/android/import-contacts

To get this branch, use:
bzr branch http://bzr.ed.am/android/import-contacts

« back to all changes in this revision

Viewing changes to src/am/ed/importcontacts/VcardImporter.java

  • Committer: Tim Marston
  • Date: 2013-05-10 12:05:10 UTC
  • Revision ID: tim@ed.am-20130510120510-xmjdrd7c7gje5shl
make checks for BEGIN:VCARD and END:VCARD case insensitive

Show diffs side-by-side

added added

removed removed

1
1
/*
2
2
 * VCFImporter.java
3
3
 *
4
 
 * Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
 
4
 * Copyright (C) 2009 to 2011 Tim Marston <tim@ed.am>
5
5
 *
6
6
 * This file is part of the Import Contacts program (hereafter referred
7
7
 * to as "this program"). For more information, see
8
 
 * http://www.waxworlds.org/edam/software/android/import-contacts
 
8
 * http://ed.am/dev/android/import-contacts
9
9
 *
10
10
 * This program is free software: you can redistribute it and/or modify
11
11
 * it under the terms of the GNU General Public License as published by
21
21
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
22
22
 */
23
23
 
24
 
package org.waxworlds.edam.importcontacts;
 
24
package am.ed.importcontacts;
25
25
 
26
26
import java.io.BufferedReader;
27
27
import java.io.File;
38
38
import java.util.HashSet;
39
39
import java.util.Iterator;
40
40
import java.util.List;
 
41
import java.util.Locale;
41
42
import java.util.NoSuchElementException;
42
43
import java.util.Set;
43
44
import java.util.Vector;
44
45
import java.util.regex.Matcher;
45
46
import java.util.regex.Pattern;
46
47
 
47
 
import org.waxworlds.edam.importcontacts.Importer.ContactData.ExtraDetail;
48
 
 
 
48
import android.annotation.SuppressLint;
49
49
import android.content.SharedPreferences;
50
 
import android.provider.Contacts;
51
 
import android.provider.Contacts.PhonesColumns;
52
50
 
53
 
public class VCFImporter extends Importer
 
51
public class VcardImporter extends Importer
54
52
{
55
 
        private int _vCardCount = 0;
 
53
        private int _vcard_count = 0;
56
54
        private int _progress = 0;
57
55
 
58
 
        public VCFImporter( Doit doit )
 
56
        public VcardImporter( Doit doit )
59
57
        {
60
58
                super( doit );
61
59
        }
62
60
 
 
61
        @SuppressLint( "SdCardPath" )
63
62
        @Override
64
63
        protected void onImport() throws AbortImportException
65
64
        {
84
83
                                // get files
85
84
                                class VCardFilter implements FilenameFilter {
86
85
                                        public boolean accept( File dir, String name ) {
87
 
                                                return name.toLowerCase().endsWith( ".vcf" );
 
86
                                                return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
88
87
                                        }
89
88
                                }
90
89
                                files = file.listFiles( new VCardFilter() );
112
111
                        countVCardFile( files[ i ] );
113
112
                        setTmpProgress( i );
114
113
                }
115
 
                setProgressMax( _vCardCount );  // will also update tmp progress
 
114
                setProgressMax( _vcard_count ); // will also update tmp progress
116
115
 
117
116
                // import them
118
117
                setProgress( 0 );
119
118
                for( int i = 0; i < files.length; i++ )
120
119
                        importVCardFile( files[ i ] );
 
120
                setProgress( _vcard_count );
121
121
        }
122
122
 
123
123
        private void countVCardFile( File file ) throws AbortImportException
130
130
 
131
131
                        // read
132
132
                        String line;
133
 
                        boolean inVCard = false;
 
133
                        boolean in_vcard = false;
134
134
                        while( ( line = reader.readLine() ) != null )
135
135
                        {
136
 
                                if( !inVCard ) {
 
136
                                if( !in_vcard )
 
137
                                {
137
138
                                        // look for vcard beginning
138
 
                                        if( line.matches( "^BEGIN:VCARD" ) ) {
139
 
                                                inVCard = true;
140
 
                                                _vCardCount++;
 
139
                                        if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
 
140
                                                in_vcard = true;
 
141
                                                _vcard_count++;
 
142
                                        }
 
143
                                        // check for vMsg files
 
144
                                        else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
 
145
                                                showError( getText( R.string.error_vcf_vmsgfile )
 
146
                                                        + file.getName() );
141
147
                                        }
142
148
                                }
143
 
                                else if( line.matches( "^END:VCARD" ) )
144
 
                                        inVCard = false;
 
149
                                else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
 
150
                                        in_vcard = false;
145
151
                        }
146
152
 
147
153
                }
170
176
                        FileInputStream istream = new FileInputStream( file );
171
177
                        byte[] content = new byte[ (int)file.length() ];
172
178
                        istream.read( content );
 
179
                        istream = null;
173
180
 
174
181
                        // import
175
182
                        importVCardFileContent( content, file.getName() );
176
183
                }
 
184
                catch( OutOfMemoryError e ) {
 
185
                        showError( R.string.error_outofmemory );
 
186
                }
177
187
                catch( FileNotFoundException e ) {
178
188
                        showError( getText( R.string.error_filenotfound ) +
179
189
                                file.getName() );
187
197
                throws AbortImportException
188
198
        {
189
199
                // go through lines
190
 
                VCard vCard = null;
 
200
                Vcard vcard = null;
 
201
                int vcard_start_line = 0;
191
202
                ContentLineIterator cli = new ContentLineIterator( content );
192
203
                while( cli.hasNext() )
193
204
                {
194
 
