentatively get name and params from line
501
472
                                String name_and_params =
502
 
                                        extractNameAndParamsFromLine( content_line );
 
473
                                        extractNameAndParamsFromLine( buffer, line );
503
474
 
504
475
                                // is it a version line?
505
476
                                if( name_and_params != null &&
506
477
                                        name_and_params.equalsIgnoreCase( "VERSION" ) )
507
478
                                {
508
479
                                        // yes, get it!
509
 
                                        String value = extractValueFromLine( content_line );
510
 
                                        if( value == null || (
511
 
                                                !value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
512
 
                                        {
 
480
                                        String value = extractValueFromLine( buffer, line );
 
481
                                        if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
482
                                                throw new ParseException( R.string.error_vcf_version );
514
 
                                        }
515
483
                                        _version = value;
516
484
 
517
485
                                        // parse any buffers we've been accumulating while we waited
518
486
                                        // 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;
 
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;
523
495
                                }
524
496
                                else
525
497
                                {
526
498
                                        // 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 );
 
499
                                        if( _buffers == null )
 
500
                                                _buffers = new Vector< ByteBuffer >();
 
501
                                        _buffers.add( buffer );
530
502
                                }
531
503
                        }
532
504
                        else
544
516
 
545
517
                                        // skip some initial line characters, depending on the type
546
518
                                        // of multi-line we're handling
547
 
                                        pos = content_line.getBuffer().position();
 
519
                                        pos = buffer.position();
548
520
                                        switch( _parser_multiline_state )
549
521
                                        {
550
522
                                        case MULTILINE_FOLDED:
551
523
                                                pos++;
552
524
                                                break;
553
525
                                        case MULTILINE_ENCODED:
554
 
                                                while( pos < content_line.getBuffer().limit() && (
555
 
                                                        content_line.getBuffer().get( pos ) == ' ' ||
556
 
                                                        content_line.getBuffer().get( pos ) == '\t' ) )
 
526
                                                while( pos < buffer.limit() && (
 
527
                                                        buffer.get( pos ) == ' ' ||
 
528
                                                        buffer.get( pos ) == '\t' ) )
557
529
                                                {
558
530
                                                        pos++;
559
531
                                                }
569
541
                                else
570
542
                                {
571
543
                                        // skip empty lines
572
 
                                        if( content_line.getUsAsciiLine().trim().length() == 0 )
573
 
                                                return;
 
544
                                        if( line.trim().length() == 0 ) return;
574
545
 
575
546
                                        // get name and params from line, and since we're not
576
547
                                        // parsing a subsequent line in a multi-line, this should
577
548
                                        // not fail, or it's an error
578
549
                                        name_and_params =
579
 
                                                extractNameAndParamsFromLine( content_line );
 
550
                                                extractNameAndParamsFromLine( buffer, line );
580
551
                                        if( name_and_params == null )
581
552
                                                throw new ParseException(
582
553
                                                        R.string.error_vcf_malformed );
583
554
 
584
555
                                        // calculate how many chars to skip from beginning of line
585
556
                                        // so we skip the property "name:" part
586
 
                                        pos = content_line.getBuffer().position() +
587
 
                                                name_and_params.length() + 1;
 
557
                                        pos = buffer.position() + name_and_params.length() + 1;
588
558
 
589
559
                                        // reset the saved multi-line state
590
560
                                        _parser_current_name_and_params = name_and_params;
593
563
 
594
564
                                // get value from buffer, as raw bytes
595
565
                                ByteBuffer value;
596
 
                                value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
597
 
                                        content_line.getBuffer().limit() - pos );
 
566
                                value = ByteBuffer.wrap( buffer.array(), pos,
 
567
                                        buffer.limit() - pos );
598
568
 
599
569
                                // get parameter parts
600
570
                                String[] name_param_parts = name_and_params.split( ";", -1 );
688
658
                                // if we know we're not in an encoding-based multi-line, check
689
659
                                // to see if we're in a folded multi-line
690
660
                                if( _parser_multiline_state == MULTILINE_NONE &&
691
 
                                        content_line.doesNextLineLookFolded() )
 
661
                                        next_line_looks_folded )
692
662
                                {
693
663
                                        _parser_multiline_state = MULTILINE_FOLDED;
694
664
                                }
724
694
                                        parseLABEL( name_param_parts, complete_value );
725
695
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
726
696
                                        parseNOTE( name_param_parts, complete_value );
727
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
728
 
                                        parseBDAY( name_param_parts, complete_value );
729
697
                        }
730
698
                }
731
699
 
755
723
                        {
756
724
                                String str = parts.get( a );
757
725
 
758
 
                                // Look for parts that end in an escape character, but ignore
759
 
                                // the final part.  We've already detected escape chars at the
 
726
                                // look for parts that end in an escape character, but ignore
 
727
                                // the final part. We've already detected escape chars at the
760
728
                                // end of the final part in parseLine() and handled multi-lines
761
729
                                // accordingly.
762
730
                                if( a < parts.size() - 1 &&
852
820
                                        for( int b = 0; b < name_part_parts.length; b++ )
853
821
                                                if( name_part_parts[ b ].length() > 0 )
854
822
                                                {
855
 
                                                        if( value.length() > 0 ) value += " ";
 
823
                                                        if( value.length() == 0 ) value += " ";
856
824
                                                        value += name_part_parts[ b ];
857
825
                                                }
858
826
                                }
1032
1000
                        addNote( unescapeValue( value ) );
1033
1001
                }
1034
1002
 
1035
 
                private void parseBDAY( String[] params, String value )
1036
 
                {
1037
 
                        setBirthday( value );
1038
 
                }
1039
 
 
1040
1003
                public void finaliseVcard()
1041
1004
                        throws ParseException, ContactNotIdentifiableException
1042
1005
                {
1043
1006
                        // missing version (and data is present)
1044
 
                        if( _version == null && _content_lines != null )
 
1007
                        if( _version == null && _buffers != null )
1045
1008
                                throw new ParseException( R.string.error_vcf_malformed );
1046
1009
 
1047
1010
                        // finalise the parent class
1050
1013
 
1051
1014
                /**
1052
1015
                 * Amongst the params, find the value of the first, only, of any with
1053
 
                 * the specified name.
1054
 
                 *
 
1016
                 * the specified name
1055
1017
                 * @param params
1056
1018
                 * @param name
1057
1019
                 * @return a value, or null
1063
1025
                }
1064
1026
 
1065
1027
                /**
1066
 
                 * Amongst the params, find the values of any with the specified name.
1067
 
                 *
 
1028
                 * Amongst the params, find the values of any with the specified name
1068
1029
                 * @param params
1069
1030
                 * @param name
1070
1031
                 * @return an array of values, or null
1086
1047
                }
1087
1048
 
1088
1049
                /**
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
 
                 *
 
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.
1093
1053
                 * @param params an array of params to look for types in
1094
1054
                 * @param valid_types an list of upper-case type values to look for
1095
1055
                 * @return a set of present type values
1147
1107
                                else if( ch == '=' && i == in.limit() - 1 )
1148
1108
                                {
1149
1109
                                        // we found a '=' at the end of a line signifying a multi-
1150
 
                                        // line string, so we don't add it
 
1110
                                        // line string, so we don't add it.
1151
1111
                                        another = true;
1152
1112
                                        continue;
1153
1113
                                }