/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/org/waxworlds/edam/importcontacts/VCFImporter.java

  • Committer: edam
  • Date: 2011-05-05 21:49:43 UTC
  • Revision ID: edam@waxworlds.org-20110505214943-bg0cn6qz0gr49dlk
- updated TODO
- made varibale names consistent (camelCaseVariables now_use_underscores)

Show diffs side-by-side

added added

removed removed

1
1
/*
2
2
 * VCFImporter.java
3
3
 *
4
 
 * Copyright (C) 2009 Tim Marston <edam@waxworlds.org>
 
4
 * Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
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
32
32
import java.io.IOException;
33
33
import java.io.UnsupportedEncodingException;
34
34
import java.nio.ByteBuffer;
 
35
import java.util.ArrayList;
35
36
import java.util.Arrays;
 
37
import java.util.HashMap;
36
38
import java.util.HashSet;
 
39
import java.util.Iterator;
37
40
import java.util.List;
 
41
import java.util.NoSuchElementException;
38
42
import java.util.Set;
39
43
import java.util.Vector;
40
44
import java.util.regex.Matcher;
46
50
 
47
51
public class VCFImporter extends Importer
48
52
{
49
 
        private int _vCardCount = 0;
 
53
        private int _vcard_count = 0;
50
54
        private int _progress = 0;
51
55
 
52
56
        public VCFImporter( Doit doit )
106
110
                        countVCardFile( files[ i ] );
107
111
                        setTmpProgress( i );
108
112
                }
109
 
                setProgressMax( _vCardCount );  // will also update tmp progress
 
113
                setProgressMax( _vcard_count ); // will also update tmp progress
110
114
 
111
115
                // import them
112
116
                setProgress( 0 );
120
124
                {
121
125
                        // open file
122
126
                        BufferedReader reader = new BufferedReader(
123
 
                                        new FileReader( file ) );
 
127
                                new FileReader( file ) );
124
128
 
125
129
                        // read
126
130
                        String line;
127
 
                        boolean inVCard = false;
 
131
                        boolean in_vcard = false;
128
132
                        while( ( line = reader.readLine() ) != null )
129
133
                        {
130
 
                                if( !inVCard ) {
 
134
                                if( !in_vcard ) {
131
135
                                        // look for vcard beginning
132
 
                                        if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
133
 
                                                inVCard = true;
134
 
                                                _vCardCount++;
 
136
                                        if( line.matches( "^BEGIN:VCARD" ) ) {
 
137
                                                in_vcard = true;
 
138
                                                _vcard_count++;
135
139
                                        }
136
140
                                }
137
 
                                else if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
138
 
                                        inVCard = false;
 
141
                                else if( line.matches( "^END:VCARD" ) )
 
142
                                        in_vcard = false;
139
143
                        }
140
144
 
141
145
                }
178
182
        }
179
183
 
180
184
        private void importVCardFileContent( byte[] content, String fileName )
181
 
                        throws AbortImportException
 
185
                throws AbortImportException
182
186
        {
183
 
                ByteBuffer buffers[] = getLinesFromContent( content );
184
 
 
185
187
                // go through lines
186
 
                VCard vCard = null;
187
 
                for( int i = 0; i < buffers.length; i++ )
 
188
                VCard vcard = null;
 
189
                ContentLineIterator cli = new ContentLineIterator( content );
 
190
                while( cli.hasNext() )
188
191
                {
 
192
                        ByteBuffer buffer = cli.next();
 
193
 
189
194
                        // get a US-ASCII version of the line for processing
190
195
                        String line;
191
196
                        try {
192
 
                                line = new String( buffers[ i ].array(), buffers[ i ].position(),
193
 
                                        buffers[ i ].limit() - buffers[ i ].position(), "US-ASCII" );
 
197
                                line = new String( buffer.array(), buffer.position(),
 
198
                                        buffer.limit() - buffer.position(), "US-ASCII" );
194
199
                        }
195
200
                        catch( UnsupportedEncodingException e ) {
196
201
                                // we know US-ASCII is supported, so appease the compiler...
197
202
                                line = "";
198
203
                        }
199
204
 
200
 
                        if( vCard == null ) {
 
205
                        if( vcard == null ) {
201
206
                                // look for vcard beginning
202
 
                                if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
 
207
                                if( line.matches( "^BEGIN:VCARD" ) ) {
203
208
                                        setProgress( ++_progress );
204
 
                                        vCard = new VCard();
 
209
                                        vcard = new VCard();
205
210
                                }
206
211
                        }
207
212
                        else {
208
213
                                // look for vcard content or ending
209
 