                        ByteBuffer buffer = cli.next();
195
 
 
196
 
                        // get a US-ASCII version of the line for processing
197
 
                        String line;
198
 
                        try {
199
 
                                line = new String( buffer.array(), buffer.position(),
200
 
                                        buffer.limit() - buffer.position(), "US-ASCII" );
201
 
                        }
202
 
                        catch( UnsupportedEncodingException e ) {
203
 
                                // we know US-ASCII is supported, so appease the compiler...
204
 
                                line = "";
205
 
                        }
206
 
 
207
 
                        if( vCard == null ) {
 
205
                        ContentLine content_line = cli.next();
 
206
 
 
207
                        // get a US-ASCII version of the string, for processing
 
208
                        String line = content_line.getUsAsciiLine();
 
209
 
 
210
                        if( vcard == null ) {
208
211
                                // look for vcard beginning
209
 
                                if( line.matches( "^BEGIN:VCARD" ) ) {
210
 
                                        setProgress( ++_progress );
211
 
                                        vCard = new VCard();
 
212
                                if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
 
213
                                        setProgress( _progress++ );
 
214
                                        vcard = new Vcard();
 
215
                                        vcard_start_line = cli.getLineNumber();
212
216
                                }
213
217
                        }
214
218
                        else {
215
219
                                // look for vcard content or ending
216
 
                                if( line.matches( "^END:VCARD" ) )
 
220
                                if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
217
221
                                {
218
 
                                        // store vcard and do away with it
 
222
                                        // finalise the vcard/contact
219
223
                                        try {
220
 
                                                vCard.finaliseParsing();
221
 
                                                importContact( vCard );
222
 
                                        }
223
 
                                        catch( VCard.ParseException e ) {
224
 
                                                skipContact();
225
 
                                                if( !showContinue(
226
 
                                                        getText( R.string.error_vcf_parse ).toString()
227
 
                                                        + fileName + "\n" + e.getMessage() ) )
228
 
                                                {
229
 
                                                        finish( ACTION_ABORT );
230
 
                                                }
231
 
                                        }
232
 
                                        catch( VCard.SkipContactException e ) {
233
 
                                                skipContact();
234
 
                                                // do nothing
235
 
                                        }
236
 
                                        vCard = null;
 
224
                                                vcard.finaliseVcard();
 
225
 
 
226
                                                // pass the finalised contact to the importer
 
227
                                                importContact( vcard );
 
228
                                        }
 
229
                                        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
 
257
                                        vcard = null;
237
258
                                }
238
259
                                else
239
260
                                {
240
261
                                        // try giving the line to the vcard
241
262
                                        try {
242
 
                                                vCard.parseLine( buffer, line,
243
 
                                                        cli.doesNextLineLookFolded() );
 
263
                                                vcard.parseLine( content_line );
244
264
                                        }
245
 
                                        catch( VCard.ParseException e ) {
 
265
                                        catch( Vcard.ParseException e ) {
246
266
                                                skipContact();
247
267
                                                if( !showContinue(
248
268
                                                        getText( R.string.error_vcf_parse ).toString()
249
 
                                                        + fileName + "\n" + e.getMessage() ) )
 
269
                                                        + fileName +
 
270
                                                        getText( R.string.error_vcf_parse_line ).toString()
 
271
                                                        + cli.getLineNumber() + "\n" + e.getMessage() ) )
250
272
                                                {
251
273
                                                        finish( ACTION_ABORT );
252
274
                                                }
254
276
                                                // although we're continuing, we still need to abort
255
277
                                                // this vCard. Further lines will be ignored until we
256
278
                                                // get to another BEGIN:VCARD line.
257
 
                                                vCard = null;
 
279
                                                vcard = null;
258
280
                                        }
259
 
                                        catch( VCard.SkipContactException e ) {
 
281
                                        catch( Vcard.SkipImportException e ) {
260
282
                                                skipContact();
261
283
                                                // abort this vCard. Further lines will be ignored until
262
284
                                                // we get to another BEGIN:VCARD line.
263
 
                                                vCard = null;
 
285
                                                vcard = null;
264
286
                                        }
265
287
                                }
266
288
                        }
267
289
                }
268
290
        }
269
291
 
270
 
        class ContentLineIterator implements Iterator< ByteBuffer >
 
292
        class ContentLine
 
293
        {
 
294
                private ByteBuffer _buffer;
 
295
                private boolean _folded_next;
 
296
                private String _line;
 
297
 
 
298
                public ContentLine( ByteBuffer buffer, boolean folded_next )
 
299
                {
 
300
                        _buffer = buffer;
 
301
                        _folded_next = folded_next;
 
302
                        _line = null;
 
303
                }
 
304
 
 
305
                public ByteBuffer getBuffer()
 
306
                {
 
307
                        return _buffer;
 
308
                }
 
309
 
 
310
                public boolean doesNextLineLookFolded()
 
311
                {
 
312
                        return _folded_next;
 
313
                }
 
314
 
 
315
                public String getUsAsciiLine()
 
316
                {
 
317
                        // generated line and cache it
 
318
                        if( _line == null ) {
 
319
                                try {
 
320
                                        _line = new String( _buffer.array(), _buffer.position(),
 
321
                                                _buffer.limit() - _buffer.position(), "US-ASCII" );
 
322
                                }
 
323
                                catch( UnsupportedEncodingException e ) {
 
324
                                        // we know US-ASCII *is* supported, so appease the
 
325
                                        // compiler...
 
