/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/VcardImporter.java

  • Committer: edam
  • Date: 2011-05-30 15:33:01 UTC
  • Revision ID: edam@waxworlds.org-20110530153301-oor6ci9b3hf9clul
- refactored some code to do with how contacts are imported
- Vcards (and ContactData) instances now generate a CacheIdentifier when they are finalised so that ContactData instances that do not have enough information to identify them can be discovered then
- importContact() now calls the private method checkForDuplicate(), renamed from isImportRequired(), and return if it is not
- importContact() and checkForDuplicate() now use the ContactData's generated CacheIdentifier

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 <edam@waxworlds.org>
5
5
 *
6
6
 * This file is part of the Import Contacts program (hereafter referred
7
 
 * to as "this program").  For more information, see
8
 
 * http://ed.am/dev/android/import-contacts
 
7
 * to as "this program"). For more information, see
 
8
 * http://www.waxworlds.org/edam/software/android/import-contacts
9
9
 *
10
10
 * This program is free software: you can redistribute it and/or modify
11
11
 * it under the terms of the GNU General Public License as published by
21
21
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
22
22
 */
23
23
 
24
 
package am.ed.importcontacts;
 
24
package org.waxworlds.edam.importcontacts;
25
25
 
26
26
import java.io.BufferedReader;
27
27
import java.io.File;
38
38
import java.util.HashSet;
39
39
import java.util.Iterator;
40
40
import java.util.List;
41
 
import java.util.Locale;
42
41
import java.util.NoSuchElementException;
43
42
import java.util.Set;
44
43
import java.util.Vector;
45
44
import java.util.regex.Matcher;
46
45
import java.util.regex.Pattern;
47
46
 
48
 
import android.annotation.SuppressLint;
49
47
import android.content.SharedPreferences;
 
48
import android.provider.Contacts;
 
49
import android.provider.Contacts.PhonesColumns;
50
50
 
51
51
public class VcardImporter extends Importer
52
52
{
58
58
                super( doit );
59
59
        }
60
60
 
61
 
        @SuppressLint( "SdCardPath" )
62
61
        @Override
63
62
        protected void onImport() throws AbortImportException
64
63
        {
83
82
                                // get files
84
83
                                class VCardFilter implements FilenameFilter {
85
84
                                        public boolean accept( File dir, String name ) {
86
 
                                                return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
 
85
                                                return name.toLowerCase().endsWith( ".vcf" );
87
86
                                        }
88
87
                                }
89
88
                                files = file.listFiles( new VCardFilter() );
117
116
                setProgress( 0 );
118
117
                for( int i = 0; i < files.length; i++ )
119
118
                        importVCardFile( files[ i ] );
120
 
                setProgress( _vcard_count );
121
119
        }
122
120
 
123
121
        private void countVCardFile( File file ) throws AbortImportException
133
131
                        boolean in_vcard = false;
134
132
                        while( ( line = reader.readLine() ) != null )
135
133
                        {
136
 
                                if( !in_vcard )
137
 
                                {
 
134
                                if( !in_vcard ) {
138
135
                                        // look for vcard beginning
139
 
                                        if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
 
136
                                        if( line.matches( "^BEGIN:VCARD" ) ) {
140
137
                                                in_vcard = true;
141
138
                                                _vcard_count++;
142
139
                                        }
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
140
                                }
149
 
                                else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
 
141
                                else if( line.matches( "^END:VCARD" ) )
150
142
                                        in_vcard = false;
151
143
                        }
152
144
 
176
168
                        FileInputStream istream = new FileInputStream( file );
177
169
                        byte[] content = new byte[ (int)file.length() ];
178
170
                        istream.read( content );
179
 
                        istream = null;
180
171
 
181
172
                        // import
182
173
                        importVCardFileContent( content, file.getName() );
183
174
                }
184
 
                catch( OutOfMemoryError e ) {
185
 
                        showError( R.string.error_outofmemory );
186
 
                }
187
175
                catch( FileNotFoundException e ) {
188
176
                        showError( getText( R.string.error_filenotfound ) +
189
177
                                file.getName() );
198
186
        {
199
187
                // go through lines
200
188
                Vcard vcard = null;
201
 
                int vcard_start_line = 0;
202
189
                ContentLineIterator cli = new ContentLineIterator( content );
203
190
                while( cli.hasNext() )
204
191
                {
205
 
                        ContentLine content_line = cli.next();
 
192
                        ByteBuffer buffer = cli.next();
206
193
 
207
 
                        // get a US-ASCII version of the string, for processing
208
 
                        String line = content_line.getUsAsciiLine();
 
194
                        // get a US-ASCII version of the line for processing
 
195
                        String line;
 
196
                        try {
 
197
                                line = new String( buffer.array(), buffer.position(),
 
198
                                        buffer.limit() - buffer.position(), "US-ASCII" );
 
199
                        }
 
200
                        catch( UnsupportedEncodingException e ) {
 
201
                                // we know US-ASCII is supported, so appease the compiler...
 
