/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 Tim Marston <edam@waxworlds.org>
 
4
 * Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
5
5
 *
6
6
 * This file is part of the Import Contacts program (hereafter referred
7
7
 * to as "this program"). For more information, see
25
25
 
26
26
import java.io.BufferedReader;
27
27
import java.io.File;
 
28
import java.io.FileInputStream;
28
29
import java.io.FileNotFoundException;
29
30
import java.io.FileReader;
30
31
import java.io.FilenameFilter;
31
32
import java.io.IOException;
32
33
import java.io.UnsupportedEncodingException;
 
34
import java.nio.ByteBuffer;
 
35
import java.util.ArrayList;
33
36
import java.util.Arrays;
 
37
import java.util.HashMap;
34
38
import java.util.HashSet;
 
39
import java.util.Iterator;
35
40
import java.util.List;
 
41
import java.util.NoSuchElementException;
36
42
import java.util.Set;
37
43
import java.util.Vector;
38
44
import java.util.regex.Matcher;
42
48
import android.provider.Contacts;
43
49
import android.provider.Contacts.PhonesColumns;
44
50
 
45
 
public class VCFImporter extends Importer
 
51
public class VcardImporter extends Importer
46
52
{
47
 
        private int _vCardCount = 0;
 
53
        private int _vcard_count = 0;
48
54
        private int _progress = 0;
49
55
 
50
 
        public VCFImporter( Doit doit )
 
56
        public VcardImporter( Doit doit )
51
57
        {
52
58
                super( doit );
53
59
        }
104
110
                        countVCardFile( files[ i ] );
105
111
                        setTmpProgress( i );
106
112
                }
107
 
                setProgressMax( _vCardCount );  // will also update tmp progress
 
113
                setProgressMax( _vcard_count ); // will also update tmp progress
108
114
 
109
115
                // import them
110
116
                setProgress( 0 );
118
124
                {
119
125
                        // open file
120
126
                        BufferedReader reader = new BufferedReader(
121
 
                                        new FileReader( file ) );
 
127
                                new FileReader( file ) );
122
128
 
123
129
                        // read
124
130
                        String line;
125
 
                        boolean inVCard = false;
 
131
                        boolean in_vcard = false;
126
132
                        while( ( line = reader.readLine() ) != null )
127
133
                        {
128
 
                                if( !inVCard ) {
 
134
                                if( !in_vcard ) {
129
135
                                        // look for vcard beginning
130
 
                                        if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
131
 
                                                inVCard = true;
132
 
                                                _vCardCount++;
 
136
                                        if( line.matches( "^BEGIN:VCARD" ) ) {
 
137
                                                in_vcard = true;
 
138
                                                _vcard_count++;
133
139
                                        }
134
140
                                }
135
 
                                else if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
136
 
                                        inVCard = false;
 
141
                                else if( line.matches( "^END:VCARD" ) )
 
142
                                        in_vcard = false;
137
143
                        }
138
144
 
139
145
                }
148
154
 
149
155
        private void importVCardFile( File file ) throws AbortImportException
150
156
        {
 
157
                // check file is good
 
158
                if( !file.exists() )
 
159
                        showError( getText( R.string.error_filenotfound ) +
 
160
                                file.getName() );
 
161
                if( file.length() == 0 )
 
162
                        showError( getText( R.string.error_fileisempty ) +
 
163
                                file.getName() );
 
164
 
151
165
                try
152
166
                {
153
 
                        // open file
154
 
                        BufferedReader reader = new BufferedReader(
155
 
                                        new FileReader( file ) );
156
 
 
157
 
                        // read
158
 
                        StringBuffer content = new StringBuffer();
159
 
                        String line;
160
 
                        while( ( line = reader.readLine() ) != null )
161
 
                                content.append( line ).append( "\n" );
162
 
 
163
 
                        importVCardFileContent( content.toString(), file.getName() );
 
167
                        // open/read file
 
168
                        FileInputStream istream = new FileInputStream( file );
 
169
                        byte[] content = new byte[ (int)file.length() ];
 
170
                        istream.read( content );
 
171
 
 
172
                        // import
 
173
                        importVCardFileContent( content, file.getName() );
164
174
                }
165
175
                catch( FileNotFoundException e ) {
166
176
                        showError( getText( R.string.error_filenotfound ) +
171
181
                }
172
182
        }
173
183
 
174
 
        private void importVCardFileContent( String content, String fileName )
175
 
                        throws AbortImportException
 
184
        private void importVCardFileContent( byte[] content, String fileName )
 
185
                throws AbortImportException
176
186
        {
177
 
                // get lines and parse them
178
 
                String[] lines = content.split( "\n" );
179
 
                VCard vCard = null;
180
 
                for( int i = 0; i < lines.length; i++ )
 
187
                // go through lines
 
188
                Vcard vcard = null;
 
189
                ContentLineIterator cli = new ContentLineIterator( content );
 
190
                while( cli.hasNext() )
181
191
                {
182
 
                        String line = lines[ i ];
183
 
 
184
 
                        if( vCard == null ) {
 
192
                        ByteBuffer buffer = cli.next();
 
193
 
 
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
                        }
 
204
 
 
205
                        if( vcard == null ) {
185
206
                                // look for vcard beginning
186
 
                                if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
 
207
                                if( line.matches( "^BEGIN:VCARD" ) ) {
187
208
                                        setProgress( ++_progress );
188
 
                                        vCard = new VCard();
 
209
                                        vcard = new Vcard();
189
210
                                }
190
211
                        }
191
212
                        else {
192
213
                                // look for vcard content or ending
193
 
                                if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
 
214
                                if( line.matches( "^END:VCARD" ) )
194
215
                                {
195
 
                                        // store vcard and do away with it
 
216
                                        // finalise the vcard/contact
196
217
                                        try {
197
 