326
                                }
 
327
                        }
 
328
 
 
329
                        // return cached line
 
330
                        return _line;
 
331
                }
 
332
        }
 
333
 
 
334
        class ContentLineIterator implements Iterator< ContentLine >
271
335
        {
272
336
                protected byte[] _content = null;
273
337
                protected int _pos = 0;
 
338
                protected int _line = 0;
274
339
 
275
340
                public ContentLineIterator( byte[] content )
276
341
                {
284
349
                }
285
350
 
286
351
                @Override
287
 
                public ByteBuffer next()
 
352
                public ContentLine next()
288
353
                {
289
354
                        int initial_pos = _pos;
290
355
 
296
361
                                        int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
297
362
                                                _pos > initial_pos )? _pos - 1 : _pos;
298
363
                                        _pos++;
299
 
                                        return ByteBuffer.wrap( _content, initial_pos,
300
 
                                                to - initial_pos );
 
364
                                        _line++;
 
365
                                        return new ContentLine(
 
366
                                                ByteBuffer.wrap( _content, initial_pos,
 
367
                                                        to - initial_pos ),
 
368
                                                doesNextLineLookFolded() );
301
369
                                }
302
370
 
303
371
                        // we didn't find one, but were there bytes left?
304
372
                        if( _pos != initial_pos ) {
305
373
                                int to = _pos;
306
374
                                _pos++;
307
 
                                return ByteBuffer.wrap( _content, initial_pos,
308
 
                                        to - initial_pos );
 
375
                                _line++;
 
376
                                return new ContentLine(
 
377
                                        ByteBuffer.wrap( _content, initial_pos,
 
378
                                                to - initial_pos ),
 
379
                                        doesNextLineLookFolded() );
309
380
                        }
310
381
 
311
382
                        // no bytes left
323
394
                 * onto the end of this one?
324
395
                 * @return
325
396
                 */
326
 
                public boolean doesNextLineLookFolded()
 
397
                private boolean doesNextLineLookFolded()
327
398
                {
328
399
                        return _pos > 0 && _pos < _content.length &&
329
 
                                _content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
 
400
                                _content[ _pos - 1 ] == '\n' &&
 
401
                                ( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
 
402
                }
 
403
 
 
404
                public int getLineNumber()
 
405
                {
 
406
                        return _line;
330
407
                }
331
408
        }
332
409
 
333
 
        private class VCard extends ContactData
 
410
        private class Vcard extends ContactData
334
411
        {
335
412
                private final static int NAMELEVEL_NONE = 0;
336
 
                private final static int NAMELEVEL_FN = 1;
337
 
                private final static int NAMELEVEL_N = 2;
 
413
                private final static int NAMELEVEL_N = 1;
 
414
                private final static int NAMELEVEL_FN = 2;
338
415
 
339
416
                private final static int MULTILINE_NONE = 0;
340
417
                private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
341
418
                private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
342
 
                private final static int MULTILINE_FOLDED = 3;  // v3.0 folding
 
419
                private final static int MULTILINE_FOLDED = 3;  // MIME-DIR folding
343
420
 
344
421
                private String _version = null;
345
 
                private Vector< ByteBuffer > _buffers = null;
 
422
                private Vector< ContentLine > _content_lines = null;
346
423
                private int _name_level = NAMELEVEL_NONE;
347
424
                private int _parser_multiline_state = MULTILINE_NONE;
348
425
                private String _parser_current_name_and_params = null;
384
461
 
385
462
                        public ParseException( int res )
386
463
                        {
387
 
                                super( VCFImporter.this.getText( res ).toString() );
 
464
                                super( VcardImporter.this.getText( res ).toString() );
388
465
                        }
389
466
                }
390
467
 
391
468
                @SuppressWarnings("serial")
392
 
                protected class SkipContactException extends Exception { }
 
469
                protected class SkipImportException extends Exception { }
393
470
 
394
 
                private String extractCollonPartFromLine( ByteBuffer buffer,
395
 
                        String line, boolean former )
 
471
                private String extractCollonPartFromLine( ContentLine content_line,
 
472
                        boolean former )
396
473
                {
397
474
                        String ret = null;
398
475
 
399
 
                        // get a US-ASCII version of the line for processing, unless we were
400
 
                        // supplied with one
401
 
                        if( line == null ) {
402
 
                                try {
403
 
                                        line = new String( buffer.array(), buffer.position(),
404
 
                                                buffer.limit() - buffer.position(), "US-ASCII" );
405
 
                                }
406
 
                                catch( UnsupportedEncodingException e ) {
407
 
                                        // we know US-ASCII is supported, so appease the compiler...
408
 
                                        line = "";
409
 
                                }
410
 
                        }
411
 
 
412
476
                        // split line into name and value parts and check to make sure we
413
477
                        // only got 2 parts and that the first part is not zero in length
414
 
                        String[] parts = line.split( ":", 2 );
 
478
                        String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
415
479
                        if( parts.length == 2 && parts[ 0 ].length() > 0 )
416
480
                                ret = parts[ former? 0 : 1 ];
417
481
 
418
482
                        return ret;
419
483
                }
420
484
 
421
 
                private String extractNameAndParamsFromLine( ByteBuffer buffer,
422
 
                        String line )
423
 
                {
424
 
                        return extractCollonPartFromLine( buffer, line, true );
425
 
                }
426
 
 
427
 
                private String extractValueFromLine( ByteBuffer buffer, String line )
428
 
                {
429
 
                        return extractCollonPartFromLine( buffer, line, false );
430
 
                }
431
 
 
432
 
                public void parseLine( ByteBuffer buffer, String line,
433
 
                        boolean next_line_looks_folded )
434
 
                        throws ParseException, SkipContactException,
 
485
                private String extractNameAndParamsFromLine( ContentLine content_line )
 
486
                {
 
487
                        return extractCollonPartFromLine( content_line, true ).trim();
 
488
                }
 
489
 
 
490
                private String extractValueFromLine( ContentLine content_line )
 
491
                {
 
492
                        return extractCollonPartFromLine( content_line, false );
 
493
                }
 
494
 
 
495
                public void parseLine( ContentLine content_line )
 
496
                        throws ParseException, SkipImportException,
435
497
                        AbortImportException
436
498
                {
437
499
                        // do we have a version yet?
439
501
                        {
440
502
                                // tentatively get name and params from line
441
503
                                String name_and_params =
442
 