202
                                line = "";
 
203
                        }
209
204
 
210
205
                        if( vcard == null ) {
211
206
                                // look for vcard beginning
212
 
                                if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
213
 
                                        setProgress( _progress++ );
 
207
                                if( line.matches( "^BEGIN:VCARD" ) ) {
 
208
                                        setProgress( ++_progress );
214
209
                                        vcard = new Vcard();
215
 
                                        vcard_start_line = cli.getLineNumber();
216
210
                                }
217
211
                        }
218
212
                        else {
219
213
                                // look for vcard content or ending
220
 
                                if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
 
214
                                if( line.matches( "^END:VCARD" ) )
221
215
                                {
222
216
                                        // finalise the vcard/contact
223
217
                                        try {
224
218
                                                vcard.finaliseVcard();
225
 
 
226
 
                                                // pass the finalised contact to the importer
227
 
                                                importContact( vcard );
228
219
                                        }
229
220
                                        catch( Vcard.ParseException e ) {
230
 
                                                if( !showContinue(
231
 
                                                        getText( R.string.error_vcf_parse ).toString()
232
 
                                                        + fileName +
233
 
                                                        getText( R.string.error_vcf_parse_line ).toString()
234
 
                                                        + cli.getLineNumber() + ":\n" + e.getMessage() ) )
235
 
                                                {
236
 
                                                        finish( ACTION_ABORT );
237
 
                                                }
238
 
                                                else
239
 
                                                        skipContact();
240
 
                                        }
241
 
                                        catch( ContactData.ContactNotIdentifiableException e ) {
242
 
                                                if( !showContinue(
243
 
                                                        getText( R.string.error_vcf_parse ).toString()
244
 
                                                        + fileName +
245
 
                                                        getText( R.string.error_vcf_parse_line ).toString()
246
 
                                                        + vcard_start_line + ":\n" + getText(
247
 
                                                                R.string.error_vcf_notenoughinfo ).toString()
248
 
                                                ) )
249
 
                                                {
250
 
                                                        finish( ACTION_ABORT );
251
 
                                                }
252
 
                                                else
253
 
                                                        skipContact();
254
 
                                        }
255
 
 
256
 
                                        // discard this vcard
 
221
                                                skipContact();
 
222
                                                if( !showContinue(
 
223
                                                        getText( R.string.error_vcf_parse ).toString()
 
224
                                                        + fileName + "\n" + e.getMessage() ) )
 
225
                                                {
 
226
                                                        finish( ACTION_ABORT );
 
227
                                                }
 
228
                                        }
 
229
 
 
230
                                        // pass the finalised contact to the importer
 
231
                                        importContact( vcard );
 
232
 
 
233
                                        // and discard it
257
234
                                        vcard = null;
258
235
                                }
259
236
                                else
260
237
                                {
261
238
                                        // try giving the line to the vcard
262
239
                                        try {
263
 
                                                vcard.parseLine( content_line );
 
240
                                                vcard.parseLine( buffer, line,
 
241
                                                        cli.doesNextLineLookFolded() );
264
242
                                        }
265
243
                                        catch( Vcard.ParseException e ) {
266
244
                                                skipContact();
267
245
                                                if( !showContinue(
268
246
                                                        getText( R.string.error_vcf_parse ).toString()
269
 
                                                        + fileName +
270
 
                                                        getText( R.string.error_vcf_parse_line ).toString()
271
 
                                                        + cli.getLineNumber() + "\n" + e.getMessage() ) )
 
247
                                                        + fileName + "\n" + e.getMessage() ) )
272
248
                                                {
273
249
                                                        finish( ACTION_ABORT );
274
250
                                                }
275
251
 
276
 
                                                // Although we're continuing, we still need to abort
277
 
                                                // this vCard.  Further lines will be ignored until we
 
252
                                                // although we're continuing, we still need to abort
 
253
                                                // this vCard. Further lines will be ignored until we
278
254
                                                // get to another BEGIN:VCARD line.
279
255
                                                vcard = null;
280
256
                                        }
281
257
                                        catch( Vcard.SkipImportException e ) {
282
258
                                                skipContact();
283
 
                                                // Abort this vCard.  Further lines will be ignored until
 
259
                                                // abort this vCard. Further lines will be ignored until
284
260
                                                // we get to another BEGIN:VCARD line.
285
261
                                                vcard = null;
286
262
                                        }
289
265
                }
290
266
        }
291
267
 
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 >
 
268
        class ContentLineIterator implements Iterator< ByteBuffer >
335
269
        {
336
270
                protected byte[] _content = null;
337
271
                protected int _pos = 0;
338
 
                protected int _line = 0;
339
272
 
340
273
                public ContentLineIterator( byte[] content )
341
274
                {
349
282
                }
350
283
 
351
284
                @Override
352
 
                public ContentLine next()
 
285
                public ByteBuffer next()
353
286
                {
354
287
                        int initial_pos = _pos;
355
288
 
361
294
                                        int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
362
295
                                                _pos > initial_pos )? _pos - 1 : _pos;
363
296
                                        _pos++;
364
 
                                        _line++;
365
 
                                        return new ContentLine(
366
 
                                                ByteBuffer.wrap( _content, initial_pos,
367
 
                                                        to - initial_pos ),
368
 
                                                doesNextLineLookFolded() );
 
297
                                        return ByteBuffer.wrap( _content, initial_pos,
 