                                                vCard.finaliseParsing();
198
 
                                                importContact( vCard );
 
218
                                                vcard.finaliseVcard();
199
219
                                        }
200
 
                                        catch( VCard.ParseException e ) {
 
220
                                        catch( Vcard.ParseException e ) {
201
221
                                                skipContact();
202
222
                                                if( !showContinue(
203
 
                                                                getText( R.string.error_vcf_parse ).toString()
204
 
                                                                + fileName + "\n" + e.getMessage() ) )
 
223
                                                        getText( R.string.error_vcf_parse ).toString()
 
224
                                                        + fileName + "\n" + e.getMessage() ) )
 
225
                                                {
205
226
                                                        finish( ACTION_ABORT );
206
 
                                        }
207
 
                                        catch( VCard.SkipContactException e ) {
208
 
                                                skipContact();
209
 
                                                // do nothing
210
 
                                        }
211
 
                                        vCard = null;
 
227
                                                }
 
228
                                        }
 
229
 
 
230
                                        // pass the finalised contact to the importer
 
231
                                        importContact( vcard );
 
232
 
 
233
                                        // and discard it
 
234
                                        vcard = null;
212
235
                                }
213
236
                                else
214
237
                                {
215
238
                                        // try giving the line to the vcard
216
239
                                        try {
217
 
                                                vCard.parseLine( line );
 
240
                                                vcard.parseLine( buffer, line,
 
241
                                                        cli.doesNextLineLookFolded() );
218
242
                                        }
219
 
                                        catch( VCard.ParseException e ) {
 
243
                                        catch( Vcard.ParseException e ) {
220
244
                                                skipContact();
221
245
                                                if( !showContinue(
222
 
                                                                getText( R.string.error_vcf_parse ).toString()
223
 
                                                                + fileName + "\n" + e.getMessage() ) )
 
246
                                                        getText( R.string.error_vcf_parse ).toString()
 
247
                                                        + fileName + "\n" + e.getMessage() ) )
 
248
                                                {
224
249
                                                        finish( ACTION_ABORT );
 
250
                                                }
225
251
 
226
252
                                                // although we're continuing, we still need to abort
227
253
                                                // this vCard. Further lines will be ignored until we
228
254
                                                // get to another BEGIN:VCARD line.
229
 
                                                vCard = null;
 
255
                                                vcard = null;
230
256
                                        }
231
 
                                        catch( VCard.SkipContactException e ) {
 
257
                                        catch( Vcard.SkipImportException e ) {
232
258
                                                skipContact();
233
259
                                                // abort this vCard. Further lines will be ignored until
234
260
                                                // we get to another BEGIN:VCARD line.
235
 
                                                vCard = null;
 
261
                                                vcard = null;
236
262
                                        }
237
263
                                }
238
264
                        }
239
265
                }
240
266
        }
241
267
 
242
 
        private class VCard extends ContactData
 
268
        class ContentLineIterator implements Iterator< ByteBuffer >
 
269
        {
 
270
                protected byte[] _content = null;
 
271
                protected int _pos = 0;
 
272
 
 
273
                public ContentLineIterator( byte[] content )
 
274
                {
 
275
                        _content = content;
 
276
                }
 
277
 
 
278
                @Override
 
279
                public boolean hasNext()
 
280
                {
 
281
                        return _pos < _content.length;
 
282
                }
 
283
 
 
284
                @Override
 
285
                public ByteBuffer next()
 
286
                {
 
287
                        int initial_pos = _pos;
 
288
 
 
289
                        // find newline
 
290
                        for( ; _pos < _content.length; _pos++ )
 
291
                                if( _content[ _pos ] == '\n' )
 
292
                                {
 
293
                                        // adjust for a \r preceding the \n
 
294
                                        int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
 
295
                                                _pos > initial_pos )? _pos - 1 : _pos;
 
296
                                        _pos++;
 
297
                                        return ByteBuffer.wrap( _content, initial_pos,
 
298
                                                to - initial_pos );
 
299
                                }
 
300
 
 
301
                        // we didn't find one, but were there bytes left?
 
302
                        if( _pos != initial_pos ) {
 
303
                                int to = _pos;
 
304
                                _pos++;
 
305
                                return ByteBuffer.wrap( _content, initial_pos,
 
306
                                        to - initial_pos );
 
307
                        }
 
308
 
 
309
                        // no bytes left
 
310
                        throw new NoSuchElementException();
 
311
                }
 
312
 
 
313
                @Override
 
314
                public void remove()
 
315
                {
 
316
                        throw new UnsupportedOperationException();
 
317
                }
 
318
 
 
319
                /**
 
320
                 * Does the next line, if there is one, look like it should be folded
 
321
                 * onto the end of this one?
 
322
                 * @return
 
323
                 */
 
324
                public boolean doesNextLineLookFolded()
 
325
                {
 
326
                        return _pos > 0 && _pos < _content.length &&
 
327
                                _content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
 
328
                }
 
329
        }
 
330
 
 
331
        private class Vcard extends ContactData
243
332
        {
244
333
                private final static int NAMELEVEL_NONE = 0;
245
 
                private final static int NAMELEVEL_ORG = 1;
246
 
                private final static int NAMELEVEL_FN = 2;
247
 
                private final static int NAMELEVEL_N = 3;
 
334
                private final static int NAMELEVEL_FN = 1;
 
335
                private final static int NAMELEVEL_N = 2;
 
336
 
 
337
                private final static int MULTILINE_NONE = 0;
 
338
                private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
 
339
                private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
 
340
                private final static int MULTILINE_FOLDED = 3;  // v3.0 folding
248
341
 
249
342
                private String _version = null;
250
 
                private Vector< String > _lines = null;
 