                                if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
 
214
                                if( line.matches( "^END:VCARD" ) )
210
215
                                {
211
216
                                        // store vcard and do away with it
212
217
                                        try {
213
 
                                                vCard.finaliseParsing();
214
 
                                                importContact( vCard );
 
218
                                                vcard.finaliseParsing();
 
219
                                                importContact( vcard );
215
220
                                        }
216
221
                                        catch( VCard.ParseException e ) {
217
222
                                                skipContact();
218
223
                                                if( !showContinue(
219
 
                                                                getText( R.string.error_vcf_parse ).toString()
220
 
                                                                + fileName + "\n" + e.getMessage() ) )
 
224
                                                        getText( R.string.error_vcf_parse ).toString()
 
225
                                                        + fileName + "\n" + e.getMessage() ) )
 
226
                                                {
221
227
                                                        finish( ACTION_ABORT );
 
228
                                                }
222
229
                                        }
223
230
                                        catch( VCard.SkipContactException e ) {
224
231
                                                skipContact();
225
232
                                                // do nothing
226
233
                                        }
227
 
                                        vCard = null;
 
234
                                        vcard = null;
228
235
                                }
229
236
                                else
230
237
                                {
231
238
                                        // try giving the line to the vcard
232
239
                                        try {
233
 
                                                vCard.parseLine( buffers[ i ] );
 
240
                                                vcard.parseLine( buffer, line,
 
241
                                                        cli.doesNextLineLookFolded() );
234
242
                                        }
235
243
                                        catch( VCard.ParseException e ) {
236
244
                                                skipContact();
237
245
                                                if( !showContinue(
238
 
                                                                getText( R.string.error_vcf_parse ).toString()
239
 
                                                                + fileName + "\n" + e.getMessage() ) )
 
246
                                                        getText( R.string.error_vcf_parse ).toString()
 
247
                                                        + fileName + "\n" + e.getMessage() ) )
 
248
                                                {
240
249
                                                        finish( ACTION_ABORT );
 
250
                                                }
241
251
 
242
252
                                                // although we're continuing, we still need to abort
243
253
                                                // this vCard. Further lines will be ignored until we
244
254
                                                // get to another BEGIN:VCARD line.
245
 
                                                vCard = null;
 
255
                                                vcard = null;
246
256
                                        }
247
257
                                        catch( VCard.SkipContactException e ) {
248
258
                                                skipContact();
249
259
                                                // abort this vCard. Further lines will be ignored until
250
260
                                                // we get to another BEGIN:VCARD line.
251
 
                                                vCard = null;
 
261
                                                vcard = null;
252
262
                                        }
253
263
                                }
254
264
                        }
255
265
                }
256
266
        }
257
267
 
258
 
        private ByteBuffer[] getLinesFromContent( byte[] content )
 
268
        class ContentLineIterator implements Iterator< ByteBuffer >
259
269
        {
260
 
                // count lines in data
261
 
                int num_lines = 1;
262
 
                for( int a = 0; a < content.length; a++ )
263
 
                        if( content[ a ] == '\n' )
264
 
                                num_lines++;
265
 
 
266
 
                // get lines, removing \r's and \n's as we go
267
 
                ByteBuffer lines[] = new ByteBuffer[ num_lines ];
268
 
                int last = 0;
269
 
                for( int a = 0, b = 0; a < content.length; a++ )
270
 
                        if( content[ a ] == '\n' ) {
271
 
                                int to = ( a > 0 && content[ a - 1 ] == '\r' &&
272
 
                                        a - 1 >= last )? a - 1 : a;
273
 
                                lines[ b++ ] = ByteBuffer.wrap( content, last, to - last );
274
 
                                last = a + 1;
 
270
                protected byte[] _content = null;
 
271
                protected int _pos = 0;
 
272
 
 
273
                public ContentLineIterator( byte[] content )
 
274
                {
 
275
                        _content = content;
 
276
                }
 
277
 
 
278
                @Override
 
279
                public boolean hasNext()
 
280
                {
 
281
                        return _pos < _content.length;
 
282
                }
 
283
 
 
284
                @Override
 
285
                public ByteBuffer next()
 
286
                {
 
287
                        int initial_pos = _pos;
 
288
 
 
289
                        // find newline
 
290
                        for( ; _pos < _content.length; _pos++ )
 
291
                                if( _content[ _pos ] == '\n' )
 
292
                                {
 
293
                                        // adjust for a \r preceding the \n
 
294
                                        int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
 
295
                                                _pos > initial_pos )? _pos - 1 : _pos;
 
296
                                        _pos++;
 
297
                                        return ByteBuffer.wrap( _content, initial_pos,
 
298
                                                to - initial_pos );
 
299
                                }
 
300
 
 
301
                        // we didn't find one, but were there bytes left?
 