298
                                                to - initial_pos );
369
299
                                }
370
300
 
371
301
                        // we didn't find one, but were there bytes left?
372
302
                        if( _pos != initial_pos ) {
373
303
                                int to = _pos;
374
304
                                _pos++;
375
 
                                _line++;
376
 
                                return new ContentLine(
377
 
                                        ByteBuffer.wrap( _content, initial_pos,
378
 
                                                to - initial_pos ),
379
 
                                        doesNextLineLookFolded() );
 
305
                                return ByteBuffer.wrap( _content, initial_pos,
 
306
                                        to - initial_pos );
380
307
                        }
381
308
 
382
309
                        // no bytes left
394
321
                 * onto the end of this one?
395
322
                 * @return
396
323
                 */
397
 
                private boolean doesNextLineLookFolded()
 
324
                public boolean doesNextLineLookFolded()
398
325
                {
399
326
                        return _pos > 0 && _pos < _content.length &&
400
 
                                _content[ _pos - 1 ] == '\n' &&
401
 
                                ( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
402
 
                }
403
 
 
404
 
                public int getLineNumber()
405
 
                {
406
 
                        return _line;
 
327
                                _content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
407
328
                }
408
329
        }
409
330
 
410
331
        private class Vcard extends ContactData
411
332
        {
412
333
                private final static int NAMELEVEL_NONE = 0;
413
 
                private final static int NAMELEVEL_N = 1;
414
 
                private final static int NAMELEVEL_FN = 2;
 
334
                private final static int NAMELEVEL_FN = 1;
 
335
                private final static int NAMELEVEL_N = 2;
415
336
 
416
337
                private final static int MULTILINE_NONE = 0;
417
338
                private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
418
339
                private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
419
 
                private final static int MULTILINE_FOLDED = 3;  // MIME-DIR folding
 
340
                private final static int MULTILINE_FOLDED = 3;  // v3.0 folding
420
341
 
421
342
                private String _version = null;
422
 
                private Vector< ContentLine > _content_lines = null;
 
343
                private Vector< ByteBuffer > _buffers = null;
423
344
                private int _name_level = NAMELEVEL_NONE;
424
345
                private int _parser_multiline_state = MULTILINE_NONE;
425
346
                private String _parser_current_name_and_params = null;
468
389
                @SuppressWarnings("serial")
469
390
                protected class SkipImportException extends Exception { }
470
391
 
471
 
                private String extractCollonPartFromLine( ContentLine content_line,
472
 
                        boolean former )
 
392
                private String extractCollonPartFromLine( ByteBuffer buffer,
 
393
                        String line, boolean former )
473
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
 
474
410
                        // split line into name and value parts and check to make sure we
475
411
                        // only got 2 parts and that the first part is not zero in length
476
 
                        String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
 
412
                        String[] parts = line.split( ":", 2 );
477
413
                        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 )
 
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 )
494
432
                        throws ParseException, SkipImportException,
495
433
                        AbortImportException
496
434
                {
499
437
                        {
500
438
                                // tentatively get name and params from line
501
439
                                String name_and_params =
502
 
                                        extractNameAndParamsFromLine( content_line );
 
440
                                        extractNameAndParamsFromLine( buffer, line );
503
441
 
504
442
                                // is it a version line?
505
443
                                if( name_and_params != null &&
506
 
                                        name_and_params.equalsIgnoreCase( "VERSION" ) )
 
444
                                        name_and_params.equals( "VERSION" ) )
507
445
                                {
508
446
                                        // yes, get it!
509
 
                                        String value = extractValueFromLine( content_line );
510
 
                                        if( value == null || (
511
 
                                                !value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
512
 
                                        {
 
447
                                        String value = extractValueFromLine( buffer, line );
 
448
                                        if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
449
                                                throw new ParseException( R.string.error_vcf_version );
514
 
                                        }
515
450
                                        _version = value;
516
451
 
517
452
                                        // parse any buffers we've been accumulating while we waited
518
453
                                        // 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;
 
454
                                        if( _buffers != null )
 
455
                                                for( int i = 0; i < _buffers.size(); i++ )
 
456
                                                        parseLine( _buffers.get( i ), null,
 
457
                                                                i + 1 < _buffers.size() &&
 
458
                                                                _buffers.get( i + 1 ).hasRemaining() &&
 
459
                                                                _buffers.get( i + 1 ).get(
 
460
                                                                        _buffers.get( i + 1 ).position() ) == ' ' );
 
461
                                        _buffers = null;
523
462
                                }
524
463
                                else
525
464
                                {
526
465
                                        // 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 );
 
466
                                        if( _buffers == null )
 
467
                                                _buffers = new Vector< ByteBuffer >();
 
468
                                        _buffers.add( buffer );
530
469
                                }
531
470
                        }
532
471
                        else
533
472
                        {
534
473
                                // name and params and the position in the buffer where the
535
 
                                // "value" part of the line starts
 
474
                                // "value" part of the line start
536
475
                                String name_and_params;
537
476
                                int pos;
538
477
 
544
483
 
545
484
                                        // skip some initial line characters, depending on the type
546
485
                                        // of multi-line we're handling
547
 