343
                private Vector< ByteBuffer > _buffers = null;
251
344
                private int _name_level = NAMELEVEL_NONE;
252
 
                private boolean _parser_in_multiline = false;
 
345
                private int _parser_multiline_state = MULTILINE_NONE;
253
346
                private String _parser_current_name_and_params = null;
254
347
                private String _parser_buffered_value_so_far = "";
 
348
                private String _cached_organisation = null;
 
349
                private String _cached_title = null;
255
350
 
256
351
                protected class UnencodeResult
257
352
                {
258
353
                        private boolean _another_line_required;
259
 
                        private byte[] _bytes;
260
 
                        private int _num_bytes;
 
354
                        private ByteBuffer _buffer;
261
355
 
262
 
                        public UnencodeResult( boolean another_line_required, byte[] bytes,
263
 
                                int num_bytes )
 
356
                        public UnencodeResult( boolean another_line_required,
 
357
                                ByteBuffer buffer )
264
358
                        {
265
359
                                _another_line_required = another_line_required;
266
 
                                _bytes = bytes;
267
 
                                _num_bytes = num_bytes;
 
360
                                _buffer = buffer;
268
361
                        }
269
362
 
270
363
                        public boolean isAnotherLineRequired()
272
365
                                return _another_line_required;
273
366
                        }
274
367
 
275
 
                        public byte[] getBytes()
276
 
                        {
277
 
                                return _bytes;
278
 
                        }
279
 
 
280
 
                        public int getNumBytes()
281
 
                        {
282
 
                                return _num_bytes;
 
368
                        public ByteBuffer getBuffer()
 
369
                        {
 
370
                                return _buffer;
283
371
                        }
284
372
                }
285
373
 
294
382
 
295
383
                        public ParseException( int res )
296
384
                        {
297
 
                                super( VCFImporter.this.getText( res ).toString() );
 
385
                                super( VcardImporter.this.getText( res ).toString() );
298
386
                        }
299
387
                }
300
388
 
301
389
                @SuppressWarnings("serial")
302
 
                protected class SkipContactException extends Exception { }
303
 
 
304
 
                public void parseLine( String line )
305
 
                                throws ParseException, SkipContactException,
306
 
                                AbortImportException
307
 
                {
308
 
                        // ignore empty lines
309
 
                        if( line.trim() == "" ) return;
310
 
 
311
 
                        // split line into name and value parts (this may turn out to be
312
 
                        // unwanted if the line is a subsequent line in a multi-line
313
 
                        // value, but we have to do this now to check for and handle VCF
314
 
                        // versions first)
315
 
                        String[] props = line.split(  ":", 2 );
316
 
                        for( int i = 0; i < props.length; i++ )
317
 
                                props[ i ] = props[ i ].trim();
318
 
 
319
 
                        // if we haven't yet got a version, we won't be paring anything!
 
390
                protected class SkipImportException extends Exception { }
 
391
 
 
392
                private String extractCollonPartFromLine( ByteBuffer buffer,
 
393
                        String line, boolean former )
 
394
                {
 
395
                        String ret = null;
 
396
 
 
397
                        // get a US-ASCII version of the line for processing, unless we were
 
398
                        // supplied with one
 
399
                        if( line == null ) {
 
400
                                try {
 
401
                                        line = new String( buffer.array(), buffer.position(),
 
402
                                                buffer.limit() - buffer.position(), "US-ASCII" );
 
403
                                }
 
404
                                catch( UnsupportedEncodingException e ) {
 
405
                                        // we know US-ASCII is supported, so appease the compiler...
 
406
                                        line = "";
 
407
                                }
 
408
                        }
 
409
 
 
410
                        // split line into name and value parts and check to make sure we
 
411
                        // only got 2 parts and that the first part is not zero in length
 
412
                        String[] parts = line.split( ":", 2 );
 
413
                        if( parts.length == 2 && parts[ 0 ].length() > 0 )
 
414
                                ret = parts[ former? 0 : 1 ];
 
415
 
 
416
                        return ret;
 
417
                }
 
418
 
 
419
                private String extractNameAndParamsFromLine( ByteBuffer buffer,
 
420
                        String line )
 
421
                {
 
422
                        return extractCollonPartFromLine( buffer, line, true );
 
423
                }
 
424
 
 
425
                private String extractValueFromLine( ByteBuffer buffer, String line )
 
426
                {
 
427
                        return extractCollonPartFromLine( buffer, line, false );
 
428
                }
 
429
 
 
430
                public void parseLine( ByteBuffer buffer, String line,
 
431
                        boolean next_line_looks_folded )
 
432
                        throws ParseException, SkipImportException,
 
433
                        AbortImportException
 
434
                {
 
435
                        // do we have a version yet?
320
436
                        if( _version == null )
321
437
                        {
322
 
                                // is this a version?
323
 
                                if( props.length == 2 && props[ 0 ].equals( "VERSION" ) )
 
438
                                // tentatively get name and params from line
 
439
                                String name_and_params =
 
440
                                        extractNameAndParamsFromLine( buffer, line );
 
441
 
 
442
                                // is it a version line?
 
443
                                if( name_and_params != null &&
 
444
                                        name_and_params.equals( "VERSION" ) )
324
445
                                {
325
 
                                        // yes, check/store it
326
 
                                        if( !props[ 1 ].equals( "2.1" ) &&
327
 
                                                        !props[ 1 ].equals( "3.0" ) )
 
446
                                        // yes, get it!
 