302
                        if( _pos != initial_pos ) {
 
303
                                int to = _pos;
 
304
                                _pos++;
 
305
                                return ByteBuffer.wrap( _content, initial_pos,
 
306
                                        to - initial_pos );
275
307
                        }
276
 
                lines[ lines.length - 1 ] = ByteBuffer.wrap( content, last,
277
 
                        content.length - last );
278
 
 
279
 
                return lines;
 
308
 
 
309
                        // no bytes left
 
310
                        throw new NoSuchElementException();
 
311
                }
 
312
 
 
313
                @Override
 
314
                public void remove()
 
315
                {
 
316
                        throw new UnsupportedOperationException();
 
317
                }
 
318
 
 
319
                /**
 
320
                 * Does the next line, if there is one, look like it should be folded
 
321
                 * onto the end of this one?
 
322
                 * @return
 
323
                 */
 
324
                public boolean doesNextLineLookFolded()
 
325
                {
 
326
                        return _pos > 0 && _pos < _content.length &&
 
327
                                _content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
 
328
                }
280
329
        }
281
330
 
282
331
        private class VCard extends ContactData
283
332
        {
284
333
                private final static int NAMELEVEL_NONE = 0;
285
 
                private final static int NAMELEVEL_ORG = 1;
286
 
                private final static int NAMELEVEL_FN = 2;
287
 
                private final static int NAMELEVEL_N = 3;
 
334
                private final static int NAMELEVEL_FN = 1;
 
335
                private final static int NAMELEVEL_N = 2;
 
336
 
 
337
                private final static int MULTILINE_NONE = 0;
 
338
                private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
 
339
                private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
 
340
                private final static int MULTILINE_FOLDED = 3;  // v3.0 folding
288
341
 
289
342
                private String _version = null;
290
343
                private Vector< ByteBuffer > _buffers = null;
291
344
                private int _name_level = NAMELEVEL_NONE;
292
 
                private boolean _parser_in_multiline = false;
 
345
                private int _parser_multiline_state = MULTILINE_NONE;
293
346
                private String _parser_current_name_and_params = null;
294
347
                private String _parser_buffered_value_so_far = "";
 
348
                private String _cached_organisation = null;
 
349
                private String _cached_title = null;
295
350
 
296
351
                protected class UnencodeResult
297
352
                {
334
389
                @SuppressWarnings("serial")
335
390
                protected class SkipContactException extends Exception { }
336
391
 
337
 
                public void parseLine( ByteBuffer buffer )
338
 
                                throws ParseException, SkipContactException,
339
 
                                AbortImportException
340
 
                {
341
 
                        // get a US-ASCII version of the line for processing
342
 
                        String line;
343
 
                        try {
344
 
                                line = new String( buffer.array(), buffer.position(),
345
 
                                        buffer.limit() - buffer.position(), "US-ASCII" );
346
 
                        }
347
 
                        catch( UnsupportedEncodingException e ) {
348
 
                                // we know US-ASCII is supported, so appease the compiler...
349
 
                                line = "";
350
 
                        }
351
 
 
352
 
                        // ignore empty lines
353
 
                        if( line.trim() == "" ) return;
354
 
 
355
 
                        // split line into name and value parts (this may turn out to be
356
 
                        // unwanted if the line is a subsequent line in a multi-line
357
 
                        // value, but we have to do this now to check for and handle VCF
358
 
                        // versions first). Also, the value part is only created tentatively
359
 
                        // because it may have an encoding/charset. Since we're treating it
360
 
                        // as UTF-8 (which is compatible with 7-bit US-ASCII) this is ok
361
 
                        // though so long as we later use the raw bytes. ALso we check for
362
 
                        // malformed property:name pairs.
363
 
                        String name_and_params, string_value;
364
 
                        {
365
 
                                String[] parts = line.split( ":", 2 );
366
 
                                if( parts.length == 2 ) {
367
 
                                        name_and_params = parts[ 0 ].trim();
368
 
                                        string_value = parts[ 1 ].trim();
369
 
                                        if( name_and_params.length() == 0 )
370
 
                                                throw new ParseException( R.string.error_vcf_malformed );
371
 
                                }
372
 
                                else
373
 
                                {
374
 
                                        if( !_parser_in_multiline )
375
 
                                                throw new ParseException( R.string.error_vcf_malformed );
376
 
                                        name_and_params = null;
377
 
                                        string_value = null;
378
 
                                }
379
 
                        }
380
 
 
381
 
                        // if we haven't yet got a version, we won't be paring anything!
 