                                        extractNameAndParamsFromLine( buffer, line );
 
504
                                        extractNameAndParamsFromLine( content_line );
443
505
 
444
506
                                // is it a version line?
445
507
                                if( name_and_params != null &&
446
 
                                        name_and_params.equals( "VERSION" ) )
 
508
                                        name_and_params.equalsIgnoreCase( "VERSION" ) )
447
509
                                {
448
510
                                        // yes, get it!
449
 
                                        String value = extractValueFromLine( buffer, line );
 
511
                                        String value = extractValueFromLine( content_line ).trim();
450
512
                                        if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
451
513
                                                throw new ParseException( R.string.error_vcf_version );
452
514
                                        _version = value;
453
515
 
454
516
                                        // parse any buffers we've been accumulating while we waited
455
517
                                        // for a version
456
 
                                        if( _buffers != null )
457
 
                                                for( int i = 0; i < _buffers.size(); i++ )
458
 
                                                        parseLine( _buffers.get( i ), null,
459
 
                                                                i + 1 < _buffers.size() &&
460
 
                                                                _buffers.get( i + 1 ).hasRemaining() &&
461
 
                                                                _buffers.get( i + 1 ).get(
462
 
                                                                        _buffers.get( i + 1 ).position() ) == ' ' );
463
 
                                        _buffers = null;
 
518
                                        if( _content_lines != null )
 
519
                                                for( int i = 0; i < _content_lines.size(); i++ )
 
520
                                                        parseLine( _content_lines.get( i ) );
 
521
                                        _content_lines = null;
464
522
                                }
465
523
                                else
466
524
                                {
467
525
                                        // no, so stash this line till we get a version
468
 
                                        if( _buffers == null )
469
 
                                                _buffers = new Vector< ByteBuffer >();
470
 
                                        _buffers.add( buffer );
 
526
                                        if( _content_lines == null )
 
527
                                                _content_lines = new Vector< ContentLine >();
 
528
                                        _content_lines.add( content_line );
471
529
                                }
472
530
                        }
473
531
                        else
474
532
                        {
475
533
                                // name and params and the position in the buffer where the
476
 
                                // "value" part of the line start
 
534
                                // "value" part of the line starts
477
535
                                String name_and_params;
478
536
                                int pos;
479
537
 
485
543
 
486
544
                                        // skip some initial line characters, depending on the type
487
545
                                        // of multi-line we're handling
488
 
                                        pos = buffer.position();
 
546
                                        pos = content_line.getBuffer().position();
489
547
                                        switch( _parser_multiline_state )
490
548
                                        {
491
549
                                        case MULTILINE_FOLDED:
492
550
                                                pos++;
493
551
                                                break;
494
552
                                        case MULTILINE_ENCODED:
495
 
                                                while( pos < buffer.limit() && (
496
 
                                                        buffer.get( pos ) == ' ' ||
497
 
                                                        buffer.get( pos ) == '\t' ) )
 
553
                                                while( pos < content_line.getBuffer().limit() && (
 
554
                                                        content_line.getBuffer().get( pos ) == ' ' ||
 
555
                                                        content_line.getBuffer().get( pos ) == '\t' ) )
498
556
                                                {
499
557
                                                        pos++;
500
558
                                                }
509
567
                                }
510
568
                                else
511
569
                                {
 
570
                                        // skip empty lines
 
571
                                        if( content_line.getUsAsciiLine().trim().length() == 0 )
 
572
                                                return;
 
573
 
512
574
                                        // get name and params from line, and since we're not
513
575
                                        // parsing a subsequent line in a multi-line, this should
514
576
                                        // not fail, or it's an error
515
577
                                        name_and_params =
516
 
                                                extractNameAndParamsFromLine( buffer, line );
 
578
                                                extractNameAndParamsFromLine( content_line );
517
579
                                        if( name_and_params == null )
518
580
                                                throw new ParseException(
519
581
                                                        R.string.error_vcf_malformed );
520
582
 
521
583
                                        // calculate how many chars to skip from beginning of line
522
584
                                        // so we skip the property "name:" part
523
 
                                        pos = buffer.position() + name_and_params.length() + 1;
 
585
                                        pos = content_line.getBuffer().position() +
 
586
                                                name_and_params.length() + 1;
524
587
 
525
588
                                        // reset the saved multi-line state
526
589
                                        _parser_current_name_and_params = name_and_params;
529
592
 
530
593
                                // get value from buffer, as raw bytes
531
594
                                ByteBuffer value;
532
 
                                value = ByteBuffer.wrap( buffer.array(), pos,
533
 
                                        buffer.limit() - pos );
 
595
                                value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
 
596
                                        content_line.getBuffer().limit() - pos );
534
597
 
535
598
                                // get parameter parts
536
599
                                String[] name_param_parts = name_and_params.split( ";", -1 );
537
600
                                for( int i = 0; i < name_param_parts.length; i++ )
538
601
                                        name_param_parts[ i ] = name_param_parts[ i ].trim();
539
602
 
 
603
                                // determine whether we care about this entry
 
604
                                final HashSet< String > interesting_fields =
 
605
                                        new HashSet< String >( Arrays.asList( new String[] { "N",
 
606
                                                "FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
 
607
                                ) );
 
608
                                boolean is_interesting_field =
 
609
                                        interesting_fields.contains(
 
610
                                                name_param_parts[ 0 ].toUpperCase( Locale.US ) );
 
611
 
540
612
                                // parse encoding parameter
541
613
                                String encoding = checkParam( name_param_parts, "ENCODING" );
542
 