447
                                        String value = extractValueFromLine( buffer, line );
 
448
                                        if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
328
449
                                                throw new ParseException( R.string.error_vcf_version );
329
 
                                        _version = props[ 1 ];
 
450
                                        _version = value;
330
451
 
331
 
                                        // parse any other lines we've accumulated so far
332
 
                                        if( _lines != null )
333
 
                                                for( int i = 0; i < _lines.size(); i++ )
334
 
                                                        parseLine( _lines.get( i ) );
335
 
                                        _lines = null;
 
452
                                        // parse any buffers we've been accumulating while we waited
 
453
                                        // for a version
 
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;
336
462
                                }
337
463
                                else
338
464
                                {
339
 
                                        // no, so stash this line till we have a version
340
 
                                        if( _lines == null )
341
 
                                                _lines = new Vector< String >();
342
 
                                        _lines.add( line );
 
465
                                        // no, so stash this line till we get a version
 
466
                                        if( _buffers == null )
 
467
                                                _buffers = new Vector< ByteBuffer >();
 
468
                                        _buffers.add( buffer );
343
469
                                }
344
470
                        }
345
471
                        else
346
472
                        {
347
 
                                if( _parser_in_multiline )
 
473
                                // name and params and the position in the buffer where the
 
474
                                // "value" part of the line start
 
475
                                String name_and_params;
 
476
                                int pos;
 
477
 
 
478
                                if( _parser_multiline_state != MULTILINE_NONE )
348
479
                                {
349
480
                                        // if we're currently in a multi-line value, use the stored
350
481
                                        // property name and parameters
351
 
                                        props = new String[ 2 ];
352
 
                                        props[ 0 ] = _parser_current_name_and_params;
353
 
                                        props[ 1 ] = line.trim();
 
482
                                        name_and_params = _parser_current_name_and_params;
 
483
 
 
484
                                        // skip some initial line characters, depending on the type
 
485
                                        // of multi-line we're handling
 
486
                                        pos = buffer.position();
 
487
                                        switch( _parser_multiline_state )
 
488
                                        {
 
489
                                        case MULTILINE_FOLDED:
 
490
                                                pos++;
 
491
                                                break;
 
492
                                        case MULTILINE_ENCODED:
 
493
                                                while( pos < buffer.limit() && (
 
494
                                                        buffer.get( pos ) == ' ' ||
 
495
                                                        buffer.get( pos ) == '\t' ) )
 
496
                                                {
 
497
                                                        pos++;
 
498
                                                }
 
499
                                                break;
 
500
                                        default:
 
501
                                                // do nothing
 
502
                                        }
 
503
 
 
504
                                        // take us out of multi-line so that we can re-detect that
 
505
                                        // this line is a multi-line or not
 
506
                                        _parser_multiline_state = MULTILINE_NONE;
354
507
                                }
355
508
                                else
356
509
                                {
357
 
                                        // for normal lines, check the property name/value bits
358
 
                                        if( props.length < 2 || props[ 0 ].length() == 0 )
 
510
                                        // get name and params from line, and since we're not
 
511
                                        // parsing a subsequent line in a multi-line, this should
 
512
                                        // not fail, or it's an error
 
513
                                        name_and_params =
 
514
                                                extractNameAndParamsFromLine( buffer, line );
 
515
                                        if( name_and_params == null )
359
516
                                                throw new ParseException(
360
517
                                                        R.string.error_vcf_malformed );
361
518
 
362
 
                                        // ignore empty properties
363
 
                                        if( props[ 1 ].length() < 1 )
364
 
                                                return;
 
519
                                        // calculate how many chars to skip from beginning of line
 
520
                                        // so we skip the property "name:" part
 
521
                                        pos = buffer.position() + name_and_params.length() + 1;
365
522
 
366
523
                                        // reset the saved multi-line state
367
 
                                        _parser_current_name_and_params = props[ 0 ];
 
524
                                        _parser_current_name_and_params = name_and_params;
368
525
                                        _parser_buffered_value_so_far = "";
369
526
                                }
370
527
 
 
528
                                // get value from buffer, as raw bytes
 
529
                                ByteBuffer value;
 
530
                                value = ByteBuffer.wrap( buffer.array(), pos,
 
531
                                        buffer.limit() - pos );
 
532
 
371
533
                                // get parameter parts
372
 
                                String[] params = props[ 0 ].split( ";" );
373
 
                                for( int i = 0; i < params.length; i++ )
374
 
                                        params[ i ] = params[ i ].trim();
 
534
                                String[] name_param_parts = name_and_params.split( ";", -1 );
 
535
                                for( int i = 0; i < name_param_parts.length; i++ )
 
536
                                        name_param_parts[ i ] = name_param_parts[ i ].trim();
375
537
 
376
 
                                // parse charset and encoding parameters
377
 
                                String charset, encoding;
378
 
                                if( ( charset = checkParam( params, "CHARSET" ) ) != null &&
379
 
                                        !charset.equals( "UTF-8" ) && !charset.equals( "UTF-16" ) )
380
 
                                {
381
 
                                        throw new ParseException( R.string.error_vcf_charset );
382
 
                                }
383
 
                                if( ( encoding = checkParam( params, "ENCODING" ) ) != null &&
384
 
                                        !encoding.equals( "QUOTED-PRINTABLE" ) &&
385
 
                                        !encoding.equals( "8BIT" ) )
 
538
                                // parse encoding parameter
 
539
                                String encoding = checkParam( name_param_parts, "ENCODING" );
 
540
                                if( encoding != null ) encoding = encoding.toUpperCase();
 
541
                                if( encoding != null && !encoding.equals( "8BIT" ) &&
 
542
                                        !encoding.equals( "QUOTED-PRINTABLE" ) )
386
543
                                        //&& !encoding.equals( "BASE64" ) )
387
544
                                {
388
545
                                        throw new ParseException( R.string.error_vcf_encoding );
389
546
                                }
390
547
 
 
548
                                // parse charset parameter
 
549
                                String charset = checkParam( name_param_parts, "CHARSET" );
 
550
                                if( charset != null ) charset = charset.toUpperCase();
 