392
                private String extractCollonPartFromLine( ByteBuffer buffer,
 
393
                        String line, boolean former )
 
394
                {
 
395
                        String ret = null;
 
396
 
 
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
                        // split line into name and value parts and check to make sure we
 
411
                        // only got 2 parts and that the first part is not zero in length
 
412
                        String[] parts = line.split( ":", 2 );
 
413
                        if( parts.length == 2 && parts[ 0 ].length() > 0 )
 
414
                                ret = parts[ former? 0 : 1 ];
 
415
 
 
416
                        return ret;
 
417
                }
 
418
 
 
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,
 
433
                        AbortImportException
 
434
                {
 
435
                        // do we have a version yet?
382
436
                        if( _version == null )
383
437
                        {
384
 
                                // is this a version?
385
 
                                if( name_and_params.equals( "VERSION" ) )
 
438
                                // tentatively get name and params from line
 
439
                                String name_and_params =
 
440
                                        extractNameAndParamsFromLine( buffer, line );
 
441
 
 
442
                                // is it a version line?
 
443
                                if( name_and_params != null &&
 
444
                                        name_and_params.equals( "VERSION" ) )
386
445
                                {
387
 
                                        // yes, check/store it
388
 
                                        if( !string_value.equals( "2.1" ) &&
389
 
                                                        !string_value.equals( "3.0" ) )
 
446
                                        // yes, get it!
 
447
                                        String value = extractValueFromLine( buffer, line );
 
448
                                        if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
390
449
                                                throw new ParseException( R.string.error_vcf_version );
391
 
                                        _version = string_value;
 
450
                                        _version = value;
392
451
 
393
 
                                        // parse any other buffers we've accumulated so far
 
452
                                        // parse any buffers we've been accumulating while we waited
 
453
                                        // for a version
394
454
                                        if( _buffers != null )
395
455
                                                for( int i = 0; i < _buffers.size(); i++ )
396
 
                                                        parseLine( _buffers.get( 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() ) == ' ' );
397
461
                                        _buffers = null;
398
462
                                }
399
463
                                else
400
464
                                {
401
 
                                        // no, so stash this buffer till we have a version
 
465
                                        // no, so stash this line till we get a version
402
466
                                        if( _buffers == null )
403
467
                                                _buffers = new Vector< ByteBuffer >();
404
468
                                        _buffers.add( buffer );
406
470
                        }
407
471
                        else
408
472
                        {
409
 
                                // value bytes, for processing
410
 
                                ByteBuffer value;
 
473
                                // name and params and the position in the buffer where the
 
474
                                // "value" part of the line start
 
475
                                String name_and_params;
 
476
                                int pos;
411
477
 
412
 
                                if( _parser_in_multiline )
 
478
                                if( _parser_multiline_state != MULTILINE_NONE )
413
479
                                {
414
480
                                        // if we're currently in a multi-line value, use the stored
415
481
                                        // property name and parameters
416
482
                                        name_and_params = _parser_current_name_and_params;
417
483
 
418
 
                                        // find start of string (skip spaces/tabs)
419
 
                                        int pos = buffer.position();
420
 
                                        byte[] buffer_array = buffer.array();
421
 
                                        while( pos < buffer.limit() && (
422
 
                                                buffer_array[ pos ] == ' ' ||
423
 
                                                buffer_array[ pos ] == '\t' ) )
 
484
                                        // skip some initial line characters, depending on the type
 
485
                                        // of multi-line we're handling
 
486
                                        pos = buffer.position();
 
487
                                        switch( _parser_multiline_state )
424
488
                                        {
 
489
                                        case MULTILINE_FOLDED:
425
490
                                                pos++;
 
491
                                                break;
 
492
                                        case MULTILINE_ENCODED:
 
493
                                                while( pos < buffer.limit() && (
 
494
                                                        buffer.get( pos ) == ' ' ||
 
495
                                                        buffer.get( pos ) == '\t' ) )
 
496
                                                {
 
497
                                                        pos++;
 
498
                                                }
 
499
                                                break;
 
500
                                        default:
 
501
                                                // do nothing
426
502
                                        }
427
503
 
428
 
                                        // get value from buffer
429
 
                                        value = ByteBuffer.wrap( buffer.array(), pos,
430
 
                                                buffer.limit() - pos );
 
504
                                        // take us out of multi-line so that we can re-detect that
 
505
                                        // this line is a multi-line or not
 
506
                                        _parser_multiline_state = MULTILINE_NONE;
431
507
                                }
432
508
                                else
433
509
                                {
434
 