                                        pos = content_line.getBuffer().position();
 
486
                                        pos = buffer.position();
548
487
                                        switch( _parser_multiline_state )
549
488
                                        {
550
489
                                        case MULTILINE_FOLDED:
551
490
                                                pos++;
552
491
                                                break;
553
492
                                        case MULTILINE_ENCODED:
554
 
                                                while( pos < content_line.getBuffer().limit() && (
555
 
                                                        content_line.getBuffer().get( pos ) == ' ' ||
556
 
                                                        content_line.getBuffer().get( pos ) == '\t' ) )
 
493
                                                while( pos < buffer.limit() && (
 
494
                                                        buffer.get( pos ) == ' ' ||
 
495
                                                        buffer.get( pos ) == '\t' ) )
557
496
                                                {
558
497
                                                        pos++;
559
498
                                                }
568
507
                                }
569
508
                                else
570
509
                                {
571
 
                                        // skip empty lines
572
 
                                        if( content_line.getUsAsciiLine().trim().length() == 0 )
573
 
                                                return;
574
 
 
575
510
                                        // get name and params from line, and since we're not
576
511
                                        // parsing a subsequent line in a multi-line, this should
577
512
                                        // not fail, or it's an error
578
513
                                        name_and_params =
579
 
                                                extractNameAndParamsFromLine( content_line );
 
514
                                                extractNameAndParamsFromLine( buffer, line );
580
515
                                        if( name_and_params == null )
581
516
                                                throw new ParseException(
582
517
                                                        R.string.error_vcf_malformed );
583
518
 
584
519
                                        // calculate how many chars to skip from beginning of line
585
520
                                        // so we skip the property "name:" part
586
 
                                        pos = content_line.getBuffer().position() +
587
 
                                                name_and_params.length() + 1;
 
521
                                        pos = buffer.position() + name_and_params.length() + 1;
588
522
 
589
523
                                        // reset the saved multi-line state
590
524
                                        _parser_current_name_and_params = name_and_params;
593
527
 
594
528
                                // get value from buffer, as raw bytes
595
529
                                ByteBuffer value;
596
 
                                value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
597
 
                                        content_line.getBuffer().limit() - pos );
 
530
                                value = ByteBuffer.wrap( buffer.array(), pos,
 
531
                                        buffer.limit() - pos );
598
532
 
599
533
                                // get parameter parts
600
534
                                String[] name_param_parts = name_and_params.split( ";", -1 );
601
535
                                for( int i = 0; i < name_param_parts.length; i++ )
602
536
                                        name_param_parts[ i ] = name_param_parts[ i ].trim();
603
537
 
604
 
                                // determine whether we care about this entry
605
 
                                final HashSet< String > interesting_fields =
606
 
                                        new HashSet< String >( Arrays.asList( new String[] { "N",
607
 
                                                "FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
608
 
                                ) );
609
 
                                boolean is_interesting_field =
610
 
                                        interesting_fields.contains(
611
 
                                                name_param_parts[ 0 ].toUpperCase( Locale.US ) );
612
 
 
613
538
                                // parse encoding parameter
614
539
                                String encoding = checkParam( name_param_parts, "ENCODING" );
615
 
                                if( encoding != null )
616
 
                                        encoding = encoding.toUpperCase( Locale.US );
617
 
                                if( is_interesting_field && encoding != null &&
618
 
                                        !encoding.equalsIgnoreCase( "8BIT" ) &&
619
 
                                        !encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
620
 
                                        //&& !encoding.equalsIgnoreCase( "BASE64" ) )
 
540
                                if( encoding != null ) encoding = encoding.toUpperCase();
 
541
                                if( encoding != null && !encoding.equals( "8BIT" ) &&
 
542
                                        !encoding.equals( "QUOTED-PRINTABLE" ) )
 
543
                                        //&& !encoding.equals( "BASE64" ) )
621
544
                                {
622
545
                                        throw new ParseException( R.string.error_vcf_encoding );
623
546
                                }
624
547
 
625
548
                                // parse charset parameter
626
549
                                String charset = checkParam( name_param_parts, "CHARSET" );
627
 
                                if( charset != null )
628
 
                                        charset = charset.toUpperCase( Locale.US );
629
 
                                if( charset != null &&
630
 
                                        !charset.equalsIgnoreCase( "US-ASCII" ) &&
631
 
                                        !charset.equalsIgnoreCase( "ASCII" ) &&
632
 
                                        !charset.equalsIgnoreCase( "UTF-8" ) )
 
550
                                if( charset != null ) charset = charset.toUpperCase();
 
551
                                if( charset != null && !charset.equals( "US-ASCII" ) &&
 
552
                                        !charset.equals( "ASCII" ) &&
 
553
                                        !charset.equals( "UTF-8" ) )
633
554
                                {
634
555
                                        throw new ParseException( R.string.error_vcf_charset );
635
556
                                }
637
558
                                // do unencoding (or default to a fake unencoding result with
638
559
                                // the raw string)
639
560
                                UnencodeResult unencoding_result = null;
640
 
                                if( encoding != null &&
641
 
                                        encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
642
 
                                {
 
561
                                if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
643
562
                                        unencoding_result = unencodeQuotedPrintable( value );
644
 