551
                                if( charset != null && !charset.equals( "US-ASCII" ) &&
 
552
                                        !charset.equals( "ASCII" ) &&
 
553
                                        !charset.equals( "UTF-8" ) )
 
554
                                {
 
555
                                        throw new ParseException( R.string.error_vcf_charset );
 
556
                                }
 
557
 
391
558
                                // do unencoding (or default to a fake unencoding result with
392
559
                                // the raw string)
393
 
                                UnencodeResult result;
 
560
                                UnencodeResult unencoding_result = null;
394
561
                                if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
395
 
                                        result = unencodeQuotedPrintable( props[ 1 ], charset );
 
562
                                        unencoding_result = unencodeQuotedPrintable( value );
396
563
//                              else if( encoding != null && encoding.equals( "BASE64" ) )
397
 
//                                      result = unencodeBase64( props[ 1 ], charset );
398
 
                                else
399
 
                                        result = new UnencodeResult( false, props[ 1 ].getBytes(),
400
 
                                                props[ 1 ].getBytes().length );
 
564
//                                      unencoding_result = unencodeBase64( props[ 1 ], charset );
 
565
                                if( unencoding_result != null ) {
 
566
                                        value = unencoding_result.getBuffer();
 
567
                                        if( unencoding_result.isAnotherLineRequired() )
 
568
                                                _parser_multiline_state = MULTILINE_ENCODED;
 
569
                                }
 
570
 
 
571
                                // convert 8-bit ASCII charset to US-ASCII
 
572
                                if( charset == null || charset.equals( "ASCII" ) ) {
 
573
                                        value = transcodeAsciiToUtf8( value );
 
574
                                        charset = "UTF-8";
 
575
                                }
401
576
 
402
577
                                // process charset
 
578
                                String string_value;
403
579
                                try {
404
 
                                        props[ 1 ] = new String( result.getBytes(), 0,
405
 
                                                result.getNumBytes(),
406
 
                                                charset == null? "UTF-8" : charset );
 
580
                                        string_value = new String( value.array(), value.position(),
 
581
                                                value.limit() - value.position(), charset );
407
582
                                } catch( UnsupportedEncodingException e ) {
408
583
                                        throw new ParseException( R.string.error_vcf_charset );
409
584
                                }
410
585
 
411
 
                                // handle multi-line requests
412
 
                                _parser_in_multiline = result.isAnotherLineRequired();
413
 
                                if( _parser_in_multiline ) {
414
 
                                        _parser_buffered_value_so_far += props[ 1 ];
 
586
                                // for some entries that have semicolon-separated value parts,
 
587
                                // check to see if the value ends in an escape character, which
 
588
                                // indicates that we have a multi-line value
 
589
                                if( ( name_param_parts[ 0 ].equals( "N" ) ||
 
590
                                        name_param_parts[ 0 ].equals( "ORG" ) ||
 
591
                                        name_param_parts[ 0 ].equals( "ADR" ) ) &&
 
592
                                        doesStringEndInAnEscapeChar( string_value ) )
 
593
                                {
 
594
                                        _parser_multiline_state = MULTILINE_ESCAPED;
 
595
                                        string_value = string_value.substring( 0,
 
596
                                                string_value.length() - 1 );
 
597
                                }
 
598
 
 
599
                                // now we know whether we're in an encoding multi-line,
 
600
                                // determine if we're in a v3 folded multi-line or not
 
601
                                if( _parser_multiline_state == MULTILINE_NONE &&
 
602
                                        _version.equals( "3.0" ) && next_line_looks_folded )
 
603
                                {
 
604
                                        _parser_multiline_state = MULTILINE_FOLDED;
 
605
                                }
 
606
 
 
607
                                // handle multi-lines by buffering them and parsing them when we
 
608
                                // are processing the last line in a multi-line sequence
 
609
                                if( _parser_multiline_state != MULTILINE_NONE ) {
 
610
                                        _parser_buffered_value_so_far += string_value;
415
611
                                        return;
416
612
                                }
 
613
                                String complete_value =
 
614
                                        ( _parser_buffered_value_so_far + string_value ).trim();
417
615
 
418
 
                                // add on buffered multi-line content
419
 
                                String value = _parser_buffered_value_so_far + props[ 1 ];
 
616
                                // ignore empty values
 
617
                                if( complete_value.length() < 1 ) return;
420
618
 
421
619
                                // parse some properties
422
 
                                if( params[ 0 ].equals( "N" ) )
423
 
                                        parseN( params, value );
424
 
                                else if( params[ 0 ].equals( "FN" ) )
425
 
                                        parseFN( params, value );
426
 
                                else if( params[ 0 ].equals( "ORG" ) )
427
 
                                        parseORG( params, value );
428
 
                                else if( params[ 0 ].equals( "TEL" ) )
429
 
                                        parseTEL( params, value );
430
 
                                else if( params[ 0 ].equals( "EMAIL" ) )
431
 
                                        parseEMAIL( params, value );
432
 
                        }
 
620
                                if( name_param_parts[ 0 ].equals( "N" ) )
 
621
                                        parseN( name_param_parts, complete_value );
 
622
                                else if( name_param_parts[ 0 ].equals( "FN" ) )
 
623
                                        parseFN( name_param_parts, complete_value );
 
624
                                else if( name_param_parts[ 0 ].equals( "ORG" ) )
 
625
                                        parseORG( name_param_parts, complete_value );
 
626
                                else if( name_param_parts[ 0 ].equals( "TITLE" ) )
 
627
                                        parseTITLE( name_param_parts, complete_value );
 
628
                                else if( name_param_parts[ 0 ].equals( "TEL" ) )
 
629
                                        parseTEL( name_param_parts, complete_value );
 
630
                                else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
 
631
                                        parseEMAIL( name_param_parts, complete_value );
 