                                        // ignore empty values
435
 
                                        if( string_value.length() < 1 ) return;
 
510
                                        // get name and params from line, and since we're not
 
511
                                        // parsing a subsequent line in a multi-line, this should
 
512
                                        // not fail, or it's an error
 
513
                                        name_and_params =
 
514
                                                extractNameAndParamsFromLine( buffer, line );
 
515
                                        if( name_and_params == null )
 
516
                                                throw new ParseException(
 
517
                                                        R.string.error_vcf_malformed );
436
518
 
437
519
                                        // calculate how many chars to skip from beginning of line
438
520
                                        // so we skip the property "name:" part
439
 
                                        int pos = buffer.position() + name_and_params.length() + 1;
440
 
 
441
 
                                        // get value from buffer
442
 
                                        value = ByteBuffer.wrap( buffer.array(), pos,
443
 
                                                buffer.limit() - pos );
 
521
                                        pos = buffer.position() + name_and_params.length() + 1;
444
522
 
445
523
                                        // reset the saved multi-line state
446
524
                                        _parser_current_name_and_params = name_and_params;
447
525
                                        _parser_buffered_value_so_far = "";
448
526
                                }
449
527
 
 
528
                                // get value from buffer, as raw bytes
 
529
                                ByteBuffer value;
 
530
                                value = ByteBuffer.wrap( buffer.array(), pos,
 
531
                                        buffer.limit() - pos );
 
532
 
450
533
                                // get parameter parts
451
534
                                String[] name_param_parts = name_and_params.split( ";", -1 );
452
535
                                for( int i = 0; i < name_param_parts.length; i++ )
466
549
                                String charset = checkParam( name_param_parts, "CHARSET" );
467
550
                                if( charset != null ) charset = charset.toUpperCase();
468
551
                                if( charset != null && !charset.equals( "US-ASCII" ) &&
469
 
                                        !charset.equals( "ASCII" ) && !charset.equals( "UTF-8" ) )
 
552
                                        !charset.equals( "ASCII" ) &&
 
553
                                        !charset.equals( "UTF-8" ) )
470
554
                                {
471
555
                                        throw new ParseException( R.string.error_vcf_charset );
472
556
                                }
477
561
                                if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
478
562
                                        unencoding_result = unencodeQuotedPrintable( value );
479
563
//                              else if( encoding != null && encoding.equals( "BASE64" ) )
480
 
//                                      result = unencodeBase64( props[ 1 ], charset );
 
564
//                                      unencoding_result = unencodeBase64( props[ 1 ], charset );
481
565
                                if( unencoding_result != null ) {
482
566
                                        value = unencoding_result.getBuffer();
483
 
                                        _parser_in_multiline =
484
 
                                                unencoding_result.isAnotherLineRequired();
 
567
                                        if( unencoding_result.isAnotherLineRequired() )
 
568
                                                _parser_multiline_state = MULTILINE_ENCODED;
485
569
                                }
486
570
 
487
571
                                // convert 8-bit ASCII charset to US-ASCII
488
 
                                if( charset == null || charset == "ASCII" ) {
 
572
                                if( charset == null || charset.equals( "ASCII" ) ) {
489
573
                                        value = transcodeAsciiToUtf8( value );
490
574
                                        charset = "UTF-8";
491
575
                                }
492
576
 
493
577
                                // process charset
 
578
                                String string_value;
494
579
                                try {
495
 
                                        string_value =
496
 
                                                new String( value.array(), value.position(),
497
 
                                                        value.limit() - value.position(), charset );
 
580
                                        string_value = new String( value.array(), value.position(),
 
581
                                                value.limit() - value.position(), charset );
498
582
                                } catch( UnsupportedEncodingException e ) {
499
583
                                        throw new ParseException( R.string.error_vcf_charset );
500
584
                                }
501
585
 
502
 
                                // handle multi-line requests
503
 
                                if( _parser_in_multiline ) {
 
586
                                // for some entries that have semicolon-separated value parts,
 
587
                                // check to see if the value ends in an escape character, which
 
588
                                // 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" ) ) &&
 
592
                                        doesStringEndInAnEscapeChar( string_value ) )
 
593
                                {
 
594
                                        _parser_multiline_state = MULTILINE_ESCAPED;
 
595
                                        string_value = string_value.substring( 0,
 
596
                                                string_value.length() - 1 );
 
597
                                }
 
598
 
 
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
 
601
                                if( _parser_multiline_state == MULTILINE_NONE &&
 
602
                                        _version.equals( "3.0" ) && next_line_looks_folded )
 