                                if( encoding != null ) encoding = encoding.toUpperCase();
543
 
                                if( encoding != null && !encoding.equals( "8BIT" ) &&
544
 
                                        !encoding.equals( "QUOTED-PRINTABLE" ) )
545
 
                                        //&& !encoding.equals( "BASE64" ) )
 
614
                                if( encoding != null )
 
615
                                        encoding = encoding.toUpperCase( Locale.US );
 
616
                                if( is_interesting_field && encoding != null &&
 
617
                                        !encoding.equalsIgnoreCase( "8BIT" ) &&
 
618
                                        !encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
 
619
                                        //&& !encoding.equalsIgnoreCase( "BASE64" ) )
546
620
                                {
547
621
                                        throw new ParseException( R.string.error_vcf_encoding );
548
622
                                }
549
623
 
550
624
                                // parse charset parameter
551
625
                                String charset = checkParam( name_param_parts, "CHARSET" );
552
 
                                if( charset != null ) charset = charset.toUpperCase();
553
 
                                if( charset != null && !charset.equals( "US-ASCII" ) &&
554
 
                                        !charset.equals( "ASCII" ) &&
555
 
                                        !charset.equals( "UTF-8" ) )
 
626
                                if( charset != null )
 
627
                                        charset = charset.toUpperCase( Locale.US );
 
628
                                if( charset != null &&
 
629
                                        !charset.equalsIgnoreCase( "US-ASCII" ) &&
 
630
                                        !charset.equalsIgnoreCase( "ASCII" ) &&
 
631
                                        !charset.equalsIgnoreCase( "UTF-8" ) )
556
632
                                {
557
633
                                        throw new ParseException( R.string.error_vcf_charset );
558
634
                                }
560
636
                                // do unencoding (or default to a fake unencoding result with
561
637
                                // the raw string)
562
638
                                UnencodeResult unencoding_result = null;
563
 
                                if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
 
639
                                if( encoding != null &&
 
640
                                        encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
 
641
                                {
564
642
                                        unencoding_result = unencodeQuotedPrintable( value );
565
 
//                              else if( encoding != null && encoding.equals( "BASE64" ) )
 
643
                                }
 
644
//                              else if( encoding != null &&
 
645
//                                      encoding.equalsIgnoreCase( "BASE64" ) )
 
646
//                              {
566
647
//                                      unencoding_result = unencodeBase64( props[ 1 ], charset );
 
648
//                              }
567
649
                                if( unencoding_result != null ) {
568
650
                                        value = unencoding_result.getBuffer();
569
651
                                        if( unencoding_result.isAnotherLineRequired() )
570
652
                                                _parser_multiline_state = MULTILINE_ENCODED;
571
653
                                }
572
654
 
573
 
                                // convert 8-bit ASCII charset to US-ASCII
574
 
                                if( charset == null || charset.equals( "ASCII" ) ) {
 
655
                                // convert 8-bit US-ASCII charset to UTF-8 (where no charset is
 
656
                                // specified for a v2.1 vcard entry, we assume it's US-ASCII)
 
657
                                if( ( charset == null && _version.equals( "2.1" ) ) ||
 
658
                                        ( charset != null && (
 
659
                                                charset.equalsIgnoreCase( "ASCII" ) ||
 
660
                                                charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
 
661
                                {
575
662
                                        value = transcodeAsciiToUtf8( value );
576
 
                                        charset = "UTF-8";
577
663
                                }
578
664
 
579
 
                                // process charset
 
665
                                // process charset (value is now in UTF-8)
580
666
                                String string_value;
581
667
                                try {
582
668
                                        string_value = new String( value.array(), value.position(),
583
 
                                                value.limit() - value.position(), charset );
 
669
                                                value.limit() - value.position(), "UTF-8" );
584
670
                                } catch( UnsupportedEncodingException e ) {
585
671
                                        throw new ParseException( R.string.error_vcf_charset );
586
672
                                }
588
674
                                // for some entries that have semicolon-separated value parts,
589
675
                                // check to see if the value ends in an escape character, which
590
676
                                // indicates that we have a multi-line value
591
 
                                if( ( name_param_parts[ 0 ].equals( "N" ) ||
592
 
                                        name_param_parts[ 0 ].equals( "ORG" ) ||
593
 
                                        name_param_parts[ 0 ].equals( "ADR" ) ) &&
 
677
                                if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
 
678
                                        name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
 
679
                                        name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
594
680
                                        doesStringEndInAnEscapeChar( string_value ) )
595
681
                                {
596
682
                                        _parser_multiline_state = MULTILINE_ESCAPED;
598
684
                                                string_value.length() - 1 );
599
685
                                }
600
686
 
601
 
                                // now we know whether we're in an encoding multi-line,
602
 
                                // determine if we're in a v3 folded multi-line or not
 
687
                                // if we know we're not in an encoding-based multi-line, check
 
688
                                // to see if we're in a folded multi-line
603
689
                                if( _parser_multiline_state == MULTILINE_NONE &&
604
 
                                        _version.equals( "3.0" ) && next_line_looks_folded )
 
690
                                        content_line.doesNextLineLookFolded() )
605
691
                                {
606
692
                                        _parser_multiline_state = MULTILINE_FOLDED;
607
693
                                }
619
705
                                if( complete_value.length() < 1 ) return;
620
706
 
621
707
                                // parse some properties
622
 
                                if( name_param_parts[ 0 ].equals( "N" ) )
 
708
                                if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
623
709
                                        parseN( name_param_parts, complete_value );
624
 
                                else if( name_param_parts[ 0 ].equals( "FN" ) )
 
710
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
625
711
                                        parseFN( name_param_parts, complete_value );
626
 
                                else if( name_param_parts[ 0 ].equals( "ORG" ) )
 
712
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
627
713
                                        parseORG( name_param_parts, complete_value );
628
 
                                else if( name_param_parts[ 0 ].equals( "TITLE" ) )
 