632
                                else if( name_param_parts[ 0 ].equals( "ADR" ) )
 
633
                                        parseADR( name_param_parts, complete_value );
 
634
                        }
 
635
                }
 
636
 
 
637
                private boolean doesStringEndInAnEscapeChar( String string )
 
638
                {
 
639
                        // count the number of backslashes at the end of the string
 
640
                        int count = 0;
 
641
                        for( int a = string.length() - 1; a >= 0; a-- )
 
642
                                if( string.charAt( a ) == '\\' )
 
643
                                        count++;
 
644
                                else
 
645
                                        break;
 
646
 
 
647
                        // if there are an even number of backslashes then the final one
 
648
                        // doesn't count
 
649
                        return ( count & 1 ) == 1;
 
650
                }
 
651
 
 
652
                private String[] splitValueBySemicolon( String value )
 
653
                {
 
654
                        // split string in to parts by semicolon
 
655
                        ArrayList< String > parts = new ArrayList< String >(
 
656
                                Arrays.asList( value.split(  ";" ) ) );
 
657
 
 
658
                        // go through parts
 
659
                        for( int a = 0; a < parts.size(); a++ )
 
660
                        {
 
661
                                String str = parts.get( a );
 
662
 
 
663
                                // look for parts that end in an escape character, but ignore
 
664
                                // the final part. We've already detected escape chars at the
 
665
                                // end of the final part in parseLine() and handled multi-lines
 
666
                                // accordingly.
 
667
                                if( a < parts.size() - 1 &&
 
668
                                        doesStringEndInAnEscapeChar( str ) )
 
669
                                {
 
670
                                        // join the next part to this part and remove the next part
 
671
                                        parts.set( a, str.substring( 0, str.length() - 1 ) +
 
672
                                                ';' + parts.get( a + 1 ) );
 
673
                                        parts.remove( a + 1 );
 
674
 
 
675
                                        // re-visit this part
 
676
                                        a--;
 
677
                                        continue;
 
678
                                }
 
679
 
 
680
                                // trim and replace string
 
681
                                str = str.trim();
 
682
                                parts.set( a, str );
 
683
                        }
 
684
 
 
685
                        String[] ret = new String[ parts.size() ];
 
686
                        return parts.toArray( ret );
433
687
                }
434
688
 
435
689
                private void parseN( String[] params, String value )
436
 
                                throws ParseException, SkipContactException,
437
 
                                AbortImportException
438
690
                {
439
691
                        // already got a better name?
440
692
                        if( _name_level >= NAMELEVEL_N ) return;
441
693
 
442
694
                        // get name parts
443
 
                        String[] nameparts = value.split( ";" );
444
 
                        for( int i = 0; i < nameparts.length; i++ )
445
 
                                nameparts[ i ] = nameparts[ i ].trim();
 
695
                        String[] name_parts = splitValueBySemicolon( value );
446
696
 
447
697
                        // build name
448
698
                        value = "";
449
 
                        if( nameparts.length > 1 && nameparts[ 1 ].length() > 0 )
450
 
                                value += nameparts[ 1 ];
451
 
                        if( nameparts[ 0 ].length() > 0 )
452
 
                                value += ( value.length() == 0? "" : " " ) + nameparts[ 0 ];
 
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 ];
453
703
 
454
704
                        // set name
455
705
                        setName( value );
456
706
                        _name_level = NAMELEVEL_N;
457
 
 
458
 
                        // check now to see if we need to import this contact (to avoid
459
 
                        // parsing the rest of the vCard unnecessarily)
460
 
                        if( !isImportRequired( getName() ) )
461
 
                                throw new SkipContactException();
462
707
                }
463
708
 
464
709
                private void parseFN( String[] params, String value )
465
 
                                throws ParseException, SkipContactException
466
710
                {
467
711
                        // already got a better name?
468
712
                        if( _name_level >= NAMELEVEL_FN ) return;
473
717
                }
474
718
 
475
719
                private void parseORG( String[] params, String value )
476
 
                                throws ParseException, SkipContactException
477
720
                {
478
 
                        // already got a better name?
479
 
                        if( _name_level >= NAMELEVEL_ORG ) return;
480
 
 
481
721
                        // get org parts
482
 
                        String[] orgparts = value.split( ";" );
483
 
                        for( int i = 0; i < orgparts.length; i++ )
484
 
                                orgparts[ i ] = orgparts[ i ].trim();
485
 
 
486
 
                        // build name
487
 
                        if( orgparts[ 0 ].length() == 0 && orgparts.length > 1 )
488
 
                                value = orgparts[ 1 ];
489
 
                        else
490
 
                                value = orgparts[ 0 ];
491
 
 
492
 
                        // set name
493
 
                        setName( value );
494
 
                        _name_level = NAMELEVEL_ORG;
 
722
                        String[] org_parts = splitValueBySemicolon( value );
 
723
                        if( org_parts == null || org_parts.length < 1 ) return;
 
724
 
 
725
                        // build organisation name
 
726
                        StringBuilder builder = new StringBuilder(
 
727
                                String.valueOf( org_parts[ 0 ] ) );
 
728
                        for( int a = 1; a < org_parts.length; a++ )
 
729
                                builder.append( ", " ).append( org_parts[ a ] );
 
730
                        String organisation = builder.toString();
 
731
 
 
732
                        // set organisation name (using a title we've previously found)
 
733
                        addOrganisation( organisation, _cached_title, true );
 
734
 
 
735
                        // if we've not previously found a title, store this organisation
 
736
                        // name (we'll need it when we find a title to update the
 
737
                        // organisation, by name), else if we *have* previously found a
 
738
                        // title, clear it (since we just used it)
 
739
                        if( _cached_title == null )
 
740
                                _cached_organisation = organisation;
 