603
                                {
 
604
                                        _parser_multiline_state = MULTILINE_FOLDED;
 
605
                                }
 
606
 
 
607
                                // handle multi-lines by buffering them and parsing them when we
 
608
                                // are processing the last line in a multi-line sequence
 
609
                                if( _parser_multiline_state != MULTILINE_NONE ) {
504
610
                                        _parser_buffered_value_so_far += string_value;
505
611
                                        return;
506
612
                                }
507
 
 
508
 
                                // add on buffered multi-line content
509
613
                                String complete_value =
510
 
                                        _parser_buffered_value_so_far + string_value;
 
614
                                        ( _parser_buffered_value_so_far + string_value ).trim();
 
615
 
 
616
                                // ignore empty values
 
617
                                if( complete_value.length() < 1 ) return;
511
618
 
512
619
                                // parse some properties
513
620
                                if( name_param_parts[ 0 ].equals( "N" ) )
516
623
                                        parseFN( name_param_parts, complete_value );
517
624
                                else if( name_param_parts[ 0 ].equals( "ORG" ) )
518
625
                                        parseORG( name_param_parts, complete_value );
 
626
                                else if( name_param_parts[ 0 ].equals( "TITLE" ) )
 
627
                                        parseTITLE( name_param_parts, complete_value );
519
628
                                else if( name_param_parts[ 0 ].equals( "TEL" ) )
520
629
                                        parseTEL( name_param_parts, complete_value );
521
630
                                else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
522
631
                                        parseEMAIL( name_param_parts, complete_value );
523
 
                        }
 
632
                                else if( name_param_parts[ 0 ].equals( "ADR" ) )
 
633
                                        parseADR( name_param_parts, complete_value );
 
634
                        }
 
635
                }
 
636
 
 
637
                private boolean doesStringEndInAnEscapeChar( String string )
 
638
                {
 
639
                        // count the number of backslashes at the end of the string
 
640
                        int count = 0;
 
641
                        for( int a = string.length() - 1; a >= 0; a-- )
 
642
                                if( string.charAt( a ) == '\\' )
 
643
                                        count++;
 
644
                                else
 
645
                                        break;
 
646
 
 
647
                        // if there are an even number of backslashes then the final one
 
648
                        // doesn't count
 
649
                        return ( count & 1 ) == 1;
 
650
                }
 
651
 
 
652
                private String[] splitValueBySemicolon( String value )
 
653
                {
 
654
                        // split string in to parts by semicolon
 
655
                        ArrayList< String > parts = new ArrayList< String >(
 
656
                                Arrays.asList( value.split(  ";" ) ) );
 
657
 
 
658
                        // go through parts
 
659
                        for( int a = 0; a < parts.size(); a++ )
 
660
                        {
 
661
                                String str = parts.get( a );
 
662
 
 
663
                                // look for parts that end in an escape character, but ignore
 
664
                                // the final part. We've already detected escape chars at the
 
665
                                // end of the final part in parseLine() and handled multi-lines
 
666
                                // accordingly.
 
667
                                if( a < parts.size() - 1 &&
 
668
                                        doesStringEndInAnEscapeChar( str ) )
 
669
                                {
 
670
                                        // join the next part to this part and remove the next part
 
671
                                        parts.set( a, str.substring( 0, str.length() - 1 ) +
 
672
                                                ';' + parts.get( a + 1 ) );
 
673
                                        parts.remove( a + 1 );
 
674
 
 
675
                                        // re-visit this part
 
676
                                        a--;
 
677
                                        continue;
 
678
                                }
 
679
 
 
680
                                // trim and replace string
 
681
                                str = str.trim();
 
682
                                parts.set( a, str );
 
683
                        }
 
684
 
 
685
                        String[] ret = new String[ parts.size() ];
 
686
                        return parts.toArray( ret );
524
687
                }
525
688
 
526
689
                private void parseN( String[] params, String value )
527
 
                                throws ParseException, SkipContactException,
528
 
                                AbortImportException
529
690
                {
530
691
                        // already got a better name?
531
692
                        if( _name_level >= NAMELEVEL_N ) return;
532
693
 
533
694
                        // get name parts
534
 
                        String[] name_parts = value.split( ";" );
535
 
                        for( int i = 0; i < name_parts.length; i++ )
536
 
                                name_parts[ i ] = name_parts[ i ].trim();
 
695
                        String[] name_parts = splitValueBySemicolon( value );
537
696
 
538
697
                        // build name
539
698
                        value = "";
545
704
                        // set name
546
705
                        setName( value );
547
706
                        _name_level = NAMELEVEL_N;
548
 