714
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
629
715
                                        parseTITLE( name_param_parts, complete_value );
630
 
                                else if( name_param_parts[ 0 ].equals( "TEL" ) )
 
716
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
631
717
                                        parseTEL( name_param_parts, complete_value );
632
 
                                else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
 
718
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
633
719
                                        parseEMAIL( name_param_parts, complete_value );
634
 
                                else if( name_param_parts[ 0 ].equals( "ADR" ) )
 
720
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
635
721
                                        parseADR( name_param_parts, complete_value );
 
722
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
 
723
                                        parseLABEL( name_param_parts, complete_value );
 
724
                                else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
 
725
                                        parseNOTE( name_param_parts, complete_value );
636
726
                        }
637
727
                }
638
728
 
651
741
                        return ( count & 1 ) == 1;
652
742
                }
653
743
 
654
 
                private String[] splitValueBySemicolon( String value )
 
744
                private String[] splitValueByCharacter( String value, char character )
655
745
                {
656
 
                        // split string in to parts by semicolon
 
746
                        // split string in to parts by specified character
657
747
                        ArrayList< String > parts = new ArrayList< String >(
658
 
                                Arrays.asList( value.split(  ";" ) ) );
 
748
                                Arrays.asList( value.split( "" + character ) ) );
659
749
 
660
750
                        // go through parts
661
751
                        for( int a = 0; a < parts.size(); a++ )
669
759
                                if( a < parts.size() - 1 &&
670
760
                                        doesStringEndInAnEscapeChar( str ) )
671
761
                                {
672
 
                                        // join the next part to this part and remove the next part
 
762
                                        // append the escaped character, join the next part to this
 
763
                                        // part and remove the next part
673
764
                                        parts.set( a, str.substring( 0, str.length() - 1 ) +
674
 
                                                ';' + parts.get( a + 1 ) );
 
765
                                                character + parts.get( a + 1 ) );
675
766
                                        parts.remove( a + 1 );
676
767
 
677
768
                                        // re-visit this part
688
779
                        return parts.toArray( ret );
689
780
                }
690
781
 
 
782
                private String unescapeValue( String value )
 
783
                {
 
784
                        StringBuilder ret = new StringBuilder( value.length() );
 
785
                        boolean in_escape = false;
 
786
                        for( int a = 0; a < value.length(); a++ )
 
787
                        {
 
788
                                int c = value.codePointAt( a );
 
789
 
 
790
                                // process a normal character
 
791
                                if( !in_escape ) {
 
792
                                        if( c == '\\' )
 
793
                                                in_escape = true;
 
794
                                        else
 
795
                                                ret.append( Character.toChars( c ) );
 
796
                                        continue;
 
797
                                }
 
798
 
 
799
                                // process an escape sequence
 
800
                                in_escape = false;
 
801
                                switch( c )
 
802
                                {
 
803
                                case 'T':
 
804
                                case 't':
 
805
                                        // add tab (invalid/non-standard, but accepted)
 
806
                                        ret.append( '\t' );
 
807
                                        break;
 
808
                                case 'N':
 
809
                                case 'n':
 
810
                                        // add newline
 
811
                                        ret.append( '\n' );
 
812
                                        break;
 
813
                                case '\\':
 
814
                                case ',':
 
815
                                case ';':
 
816
                                        // add escaped character
 
817
                                        ret.append( Character.toChars( c ) );
 
818
                                        break;
 
819
                                default:
 
820
                                        // unknown escape sequence, so add it unescaped
 
821
                                        // (invalid/non-standard, but accepted)
 
822
                                        ret.append( "\\" );
 
823
                                        ret.append( Character.toChars( c ) );
 
824
                                        break;
 
825
                                }
 
826
                        }
 
827
 
 
828
                        return ret.toString();
 
829
                }
 
830
 
691
831
                private void parseN( String[] params, String value )
692
832
                {
693
833
                        // already got a better name?
694
834
                        if( _name_level >= NAMELEVEL_N ) return;
695
835
 
696
836
                        // get name parts
697
 
                        String[] name_parts = splitValueBySemicolon( value );
 
837
                        String[] name_parts = splitValueByCharacter( value, ';' );
698
838
 
699
839
                        // build name
700
840
                        value = "";
701
 
                        if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
702
 
                                value += name_parts[ 1 ];
703
 
                        if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
704
 
                                value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
 
841
                        final int[] part_order = { 3, 1, 2, 0, 4 };
 
842
                        for( int a = 0; a < part_order.length; a++ )
 
843
                                if( name_parts.length > part_order[ a ] &&
 
844
                                        name_parts[ part_order[ a ] ].length() > 0 )
 
845
                                {
 
846
                                        // split this part in to it's comma-separated bits
 
847
                                        String[] name_part_parts = splitValueByCharacter(
 
848
                                                name_parts[ part_order[ a ] ], ',' );
 
849
                                        for( int b = 0; b < name_part_parts.length; b++ )
 
850
                                                if( name_part_parts[ b ].length() > 0 )
 
851
                                                {
 
852
                                                        if( value.length() > 0 ) value += " ";
 
853
                                                        value += name_part_parts[ b ];
 
854
                                                }
 
855
                                }
705
856
 
706
857
                        // set name
707
 
                        setName( value );
 
858
                        setName( unescapeValue( value ) );
708
859
                        _name_level = NAMELEVEL_N;
709
860
                }
710
861
 
714
865
                        if( _name_level >= NAMELEVEL_FN ) return;
715
866
 
716
867
                        // set name
717
 
                        setName( value );
 
868
                        setName( unescapeValue( value ) );
718
869
                        _name_level = NAMELEVEL_FN;
719
870
                }
720
871
 
721
872
                private void parseORG( String[] params, String value )
722
873
                {
723
874
                        // get org parts
724
 