741
                        else
 
742
                                _cached_title = null;
 
743
                }
 
744
 
 
745
                private void parseTITLE( String[] params, String value )
 
746
                {
 
747
                        // if we previously had an organisation, look it up and append this
 
748
                        // title to it
 
749
                        if( _cached_organisation != null && hasOrganisations() ) {
 
750
                                HashMap< String, ExtraDetail > datas = getOrganisations();
 
751
                                ExtraDetail detail = datas.get( _cached_organisation );
 
752
                                if( detail != null )
 
753
                                        detail.setExtra( value );
 
754
                        }
 
755
 
 
756
                        // same as when handling organisation, if we've not previously found
 
757
                        // an organisation we store this title, else we clear it (since we
 
758
                        // just appended this title to it)
 
759
                        if( _cached_organisation == null )
 
760
                                _cached_title = value;
 
761
                        else
 
762
                                _cached_organisation = null;
495
763
                }
496
764
 
497
765
                private void parseTEL( String[] params, String value )
498
 
                                throws ParseException
499
766
                {
500
767
                        if( value.length() == 0 ) return;
501
768
 
502
769
                        Set< String > types = extractTypes( params, Arrays.asList(
503
 
                                        "PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
504
 
                                        "PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
 
770
                                "PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
 
771
                                "PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
505
772
 
506
773
                        // here's the logic...
507
 
                        boolean preferred = types.contains( "PREF" );
508
 
                        if( types.contains( "VOICE" ) )
509
 
                                if( types.contains( "WORK" ) )
510
 
                                        addPhone( value, PhonesColumns.TYPE_WORK, preferred );
511
 
                                else
512
 
                                        addPhone( value, PhonesColumns.TYPE_HOME, preferred );
513
 
                        else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
514
 
                                addPhone( value, PhonesColumns.TYPE_MOBILE, preferred );
 
774
                        boolean is_preferred = types.contains( "PREF" );
 
775
                        int type;
515
776
                        if( types.contains( "FAX" ) )
516
777
                                if( types.contains( "HOME" ) )
517
 
                                        addPhone( value, PhonesColumns.TYPE_FAX_HOME, preferred );
 
778
                                        type = PhonesColumns.TYPE_FAX_HOME;
518
779
                                else
519
 
                                        addPhone( value, PhonesColumns.TYPE_FAX_WORK, preferred );
520
 
                        if( types.contains( "PAGER" ) )
521
 
                                addPhone( value, PhonesColumns.TYPE_PAGER, preferred );
 
780
                                        type = PhonesColumns.TYPE_FAX_WORK;
 
781
                        else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
 
782
                                type = PhonesColumns.TYPE_MOBILE;
 
783
                        else if( types.contains( "PAGER" ) )
 
784
                                type = PhonesColumns.TYPE_PAGER;
 
785
                        else if( types.contains( "WORK" ) )
 
786
                                type = PhonesColumns.TYPE_WORK;
 
787
                        else
 
788
                                type = PhonesColumns.TYPE_HOME;
 
789
 
 
790
                        // add phone number
 
791
                        addNumber( value, type, is_preferred );
522
792
                }
523
793
 
524
794
                public void parseEMAIL( String[] params, String value )
525
 
                                throws ParseException
526
795
                {
527
796
                        if( value.length() == 0 ) return;
528
797
 
529
798
                        Set< String > types = extractTypes( params, Arrays.asList(
530
 
                                        "PREF", "WORK", "HOME", "INTERNET" ) );
531
 
 
532
 
                        // here's the logic...
533
 
                        boolean preferred = types.contains( "PREF" );
534
 
                        if( types.contains( "WORK" ) )
535
 
                                addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
536
 
                        else
537
 
                                addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
538
 
                }
539
 
 
540
 
                public void finaliseParsing()
541
 
                                throws ParseException, SkipContactException,
542
 
                                AbortImportException
 
799
                                "PREF", "WORK", "HOME", "INTERNET" ) );
 
800
 
 
801
                        // add email address
 
802
                        boolean is_preferred = types.contains( "PREF" );
 
803
                        int type;
 
804
                        if( types.contains( "WORK" ) )
 
805
                                type = Contacts.ContactMethods.TYPE_WORK;
 
806
                        else
 
807
                                type = Contacts.ContactMethods.TYPE_HOME;
 
808
 
 
809
                        addEmail( value, type, is_preferred );
 
810
                }
 
811
 
 
812
                private void parseADR( String[] params, String value )
 
813
                {
 
814
                        // get address parts
 
815
                        String[] adr_parts = splitValueBySemicolon( value );
 
816
 
 
817
                        // build address
 
818
                        value = "";
 
819
                        for( int a = 0; a < adr_parts.length; a++ ) {
 
820
                                if( value.length() > 0 ) value += "\n";
 
821
                                value += adr_parts[ a ].trim();
 
822
                        }
 
823
 
 
824
                        Set< String > types = extractTypes( params, Arrays.asList(
 
825
                                "PREF", "WORK", "HOME", "INTERNET" ) );
 
826
 
 
827
                        // add address
 
828
                        int type;
 
829
                        if( types.contains( "WORK" ) )
 
830
                                type = Contacts.ContactMethods.TYPE_WORK;
 
831
                        else
 
832
                                type = Contacts.ContactMethods.TYPE_HOME;
 
833
 
 
834
                        addAddress( value, type );
 
835
                }
 
836
 
 
837
                public void finaliseVcard()
 
838
                        throws ParseException
543
839
                {
544
840
                        // missing version (and data is present)
545
 
                        if( _version == null && _lines != null )
 
841
                        if( _version == null && _buffers != null )
546
842
                                throw new ParseException( R.string.error_vcf_malformed );
547
843
 