 
549
 
                        // check now to see if we need to import this contact (to avoid
550
 
                        // parsing the rest of the vCard unnecessarily)
551
 
                        if( !isImportRequired( getName() ) )
552
 
                                throw new SkipContactException();
553
707
                }
554
708
 
555
709
                private void parseFN( String[] params, String value )
556
 
                                throws ParseException, SkipContactException
557
710
                {
558
711
                        // already got a better name?
559
712
                        if( _name_level >= NAMELEVEL_FN ) return;
564
717
                }
565
718
 
566
719
                private void parseORG( String[] params, String value )
567
 
                                throws ParseException, SkipContactException
568
720
                {
569
 
                        // already got a better name?
570
 
                        if( _name_level >= NAMELEVEL_ORG ) return;
571
 
 
572
721
                        // get org parts
573
 
                        String[] org_parts = value.split( ";" );
574
 
                        for( int i = 0; i < org_parts.length; i++ )
575
 
                                org_parts[ i ] = org_parts[ i ].trim();
576
 
 
577
 
                        // build name
578
 
                        if( org_parts.length > 1 && org_parts[ 0 ].length() == 0 )
579
 
                                value = org_parts[ 1 ];
580
 
                        else
581
 
                                value = org_parts[ 0 ];
582
 
 
583
 
                        // set name
584
 
                        setName( value );
585
 
                        _name_level = NAMELEVEL_ORG;
 
722
                        String[] org_parts = splitValueBySemicolon( value );
 
723
                        if( org_parts == null || org_parts.length < 1 ) return;
 
724
 
 
725
                        // build organisation name
 
726
                        StringBuilder builder = new StringBuilder(
 
727
                                String.valueOf( org_parts[ 0 ] ) );
 
728
                        for( int a = 1; a < org_parts.length; a++ )
 
729
                                builder.append( ", " ).append( org_parts[ a ] );
 
730
                        String organisation = builder.toString();
 
731
 
 
732
                        // set organisation name (using a title we've previously found)
 
733
                        addOrganisation( organisation, _cached_title, true );
 
734
 
 
735
                        // if we've not previously found a title, store this organisation
 
736
                        // name (we'll need it when we find a title to update the
 
737
                        // organisation, by name), else if we *have* previously found a
 
738
                        // title, clear it (since we just used it)
 
739
                        if( _cached_title == null )
 
740
                                _cached_organisation = organisation;
 
741
                        else
 
742
                                _cached_title = null;
 
743
                }
 
744
 
 
745
                private void parseTITLE( String[] params, String value )
 
746
                {
 
747
                        // if we previously had an organisation, look it up and append this
 
748
                        // title to it
 
749
                        if( _cached_organisation != null && hasOrganisations() ) {
 
750
                                HashMap< String, ExtraDetail > datas = getOrganisations();
 
751
                                ExtraDetail detail = datas.get( _cached_organisation );
 
752
                                if( detail != null )
 
753
                                        detail.setExtra( value );
 
754
                        }
 
755
 
 
756
                        // same as when handling organisation, if we've not previously found
 
757
                        // an organisation we store this title, else we clear it (since we
 
758
                        // just appended this title to it)
 
759
                        if( _cached_organisation == null )
 
760
                                _cached_title = value;
 
761
                        else
 
762
                                _cached_organisation = null;
586
763
                }
587
764
 
588
765
                private void parseTEL( String[] params, String value )
589
 
                                throws ParseException
590
766
                {
591
767
                        if( value.length() == 0 ) return;
592
768
 
593
769
                        Set< String > types = extractTypes( params, Arrays.asList(
594
 
                                        "PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
595
 
                                        "PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
 
770
                                "PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
 
771
                                "PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
596
772
 
597
773
                        // here's the logic...
598
 
                        boolean preferred = types.contains( "PREF" );
599
 
                        int type = PhonesColumns.TYPE_MOBILE;
600
 
                        if( types.contains( "VOICE" ) )
601
 
                                if( types.contains( "WORK" ) )
602
 
                                        type = PhonesColumns.TYPE_WORK;
603
 
                                else
604
 
                                        type = PhonesColumns.TYPE_HOME;
605
 
                        else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
606
 
                                type = PhonesColumns.TYPE_MOBILE;
 
774
                        boolean is_preferred = types.contains( "PREF" );
 
775
                        int type;
607
776
                        if( types.contains( "FAX" ) )
608
777
                                if( types.contains( "HOME" ) )
609
778
                                        type = PhonesColumns.TYPE_FAX_HOME;
610
779
                                else
611
780
                                        type = PhonesColumns.TYPE_FAX_WORK;
612
 