                                }
645
 
//                              else if( encoding != null &&
646
 
//                                      encoding.equalsIgnoreCase( "BASE64" ) )
647
 
//                              {
 
563
//                              else if( encoding != null && encoding.equals( "BASE64" ) )
648
564
//                                      unencoding_result = unencodeBase64( props[ 1 ], charset );
649
 
//                              }
650
565
                                if( unencoding_result != null ) {
651
566
                                        value = unencoding_result.getBuffer();
652
567
                                        if( unencoding_result.isAnotherLineRequired() )
653
568
                                                _parser_multiline_state = MULTILINE_ENCODED;
654
569
                                }
655
570
 
656
 
                                // convert 8-bit US-ASCII charset to UTF-8 (where no charset is
657
 
                                // specified for a v2.1 vcard entry, we assume it's US-ASCII)
658
 
                                if( ( charset == null && _version.equals( "2.1" ) ) ||
659
 
                                        ( charset != null && (
660
 
                                                charset.equalsIgnoreCase( "ASCII" ) ||
661
 
                                                charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
662
 
                                {
 
571
                                // convert 8-bit ASCII charset to US-ASCII
 
572
                                if( charset == null || charset.equals( "ASCII" ) ) {
663
573
                                        value = transcodeAsciiToUtf8( value );
 
574
                                        charset = "UTF-8";
664
575
                                }
665
576
 
666
 
                                // process charset (value is now in UTF-8)
 
577
                                // process charset
667
578
                                String string_value;
668
579
                                try {
669
580
                                        string_value = new String( value.array(), value.position(),
670
 
                                                value.limit() - value.position(), "UTF-8" );
 
581
                                                value.limit() - value.position(), charset );
671
582
                                } catch( UnsupportedEncodingException e ) {
672
583
                                        throw new ParseException( R.string.error_vcf_charset );
673
584
                                }
675
586
                                // for some entries that have semicolon-separated value parts,
676
587
                                // check to see if the value ends in an escape character, which
677
588
                                // indicates that we have a multi-line value
678
 
                                if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
679
 
                                        name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
680
 
                                        name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
 
589
                                if( ( name_param_parts[ 0 ].equals( "N" ) ||
 
590
                                        name_param_parts[ 0 ].equals( "ORG" ) ||
 
591
                                        name_param_parts[ 0 ].equals( "ADR" ) ) &&
681
592
                                        doesStringEndInAnEscapeChar( string_value ) )
682
593
                                {
683
594
                                        _parser_multiline_state = MULTILINE_ESCAPED;
685
596
                                                string_value.length() - 1 );
686
597
                                }
687
598
 
688
 
                                // if we know we're not in an encoding-based multi-line, check
689
 
                                // to see if we're in a folded multi-line
 
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
690
601
                                if( _parser_multiline_state == MULTILINE_NONE &&
691
 
                                        content_line.doesNextLineLookFolded() )
 
602
                                        _version.equals( "3.0" ) && next_line_looks_folded )
692
603
                                {
693
604
                                        _parser_multiline_state = MULTILINE_FOLDED;
694
605
                                }
706
617
                                if( complete_value.length() < 1 ) return;
707
618
 
708
619
                                // parse some properties
709
 
                                if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
 
620
                                if( name_param_parts[ 0 ].equals( "N" ) )
710
621
                                        parseN( name_param_parts, complete_value );
711
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
 
622
                                else if( name_param_parts[ 0 ].equals( "FN" ) )
712
623
                                        parseFN( name_param_parts, complete_value );
713
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
 
624
                                else if( name_param_parts[ 0 ].equals( "ORG" ) )
714
625
                                        parseORG( name_param_parts, complete_value );
715
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
 
626
                                else if( name_param_parts[ 0 ].equals( "TITLE" ) )
716
627
                                        parseTITLE( name_param_parts, complete_value );
717
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
 
628
                                else if( name_param_parts[ 0 ].equals( "TEL" ) )
718
629
                                        parseTEL( name_param_parts, complete_value );
719
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
 
630
                                else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
720
631
                                        parseEMAIL( name_param_parts, complete_value );
721
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
 
632
                                else if( name_param_parts[ 0 ].equals( "ADR" ) )
722
633
                                        parseADR( name_param_parts, complete_value );
723
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
724
 
                                        parseLABEL( name_param_parts, complete_value );
725
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
726
 
                                        parseNOTE( name_param_parts, complete_value );
727
 
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
728
 
                                        parseBDAY( name_param_parts, complete_value );
729
634
                        }
730
635
                }
731
636
 
744
649
                        return ( count & 1 ) == 1;
745
650
                }
746
651
 
747
 
                private String[] splitValueByCharacter( String value, char character )
 
652
                private String[] splitValueBySemicolon( String value )
748
653
                {
749
 
                        // split string in to parts by specified character
 
654
                        // split string in to parts by semicolon
750
655
                        ArrayList< String > parts = new ArrayList< String >(
751
 
                                Arrays.asList( value.split( "" + character ) ) );
 
656
                                Arrays.asList( value.split(  ";" ) ) );
752
657
 
753
658
                        // go through parts
754
659
                        for( int a = 0; a < parts.size(); a++ )
755
660
                        {
756
661
                                String str = parts.get( a );
757
662
 