548
 
                        //  missing name properties?
549
 
                        if( _name_level == NAMELEVEL_NONE )
550
 
                                throw new ParseException( R.string.error_vcf_noname );
551
 
 
552
 
                        // check if we should import this one? If we've already got an 'N'-
553
 
                        // type name, this will already have been done by parseN() so we
554
 
                        // mustn't do this here (or it could prompt twice!)
555
 
                        if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
556
 
                                throw new SkipContactException();
 
844
                        // finalise the parent class
 
845
                        try {
 
846
                                finalise();
 
847
                        }
 
848
                        catch( ContactNotIdentifiableException e ) {
 
849
                                throw new ParseException( R.string.error_vcf_notenoughinfo );
 
850
                        }
557
851
                }
558
852
 
559
853
                private String checkParam( String[] params, String name )
560
854
                {
561
 
                        Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
 
855
                        Pattern p = Pattern.compile(
 
856
                                "^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
562
857
                        for( int i = 0; i < params.length; i++ ) {
563
858
                                Matcher m = p.matcher( params[ i ] );
564
859
                                if( m.matches() )
565
 
                                        return m.group( 1 );
 
860
                                        return m.group( 2 );
566
861
                        }
567
862
                        return null;
568
863
                }
569
864
 
570
865
                private Set< String > extractTypes( String[] params,
571
 
                                List< String > validTypes )
 
866
                        List< String > valid_types )
572
867
                {
573
868
                        HashSet< String > types = new HashSet< String >();
574
869
 
575
870
                        // get 3.0-style TYPE= param
576
 
                        String typeParam;
577
 
                        if( ( typeParam = checkParam( params, "TYPE" ) ) != null ) {
578
 
                                String[] bits = typeParam.split( "," );
579
 
                                for( int i = 0; i < bits.length; i++ )
580
 
                                        if( validTypes.contains( bits[ i ] ) )
581
 
                                                types.add( bits[ i ] );
 
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 ] );
582
877
                        }
583
878
 
584
879
                        // get 2.1-style type param
585
880
                        if( _version.equals( "2.1" ) ) {
586
881
                                for( int i = 1; i < params.length; i++ )
587
 
                                        if( validTypes.contains( params[ i ] ) )
 
882
                                        if( valid_types.contains( params[ i ] ) )
588
883
                                                types.add( params[ i ] );
589
884
                        }
590
885
 
591
886
                        return types;
592
887
                }
593
888
 
594
 
                private UnencodeResult unencodeQuotedPrintable( String str, String charset )
 
889
                private UnencodeResult unencodeQuotedPrintable( ByteBuffer in )
595
890
                {
596
891
                        boolean another = false;
597
892
 
598
 
                        // default encoding scheme
599
 
                        if( charset == null ) charset = "UTF-8";
600
 
 
601
 
                        // unencode quoted-pritable encoding, as per RFC1521 section 5.1
602
 
                        byte[] bytes = new byte[ str.length() ];
 
893
                        // unencode quoted-printable encoding, as per RFC1521 section 5.1
 
894
                        byte[] out = new byte[ in.limit() - in.position() ];
603
895
                        int j = 0;
604
 
                        for( int i = 0; i < str.length(); i++ )
 
896
                        for( int i = in.position(); i < in.limit(); i++ )
605
897
                        {
606
898
                                // get next char and process...
607
 
                                char ch = str.charAt( i );
608
 
                                if( ch == '=' && i < str.length() - 2 )
 
899
                                byte ch = in.array()[ i ];
 
900
                                if( ch == '=' && i < in.limit() - 2 )
609
901
                                {
610
902
                                        // we found a =XX format byte, add it
611
 
                                        bytes[ j ] = (byte)(
612
 
                                                        Character.digit( str.charAt( i + 1 ), 16 ) * 16 +
613
 
                                                        Character.digit( str.charAt( i + 2 ), 16 ) );
 
903
                                        out[ j ] = (byte)(
 
904
                                                        Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
 
905
                                                        Character.digit( in.array()[ i + 2 ], 16 ) );
614
906
                                        i += 2;
615
907
                                }
616
 
                                else if( ch == '=' && i == str.length() - 1 )
 
908
                                else if( ch == '=' && i == in.limit() - 1 )
617
909
                                {
618
910
                                        // we found a '=' at the end of a line signifying a multi-
619
911
                                        // line string, so we don't add it.
622
914
                                }
623
915
                                else
624
916
                                        // just a normal char...
625
 
                                        bytes[ j ] = (byte)ch;
 
917
                                        out[ j ] = (byte)ch;
626
918
                                j++;
627
919
                        }
628
920
 
629
 
                        return new UnencodeResult( another, bytes, j );
 
921
                        return new UnencodeResult( another, ByteBuffer.wrap( out, 0, j ) );
 
922
                }
 
923
 
 
924
                private ByteBuffer transcodeAsciiToUtf8( ByteBuffer in )
 
925
                {
 
926
                        // transcode
 
927
                        byte[] out = new byte[ ( in.limit() - in.position() ) * 2 ];
 
928
                        int j = 0;
 
929
                        for( int a = in.position(); a < in.limit(); a++ )
 
930
                        {
 
931
                                // if char is < 127, keep it as-is
 
932
                                if( in.array()[ a ] >= 0 )
 
933
                                        out[ j++ ] = in.array()[ a ];
 
934
 
 
935
                                // else, convert it to UTF-8
 
936
                                else {
 
937
                                        int b = 0xff & (int)in.array()[ a ];
 
938
                                        out[ j++ ] = (byte)( 0xc0 | ( b >> 6 ) );
 
939
                                        out[ j++ ] = (byte)( 0x80 | ( b & 0x3f ) );
 
940
                                }
 
941
                        }
 
942
 
 
943
                        return ByteBuffer.wrap( out, 0, j );
630
944
                }
631
945
        }
632
946
}