                        String[] org_parts = splitValueBySemicolon( value );
 
875
                        String[] org_parts = splitValueByCharacter( value, ';' );
725
876
                        if( org_parts == null || org_parts.length < 1 ) return;
726
877
 
727
878
                        // build organisation name
729
880
                                String.valueOf( org_parts[ 0 ] ) );
730
881
                        for( int a = 1; a < org_parts.length; a++ )
731
882
                                builder.append( ", " ).append( org_parts[ a ] );
732
 
                        String organisation = builder.toString();
 
883
                        String organisation = unescapeValue( builder.toString() );
733
884
 
734
885
                        // set organisation name (using a title we've previously found)
735
886
                        addOrganisation( organisation, _cached_title, true );
746
897
 
747
898
                private void parseTITLE( String[] params, String value )
748
899
                {
 
900
                        value = unescapeValue( value );
 
901
 
749
902
                        // if we previously had an organisation, look it up and append this
750
903
                        // title to it
751
904
                        if( _cached_organisation != null && hasOrganisations() ) {
773
926
                                "PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
774
927
 
775
928
                        // here's the logic...
776
 
                        boolean preferred = types.contains( "PREF" );
 
929
                        boolean is_preferred = types.contains( "PREF" );
777
930
                        int type;
778
931
                        if( types.contains( "FAX" ) )
779
932
                                if( types.contains( "HOME" ) )
780
 
                                        type = PhonesColumns.TYPE_FAX_HOME;
 
933
                                        type = TYPE_FAX_HOME;
781
934
                                else
782
 
                                        type = PhonesColumns.TYPE_FAX_WORK;
 
935
                                        type = TYPE_FAX_WORK;
783
936
                        else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
784
 
                                type = PhonesColumns.TYPE_MOBILE;
 
937
                                type = TYPE_MOBILE;
785
938
                        else if( types.contains( "PAGER" ) )
786
 
                                type = PhonesColumns.TYPE_PAGER;
 
939
                                type = TYPE_PAGER;
787
940
                        else if( types.contains( "WORK" ) )
788
 
                                type = PhonesColumns.TYPE_WORK;
 
941
                                type = TYPE_WORK;
789
942
                        else
790
 
                                type = PhonesColumns.TYPE_HOME;
 
943
                                type = TYPE_HOME;
791
944
 
792
945
                        // add phone number
793
 
                        addNumber( value, type, preferred );
 
946
                        addNumber( value, type, is_preferred );
794
947
                }
795
948
 
796
949
                public void parseEMAIL( String[] params, String value )
801
954
                                "PREF", "WORK", "HOME", "INTERNET" ) );
802
955
 
803
956
                        // add email address
804
 
                        boolean preferred = types.contains( "PREF" );
 
957
                        boolean is_preferred = types.contains( "PREF" );
805
958
                        int type;
806
959
                        if( types.contains( "WORK" ) )
807
 
                                type = Contacts.ContactMethods.TYPE_WORK;
 
960
                                type = TYPE_WORK;
808
961
                        else
809
 
                                type = Contacts.ContactMethods.TYPE_HOME;
 
962
                                type = TYPE_HOME;
810
963
 
811
 
                        addEmail( value, type, preferred );
 
964
                        addEmail( unescapeValue( value ), type, is_preferred );
812
965
                }
813
966
 
814
967
                private void parseADR( String[] params, String value )
815
968
                {
816
969
                        // get address parts
817
 
                        String[] adr_parts = splitValueBySemicolon( value );
 
970
                        String[] adr_parts = splitValueByCharacter( value, ';' );
818
971
 
819
972
                        // build address
820
973
                        value = "";
821
 
                        for( int a = 0; a < adr_parts.length; a++ ) {
822
 
                                if( value.length() > 0 ) value += "\n";
823
 
                                value += adr_parts[ a ].trim();
824
 
                        }
825
 
 
826
 
                        Set< String > types = extractTypes( params, Arrays.asList(
827
 
                                "PREF", "WORK", "HOME", "INTERNET" ) );
828
 
 
829
 
                        // add address
830
 
                        int type;
831
 
                        if( types.contains( "WORK" ) )
832
 
                                type = Contacts.ContactMethods.TYPE_WORK;
833
 
                        else
834
 
                                type = Contacts.ContactMethods.TYPE_HOME;
835
 
 
836
 
                        addAddress( value, type );
837
 
                }
838
 
 
839
 
                public void finaliseParsing()
840
 
                        throws ParseException, SkipContactException,
841
 
                        AbortImportException
 
974
                        for( int a = 0; a < adr_parts.length; a++ )
 
975
                                if( adr_parts[ a ].length() > 0 )
 
976
                                {
 
977
                                        // version 3.0 vCards allow further splitting by comma
 
978
                                        if( _version.equals( "3.0" ) )
 
979
                                        {
 
980
                                                // split this part in to it's comma-separated bits and
 
981
                                                // add them on individual lines
 
982
                                                String[] adr_part_parts =
 
983
                                                        splitValueByCharacter( adr_parts[ a ], ',' );
 
984
                                                for( int b = 0; b < adr_part_parts.length; b++ )
 
985
                                                        if( adr_part_parts[ b ].length() > 0 )
 
986
                                                        {
 
987
                                                                if( value.length() > 0 ) value += "\n";
 
988
                                                                value += adr_part_parts[ b ];
 
989
                                                        }
 
990
                                        }
 
991
                                        else
 
992
                                        {
 
993
                                                // add this part on an individual line
 
994
                                                if( value.length() > 0 ) value += "\n";
 
995
                                                value += adr_parts[ a ];
 
996
                                        }
 
997
                                }
 
998
 
 
999
                        Set< String > types = extractTypes( params, Arrays.asList(
 
1000
                                "PREF", "WORK", "HOME" ) );
 