758
 
                                // Look for parts that end in an escape character, but ignore
759
 
                                // the final part.  We've already detected escape chars at the
 
663
                                // look for parts that end in an escape character, but ignore
 
664
                                // the final part. We've already detected escape chars at the
760
665
                                // end of the final part in parseLine() and handled multi-lines
761
666
                                // accordingly.
762
667
                                if( a < parts.size() - 1 &&
763
668
                                        doesStringEndInAnEscapeChar( str ) )
764
669
                                {
765
 
                                        // append the escaped character, join the next part to this
766
 
                                        // part and remove the next part
 
670
                                        // join the next part to this part and remove the next part
767
671
                                        parts.set( a, str.substring( 0, str.length() - 1 ) +
768
 
                                                character + parts.get( a + 1 ) );
 
672
                                                ';' + parts.get( a + 1 ) );
769
673
                                        parts.remove( a + 1 );
770
674
 
771
675
                                        // re-visit this part
782
686
                        return parts.toArray( ret );
783
687
                }
784
688
 
785
 
                private String unescapeValue( String value )
786
 
                {
787
 
                        StringBuilder ret = new StringBuilder( value.length() );
788
 
                        boolean in_escape = false;
789
 
                        for( int a = 0; a < value.length(); a++ )
790
 
                        {
791
 
                                int c = value.codePointAt( a );
792
 
 
793
 
                                // process a normal character
794
 
                                if( !in_escape ) {
795
 
                                        if( c == '\\' )
796
 
                                                in_escape = true;
797
 
                                        else
798
 
                                                ret.append( Character.toChars( c ) );
799
 
                                        continue;
800
 
                                }
801
 
 
802
 
                                // process an escape sequence
803
 
                                in_escape = false;
804
 
                                switch( c )
805
 
                                {
806
 
                                case 'T':
807
 
                                case 't':
808
 
                                        // add tab (invalid/non-standard, but accepted)
809
 
                                        ret.append( '\t' );
810
 
                                        break;
811
 
                                case 'N':
812
 
                                case 'n':
813
 
                                        // add newline
814
 
                                        ret.append( '\n' );
815
 
                                        break;
816
 
                                case '\\':
817
 
                                case ',':
818
 
                                case ';':
819
 
                                        // add escaped character
820
 
                                        ret.append( Character.toChars( c ) );
821
 
                                        break;
822
 
                                default:
823
 
                                        // unknown escape sequence, so add it unescaped
824
 
                                        // (invalid/non-standard, but accepted)
825
 
                                        ret.append( "\\" );
826
 
                                        ret.append( Character.toChars( c ) );
827
 
                                        break;
828
 
                                }
829
 
                        }
830
 
 
831
 
                        return ret.toString();
832
 
                }
833
 
 
834
689
                private void parseN( String[] params, String value )
835
690
                {
836
691
                        // already got a better name?
837
692
                        if( _name_level >= NAMELEVEL_N ) return;
838
693
 
839
694
                        // get name parts
840
 
                        String[] name_parts = splitValueByCharacter( value, ';' );
 
695
                        String[] name_parts = splitValueBySemicolon( value );
841
696
 
842
697
                        // build name
843
698
                        value = "";
844
 
                        final int[] part_order = { 3, 1, 2, 0, 4 };
845
 
                        for( int a = 0; a < part_order.length; a++ )
846
 
                                if( name_parts.length > part_order[ a ] &&
847
 
                                        name_parts[ part_order[ a ] ].length() > 0 )
848
 
                                {
849
 
                                        // split this part in to it's comma-separated bits
850
 
                                        String[] name_part_parts = splitValueByCharacter(
851
 
                                                name_parts[ part_order[ a ] ], ',' );
852
 
                                        for( int b = 0; b < name_part_parts.length; b++ )
853
 
                                                if( name_part_parts[ b ].length() > 0 )
854
 
                                                {
855
 
                                                        if( value.length() > 0 ) value += " ";
856
 
                                                        value += name_part_parts[ b ];
857
 
                                                }
858
 
                                }
 
699
                        if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
 
700
                                value += name_parts[ 1 ];
 
701
                        if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
 
702
                                value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
859
703
 
860
704
                        // set name
861
 
                        setName( unescapeValue( value ) );
 
705
                        setName( value );
862
706
                        _name_level = NAMELEVEL_N;
863
707
                }
864
708
 
868
712
                        if( _name_level >= NAMELEVEL_FN ) return;
869
713
 
870
714
                        // set name
871
 
                        setName( unescapeValue( value ) );
 
715
                        setName( value );
872
716
                        _name_level = NAMELEVEL_FN;
873
717
                }
874
718
 
875
719
                private void parseORG( String[] params, String value )
876
720
                {
877
721
                        // get org parts
878
 
                        String[] org_parts = splitValueByCharacter( value, ';' );
 
722
                        String[] org_parts = splitValueBySemicolon( value );
879
723
                        if( org_parts == null || org_parts.length < 1 ) return;
880
724
 
