(root)/android/import-contacts
» Revision
36
: src/org/waxworlds/edam/importcontacts/VCFImporter.java
To get this branch, use:
bzr branch
http://bzr.ed.am/android/import-contacts
« back to all changes in this revision
Viewing changes to src/org/waxworlds/edam/importcontacts/VCFImporter.java
- browse files at revision 36
- compare with another revision
- download diff
- download tarball
- view history from revision 36
- Committer: edam
- Date: 2011-03-19 20:33:09 UTC
- Revision ID: edam@waxworlds.org-20110319203309-5dzfyqrxwk94jtin
- formatting: removed some double-indents on overrunning lines
- updated TODO and NEWS
- rewrote central logic of parser so it makes more sense, looks nicer and has a small optimisation (getting name and params from line only when necessary)
- optimised unnecessary mutliple converting of lines to US-ASCII
- re-wrote line extraction from vcards so that we can lookahead for v3 folded lines
- added support for v3 folded lines
- updated TODO and NEWS
- rewrote central logic of parser so it makes more sense, looks nicer and has a small optimisation (getting name and params from line only when necessary)
- optimised unnecessary mutliple converting of lines to US-ASCII
- re-wrote line extraction from vcards so that we can lookahead for v3 folded lines
- added support for v3 folded lines
added
removed
src/org/waxworlds/edam/importcontacts/VCFImporter.java
34
34
import java.nio.ByteBuffer;
35
35
import java.util.Arrays;
36
36
import java.util.HashSet;
37
import java.util.Iterator;
37
38
import java.util.List;
38
39
import java.util.Set;
39
40
import java.util.Vector;
40
41
import java.util.regex.Matcher;
41
42
import java.util.regex.Pattern;
43
import java.util.NoSuchElementException;
44
import java.lang.UnsupportedOperationException;
42
45
43
46
import android.content.SharedPreferences;
44
47
import android.provider.Contacts;
120
123
{
121
124
// open file
122
125
BufferedReader reader = new BufferedReader(
123
new FileReader( file ) );
126
new FileReader( file ) );
124
127
125
128
// read
126
129
String line;
129
132
{
130
133
if( !inVCard ) {
131
134
// look for vcard beginning
132
if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
135
if( line.matches( "^BEGIN:VCARD" ) ) {
133
136
inVCard = true;
134
137
_vCardCount++;
135
138
}
136
139
}
137
else if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
140
else if( line.matches( "^END:VCARD" ) )
138
141
inVCard = false;
139
142
}
140
143
178
181
}
179
182
180
183
private void importVCardFileContent( byte[] content, String fileName )
181
throws AbortImportException
184
throws AbortImportException
182
185
{
183
ByteBuffer buffers[] = getLinesFromContent( content );
184
185
186
// go through lines
186
187
VCard vCard = null;
187
for( int i = 0; i < buffers.length; i++ )
188
ContentLineIterator cli = new ContentLineIterator( content );
189
while( cli.hasNext() )
188
190
{
191
ByteBuffer buffer = cli.next();
192
189
193
// get a US-ASCII version of the line for processing
190
194
String line;
191
195
try {
192
line = new String( buffers[ i ].array(), buffers[ i ].position(),
193
buffers[ i ].limit() - buffers[ i ].position(), "US-ASCII" );
196
line = new String( buffer.array(), buffer.position(),
197
buffer.limit() - buffer.position(), "US-ASCII" );
194
198
}
195
199
catch( UnsupportedEncodingException e ) {
196
200
// we know US-ASCII is supported, so appease the compiler...
199
203
200
204
if( vCard == null ) {
201
205
// look for vcard beginning
202
if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
206
if( line.matches( "^BEGIN:VCARD" ) ) {
203
207
setProgress( ++_progress );
204
208
vCard = new VCard();
205
209
}
206
210
}
207
211
else {
208
212
// look for vcard content or ending
209
if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
213
if( line.matches( "^END:VCARD" ) )
210
214
{
211
215
// store vcard and do away with it
212
216
try {
216
220
catch( VCard.ParseException e ) {
217
221
skipContact();
218
222
if( !showContinue(
219
getText( R.string.error_vcf_parse ).toString()
220
+ fileName + "\n" + e.getMessage() ) )
223
getText( R.string.error_vcf_parse ).toString()
224
+ fileName + "\n" + e.getMessage() ) )
225
{
221
226
finish( ACTION_ABORT );
227
}
222
228
}
223
229
catch( VCard.SkipContactException e ) {
224
230
skipContact();
230
236
{
231
237
// try giving the line to the vcard
232
238
try {
233
vCard.parseLine( buffers[ i ] );
239
vCard.parseLine( buffer, line,
240
cli.doesNextLineLookFolded() );
234
241
}
235
242
catch( VCard.ParseException e ) {
236
243
skipContact();
237
244
if( !showContinue(
238
getText( R.string.error_vcf_parse ).toString()
239
+ fileName + "\n" + e.getMessage() ) )
245
getText( R.string.error_vcf_parse ).toString()
246
+ fileName + "\n" + e.getMessage() ) )
247
{
240
248
finish( ACTION_ABORT );
249
}
241
250
242
251
// although we're continuing, we still need to abort
243
252
// this vCard. Further lines will be ignored until we
255
264
}
256
265
}
257
266
258
private ByteBuffer[] getLinesFromContent( byte[] content )
267
class ContentLineIterator implements Iterator< ByteBuffer >
259
268
{
260
// count lines in data
261
int num_lines = 1;
262
for( int a = 0; a < content.length; a++ )
263
if( content[ a ] == '\n' )
264
num_lines++;
265
266
// get lines, removing \r's and \n's as we go
267
ByteBuffer lines[] = new ByteBuffer[ num_lines ];
268
int last = 0;
269
for( int a = 0, b = 0; a < content.length; a++ )
270
if( content[ a ] == '\n' ) {
271
int to = ( a > 0 && content[ a - 1 ] == '\r' &&
272
a - 1 >= last )? a - 1 : a;
273
lines[ b++ ] = ByteBuffer.wrap( content, last, to - last );
274
last = a + 1;
269
protected byte[] _content = null;
270
protected int _pos = 0;
271
272
public ContentLineIterator( byte[] content )
273
{
274
_content = content;
275
}
276
277
@Override
278
public boolean hasNext()
279
{
280
return _pos < _content.length;
281
}
282
283
@Override
284
public ByteBuffer next()
285
{
286
int initial_pos = _pos;
287
288
// find newline
289
for( ; _pos < _content.length; _pos++ )
290
if( _content[ _pos ] == '\n' )
291
{
292
// adjust for a \r preceding the \n
293
int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
294
_pos > initial_pos )? _pos - 1 : _pos;
295
_pos++;
296
return ByteBuffer.wrap( _content, initial_pos,
297
to - initial_pos );
298
}
299
300
// we didn't find one, but were there bytes left?