1001
 
 
1002
                        // add address
 
1003
                        int type;
 
1004
                        if( types.contains( "WORK" ) )
 
1005
                                type = TYPE_WORK;
 
1006
                        else
 
1007
                                type = TYPE_HOME;
 
1008
 
 
1009
                        addAddress( unescapeValue( value ), type );
 
1010
                }
 
1011
 
 
1012
                private void parseLABEL( String[] params, String value )
 
1013
                {
 
1014
                        Set< String > types = extractTypes( params, Arrays.asList(
 
1015
                                "PREF", "WORK", "HOME" ) );
 
1016
 
 
1017
                        // add address
 
1018
                        int type;
 
1019
                        if( types.contains( "WORK" ) )
 
1020
                                type = TYPE_WORK;
 
1021
                        else
 
1022
                                type = TYPE_HOME;
 
1023
 
 
1024
                        addAddress( unescapeValue( value ), type );
 
1025
                }
 
1026
 
 
1027
                private void parseNOTE( String[] params, String value )
 
1028
                {
 
1029
                        addNote( unescapeValue( value ) );
 
1030
                }
 
1031
 
 
1032
                public void finaliseVcard()
 
1033
                        throws ParseException, ContactNotIdentifiableException
842
1034
                {
843
1035
                        // missing version (and data is present)
844
 
                        if( _version == null && _buffers != null )
 
1036
                        if( _version == null && _content_lines != null )
845
1037
                                throw new ParseException( R.string.error_vcf_malformed );
846
1038
 
847
 
                        // check if we should import this contact
848
 
                        try {
849
 
                                if( !isImportRequired( this ) )
850
 
                                        throw new SkipContactException();
851
 
                        }
852
 
                        catch( ContactNeedsMoreInfoException e ) {
853
 
                                throw new ParseException( R.string.error_vcf_notenoughinfo );
854
 
                        }
 
1039
                        // finalise the parent class
 
1040
                        finalise();
855
1041
                }
856
1042
 
 
1043
                /**
 
1044
                 * Amongst the params, find the value of the first, only, of any with
 
1045
                 * the specified name
 
1046
                 * @param params
 
1047
                 * @param name
 
1048
                 * @return a value, or null
 
1049
                 */
857
1050
                private String checkParam( String[] params, String name )
858
1051
                {
 
1052
                        String[] res = checkParams( params, name );
 
1053
                        return res.length > 0? res[ 0 ] : null;
 
1054
                }
 
1055
 
 
1056
                /**
 
1057
                 * Amongst the params, find the values of any with the specified name
 
1058
                 * @param params
 
1059
                 * @param name
 
1060
                 * @return an array of values, or null
 
1061
                 */
 
1062
                private String[] checkParams( String[] params, String name )
 
1063
                {
 
1064
                        HashSet< String > ret = new HashSet< String >();
 
1065
 
859
1066
                        Pattern p = Pattern.compile(
860
 
                                "^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
 
1067
                                "^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
 
1068
                                Pattern.CASE_INSENSITIVE );
861
1069
                        for( int i = 0; i < params.length; i++ ) {
862
1070
                                Matcher m = p.matcher( params[ i ] );
863
1071
                                if( m.matches() )
864
 
                                        return m.group( 2 );
 
1072
                                        ret.add( m.group( 2 ) );
865
1073
                        }
866
 
                        return null;
 
1074
 
 
1075
                        return (String[]) ret.toArray( new String[ ret.size() ] );
867
1076
                }
868
1077
 
 
1078
                /**
 
1079
                 * Amongst the params, return any type values present. For v2.1 vCards,
 
1080
                 * those types are just parameters. For v3.0, they are prefixed with
 
1081
                 * "TYPE=". There may also be multiple type parameters.
 
1082
                 * @param params an array of params to look for types in
 
1083
                 * @param valid_types an list of upper-case type values to look for
 
1084
                 * @return a set of present type values
 
1085
                 */
869
1086
                private Set< String > extractTypes( String[] params,
870
1087
                        List< String > valid_types )
871
1088
                {
872
1089
                        HashSet< String > types = new HashSet< String >();
873
1090
 
874
1091
                        // get 3.0-style TYPE= param
875
 
                        String type_param;
876
 
                        if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
877
 
                                String[] parts = type_param.split( "," );
878
 
                                for( int i = 0; i < parts.length; i++ )
879
 
                                        if( valid_types.contains( parts[ i ] ) )
880
 
                                                types.add( parts[ i ] );
 
1092
                        String type_params[] = checkParams( params, "TYPE" );
 
1093
                        for( int a = 0; a < type_params.length; a++ )
 
1094
                        {
 
1095
                                // check for a comma-separated list of types (why? I don't think
 
1096
                                // this is in the specs!)
 
1097
                                String[] parts = type_params[ a ].split( "," );
 
1098
                                for( int i = 0; i < parts.length; i++ ) {
 
1099
                                        String ucpart = parts[ i ].toUpperCase( Locale.US );
 
1100
                                        if( valid_types.contains( ucpart ) )
 
1101
                                                types.add( ucpart );
 
1102
                                }
881
1103
                        }
882
1104
 
883
1105
                        // get 2.1-style type param
884
1106
                        if( _version.equals( "2.1" ) ) {
885
 
                                for( int i = 1; i < params.length; i++ )
886
 
                                        if( valid_types.contains( params[ i ] ) )
887
 
                                                types.add( params[ i ] );
 
1107
                                for( int i = 1; i < params.length; i++ ) {
 
1108
                                        String ucparam = params[ i ].toUpperCase( Locale.US );
 
1109
                                        if( valid_types.contains( ucparam ) )
 
1110
                                                types.add( ucparam );
 
1111
                                }
888
1112
                        }
889
1113
 
890
1114
                        return types;