881
725
                        // build organisation name
883
727
                                String.valueOf( org_parts[ 0 ] ) );
884
728
                        for( int a = 1; a < org_parts.length; a++ )
885
729
                                builder.append( ", " ).append( org_parts[ a ] );
886
 
                        String organisation = unescapeValue( builder.toString() );
 
730
                        String organisation = builder.toString();
887
731
 
888
732
                        // set organisation name (using a title we've previously found)
889
733
                        addOrganisation( organisation, _cached_title, true );
900
744
 
901
745
                private void parseTITLE( String[] params, String value )
902
746
                {
903
 
                        value = unescapeValue( value );
904
 
 
905
747
                        // if we previously had an organisation, look it up and append this
906
748
                        // title to it
907
749
                        if( _cached_organisation != null && hasOrganisations() ) {
933
775
                        int type;
934
776
                        if( types.contains( "FAX" ) )
935
777
                                if( types.contains( "HOME" ) )
936
 
                                        type = TYPE_FAX_HOME;
 
778
                                        type = PhonesColumns.TYPE_FAX_HOME;
937
779
                                else
938
 
                                        type = TYPE_FAX_WORK;
 
780
                                        type = PhonesColumns.TYPE_FAX_WORK;
939
781
                        else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
940
 
                                type = TYPE_MOBILE;
 
782
                                type = PhonesColumns.TYPE_MOBILE;
941
783
                        else if( types.contains( "PAGER" ) )
942
 
                                type = TYPE_PAGER;
 
784
                                type = PhonesColumns.TYPE_PAGER;
943
785
                        else if( types.contains( "WORK" ) )
944
 
                                type = TYPE_WORK;
 
786
                                type = PhonesColumns.TYPE_WORK;
945
787
                        else
946
 
                                type = TYPE_HOME;
 
788
                                type = PhonesColumns.TYPE_HOME;
947
789
 
948
790
                        // add phone number
949
791
                        addNumber( value, type, is_preferred );
960
802
                        boolean is_preferred = types.contains( "PREF" );
961
803
                        int type;
962
804
                        if( types.contains( "WORK" ) )
963
 
                                type = TYPE_WORK;
 
805
                                type = Contacts.ContactMethods.TYPE_WORK;
964
806
                        else
965
 
                                type = TYPE_HOME;
 
807
                                type = Contacts.ContactMethods.TYPE_HOME;
966
808
 
967
 
                        addEmail( unescapeValue( value ), type, is_preferred );
 
809
                        addEmail( value, type, is_preferred );
968
810
                }
969
811
 
970
812
                private void parseADR( String[] params, String value )
971
813
                {
972
814
                        // get address parts
973
 
                        String[] adr_parts = splitValueByCharacter( value, ';' );
 
815
                        String[] adr_parts = splitValueBySemicolon( value );
974
816
 
975
817
                        // build address
976
818
                        value = "";
977
 
                        for( int a = 0; a < adr_parts.length; a++ )
978
 
                                if( adr_parts[ a ].length() > 0 )
979
 
                                {
980
 
                                        // version 3.0 vCards allow further splitting by comma
981
 
                                        if( _version.equals( "3.0" ) )
982
 
                                        {
983
 
                                                // split this part in to it's comma-separated bits and
984
 
                                                // add them on individual lines
985
 
                                                String[] adr_part_parts =
986
 
                                                        splitValueByCharacter( adr_parts[ a ], ',' );
987
 
                                                for( int b = 0; b < adr_part_parts.length; b++ )
988
 
                                                        if( adr_part_parts[ b ].length() > 0 )
989
 
                                                        {
990
 
                                                                if( value.length() > 0 ) value += "\n";
991
 
                                                                value += adr_part_parts[ b ];
992
 
                                                        }
993
 
                                        }
994
 
                                        else
995
 
                                        {
996
 
                                                // add this part on an individual line
997
 
                                                if( value.length() > 0 ) value += "\n";
998
 
                                                value += adr_parts[ a ];
999
 
                                        }
1000
 
                                }
1001
 
 
1002
 
                        Set< String > types = extractTypes( params, Arrays.asList(
1003
 
                                "PREF", "WORK", "HOME" ) );
1004
 
 
1005
 
                        // add address
1006
 
                        int type;
1007
 
                        if( types.contains( "WORK" ) )
1008
 
                                type = TYPE_WORK;
1009
 
                        else
1010
 
                                type = TYPE_HOME;
1011
 
 
1012
 
                        addAddress( unescapeValue( value ), type );
1013
 
                }
1014
 
 
1015
 
                private void parseLABEL( String[] params, String value )
1016
 
                {
1017
 
                        Set< String > types = extractTypes( params, Arrays.asList(
1018
 
                                "PREF", "WORK", "HOME" ) );
1019
 
 
1020
 
                        // add address
1021
 
                        int type;
1022
 
                        if( types.contains( "WORK" ) )
1023
 
                                type = TYPE_WORK;
1024
 
                        else
1025
 
                                type = TYPE_HOME;
1026
 
 
1027
 
                        addAddress( unescapeValue( value ), type );
1028
 
                }
1029
 
 
1030
 
                private void parseNOTE( String[] params, String value )
1031
 
                {
1032
 
                        addNote( unescapeValue( value ) );
1033
 
                }
1034
 
 
1035
 
                private void parseBDAY( String[] params, String value )
1036
 
                {
1037
 
                        setBirthday( value );
 