                        if( types.contains( "PAGER" ) )
 
781
                        else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
 
782
                                type = PhonesColumns.TYPE_MOBILE;
 
783
                        else if( types.contains( "PAGER" ) )
613
784
                                type = PhonesColumns.TYPE_PAGER;
 
785
                        else if( types.contains( "WORK" ) )
 
786
                                type = PhonesColumns.TYPE_WORK;
 
787
                        else
 
788
                                type = PhonesColumns.TYPE_HOME;
614
789
 
615
790
                        // add phone number
616
 
                        addPhone( value, type, preferred );
 
791
                        addNumber( value, type, is_preferred );
617
792
                }
618
793
 
619
794
                public void parseEMAIL( String[] params, String value )
620
 
                                throws ParseException
621
795
                {
622
796
                        if( value.length() == 0 ) return;
623
797
 
624
798
                        Set< String > types = extractTypes( params, Arrays.asList(
625
 
                                        "PREF", "WORK", "HOME", "INTERNET" ) );
626
 
 
627
 
                        // here's the logic...
628
 
                        boolean preferred = types.contains( "PREF" );
629
 
                        if( types.contains( "WORK" ) )
630
 
                                addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
631
 
                        else
632
 
                                addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
 
799
                                "PREF", "WORK", "HOME", "INTERNET" ) );
 
800
 
 
801
                        // add email address
 
802
                        boolean is_preferred = types.contains( "PREF" );
 
803
                        int type;
 
804
                        if( types.contains( "WORK" ) )
 
805
                                type = Contacts.ContactMethods.TYPE_WORK;
 
806
                        else
 
807
                                type = Contacts.ContactMethods.TYPE_HOME;
 
808
 
 
809
                        addEmail( value, type, is_preferred );
 
810
                }
 
811
 
 
812
                private void parseADR( String[] params, String value )
 
813
                {
 
814
                        // get address parts
 
815
                        String[] adr_parts = splitValueBySemicolon( value );
 
816
 
 
817
                        // build address
 
818
                        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 );
633
835
                }
634
836
 
635
837
                public void finaliseParsing()
636
 
                                throws ParseException, SkipContactException,
637
 
                                AbortImportException
 
838
                        throws ParseException, SkipContactException,
 
839
                        AbortImportException
638
840
                {
639
841
                        // missing version (and data is present)
640
842
                        if( _version == null && _buffers != null )
641
843
                                throw new ParseException( R.string.error_vcf_malformed );
642
844
 
643
 
                        //  missing name properties?
644
 
                        if( _name_level == NAMELEVEL_NONE )
645
 
                                throw new ParseException( R.string.error_vcf_noname );
646
 
 
647
 
                        // check if we should import this one? If we've already got an 'N'-
648
 
                        // type name, this will already have been done by parseN() so we
649
 
                        // mustn't do this here (or it could prompt twice!)
650
 
                        if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
651
 
                                throw new SkipContactException();
 
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
                        }
652
853
                }
653
854
 
654
855
                private String checkParam( String[] params, String name )
655
856
                {
656
 
                        Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
 
857
                        Pattern p = Pattern.compile(
 
858
                                "^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
657
859
                        for( int i = 0; i < params.length; i++ ) {
658
860
                                Matcher m = p.matcher( params[ i ] );
659
861
                                if( m.matches() )
660
 
                                        return m.group( 1 );
 
862
                                        return m.group( 2 );
661
863
                        }
662
864
                        return null;
663
865
                }
664
866
 
665
867
                private Set< String > extractTypes( String[] params,
666
 
                                List< String > valid_types )
 
868
                        List< String > valid_types )
667
869
                {
668
870
                        HashSet< String > types = new HashSet< String >();
669
871
 
690
892
                {
691
893
                        boolean another = false;
692
894
 
693
 
                        // unencode quoted-pritable encoding, as per RFC1521 section 5.1
 
895
                        // unencode quoted-printable encoding, as per RFC1521 section 5.1
694
896
                        byte[] out = new byte[ in.limit() - in.position() ];
695
897
                        int j = 0;
696
898
                        for( int i = in.position(); i < in.limit(); i++ )
701
903
                                {
702
904
                                        // we found a =XX format byte, add it
703
905
                                        out[ j ] = (byte)(
704
 
                                                Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
705
 
                                                Character.digit( in.array()[ i + 2 ], 16 ) );
 
906
                                                        Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
 
907
                                                        Character.digit( in.array()[ i + 2 ], 16 ) );
706
908
                                        i += 2;
707
909
                                }
708
910
                                else if( ch == '=' && i == in.limit() - 1 )