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 );
1038
835
                }
1039
836
 
1040
837
                public void finaliseVcard()
1041
 
                        throws ParseException, ContactNotIdentifiableException
 
838
                        throws ParseException
1042
839
                {
1043
840
                        // missing version (and data is present)
1044
 
                        if( _version == null && _content_lines != null )
 
841
                        if( _version == null && _buffers != null )
1045
842
                                throw new ParseException( R.string.error_vcf_malformed );
1046
843
 
1047
844
                        // finalise the parent class
1048
 
                        finalise();
 
845
                        try {
 
846
                                finalise();
 
847
                        }
 
848
                        catch( ContactNotIdentifiableException e ) {
 
849
                                throw new ParseException( R.string.error_vcf_notenoughinfo );
 
850
                        }
1049
851
                }
1050
852
 
1051
 
                /**
1052
 
                 * Amongst the params, find the value of the first, only, of any with
1053
 
                 * the specified name.
1054
 
                 *
1055
 
                 * @param params
1056
 
                 * @param name
1057
 
                 * @return a value, or null
1058
 
                 */
1059
853
                private String checkParam( String[] params, String name )
1060
854
                {
1061
 
                        String[] res = checkParams( params, name );
1062
 
                        return res.length > 0? res[ 0 ] : null;
1063
 
                }
1064
 
 
1065
 
                /**
1066
 
                 * Amongst the params, find the values of any with the specified name.
1067
 
                 *
1068
 
                 * @param params
1069
 
                 * @param name
1070
 
                 * @return an array of values, or null
1071
 
                 */
1072
 
                private String[] checkParams( String[] params, String name )
1073
 
                {
1074
 
                        HashSet< String > ret = new HashSet< String >();
1075
 
 
1076
855
                        Pattern p = Pattern.compile(
1077
 
                                "^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1078
 
                                Pattern.CASE_INSENSITIVE );
 
856
                                "^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1079
857
                        for( int i = 0; i < params.length; i++ ) {
1080
858
                                Matcher m = p.matcher( params[ i ] );
1081
859
                                if( m.matches() )
1082
 
                                        ret.add( m.group( 2 ) );
 
860
                                        return m.group( 2 );
1083
861
                        }
1084
 
 
1085
 
                        return (String[]) ret.toArray( new String[ ret.size() ] );
 
862
                        return null;
1086
863
                }
1087
864
 
1088
 
                /**
1089
 
                 * Amongst the params, return any type values present.  For v2.1 vCards,
1090
 
                 * those types are just parameters.  For v3.0, they are prefixed with
1091
 
                 * "TYPE=".  There may also be multiple type parameters.
1092
 
                 *
1093
 
                 * @param params an array of params to look for types in
1094
 
                 * @param valid_types an list of upper-case type values to look for
1095
 
                 * @return a set of present type values
1096
 
                 */
1097
865
                private Set< String > extractTypes( String[] params,
1098
866
                        List< String > valid_types )
1099
867
                {
1100
868
                        HashSet< String > types = new HashSet< String >();
1101
869
 
1102
870
                        // get 3.0-style TYPE= param
1103
 
                        String type_params[] = checkParams( params, "TYPE" );
1104
 
                        for( int a = 0; a < type_params.length; a++ )
1105
 
                        {
1106
 
                                // check for a comma-separated list of types (why? I don't think
1107
 
                                // this is in the specs!)
1108
 
                                String[] parts = type_params[ a ].split( "," );
1109
 
                                for( int i = 0; i < parts.length; i++ ) {
1110
 
                                        String ucpart = parts[ i ].toUpperCase( Locale.US );
1111
 
                                        if( valid_types.contains( ucpart ) )
1112
 
                                                types.add( ucpart );
1113
 
                                }
 
871
                        String type_param;
 
872
                        if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
 
873
                                String[] parts = type_param.split( "," );
 
874
                                for( int i = 0; i < parts.length; i++ )
 
875
                                        if( valid_types.contains( parts[ i ] ) )
 
876
                                                types.add( parts[ i ] );
1114
877
                        }
1115
878
 
1116
879
                        // get 2.1-style type param
1117
880
                        if( _version.equals( "2.1" ) ) {
1118
 
                                for( int i = 1; i < params.length; i++ ) {
1119
 
                                        String ucparam = params[ i ].toUpperCase( Locale.US );
1120
 
                                        if( valid_types.contains( ucparam ) )
1121
 
                                                types.add( ucparam );
1122
 
                                }
 
881
                                for( int i = 1; i < params.length; i++ )
 
882
                                        if( valid_types.contains( params[ i ] ) )
 
883
                                                types.add( params[ i ] );
1123
884
                        }
1124
885
 
1125
886
                        return types;
1147
908
                                else if( ch == '=' && i == in.limit() - 1 )
1148
909
                                {
1149
910
                                        // we found a '=' at the end of a line signifying a multi-
1150
 
                                        // line string, so we don't add it
 
911
                                        // line string, so we don't add it.
1151
912
                                        another = true;
1152
913
                                        continue;
1153
914
                                }