301
if( _pos != initial_pos ) {
302
int to = _pos;
303
_pos++;
304
return ByteBuffer.wrap( _content, initial_pos,
305
to - initial_pos );
275
306
}
276
lines[ lines.length - 1 ] = ByteBuffer.wrap( content, last,
277
content.length - last );
278
279
return lines;
307
308
// no bytes left
309
throw new NoSuchElementException();
310
}
311
312
@Override
313
public void remove()
314
{
315
throw new UnsupportedOperationException();
316
}
317
318
/**
319
* Does the next line, if there is one, look like it should be folded
320
* onto the end of this one?
321
* @return
322
*/
323
public boolean doesNextLineLookFolded()
324
{
325
return _pos > 0 && _pos < _content.length &&
326
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
327
}
280
328
}
281
329
282
330
private class VCard extends ContactData
289
337
private String _version = null;
290
338
private Vector< ByteBuffer > _buffers = null;
291
339
private int _name_level = NAMELEVEL_NONE;
292
private boolean _parser_in_multiline = false;
340
private boolean _parser_in_encoded_multiline = false;
341
private boolean _parser_in_folded_multiline = false;
293
342
private String _parser_current_name_and_params = null;
294
343
private String _parser_buffered_value_so_far = "";
295
344
334
383
@SuppressWarnings("serial")
335
384
protected class SkipContactException extends Exception { }
336
385
337
public void parseLine( ByteBuffer buffer )
338
throws ParseException, SkipContactException,
339
AbortImportException
340
{
341
// get a US-ASCII version of the line for processing
342
String line;
343
try {
344
line = new String( buffer.array(), buffer.position(),
345
buffer.limit() - buffer.position(), "US-ASCII" );
346
}
347
catch( UnsupportedEncodingException e ) {
348
// we know US-ASCII is supported, so appease the compiler...
349
line = "";
350
}
351
352
// ignore empty lines
353
if( line.trim().equals( "" ) ) return;
354
355
// split line into name and value parts (this may turn out to be
356
// unwanted if the line is a subsequent line in a multi-line
357
// value, but we have to do this now to check for and handle VCF
358
// versions first). Also, the value part is only created tentatively
359
// because it may have an encoding/charset. Since we're treating it
360
// as UTF-8 (which is compatible with 7-bit US-ASCII) this is ok
361
// though so long as we later use the raw bytes. ALso we check for
362
// malformed property:name pairs.
363
String name_and_params, string_value;
364
{
365
String[] parts = line.split( ":", 2 );
366
if( parts.length == 2 ) {
367
name_and_params = parts[ 0 ].trim();
368
string_value = parts[ 1 ].trim();
369
if( name_and_params.length() == 0 )
370
throw new ParseException( R.string.error_vcf_malformed );
371
}
372
else
373
{
374
if( !_parser_in_multiline )
375
throw new ParseException( R.string.error_vcf_malformed );
376
name_and_params = null;
377
string_value = null;
378
}
379
}
380
381
// if we haven't yet got a version, we won't be paring anything!
386
private String extractCollonPartFromLine( ByteBuffer buffer,
387
String line, boolean former )
388
{
389
String ret = null;
390
391
// get a US-ASCII version of the line for processing, unless we were
392
// supplied with one
393
if( line == null ) {
394
try {
395
line = new String( buffer.array(), buffer.position(),
396
buffer.limit() - buffer.position(), "US-ASCII" );
397
}
398
catch( UnsupportedEncodingException e ) {
399
// we know US-ASCII is supported, so appease the compiler...
400
line = "";
401
}
402
}
403
404
// split line into name and value parts and check to make sure we
405
// only got 2 parts and that the first part is not zero in length
406
String[] parts = line.split( ":", 2 );
407
if( parts.length == 2 && parts[ 0 ].length() > 0 )
408
ret = parts[ former? 0 : 1 ];
409
410
return ret;
411
}
412
413
private String extractNameAndParamsFromLine( ByteBuffer buffer,
414
String line )
415
{
416
return extractCollonPartFromLine( buffer, line, true );
417
}
418
419
private String extractValueFromLine( ByteBuffer buffer, String line )
420
{
421
return extractCollonPartFromLine( buffer, line, false );
422
}
423
424
public void parseLine( ByteBuffer buffer, String line,
425
boolean next_line_looks_folded )
426
throws ParseException, SkipContactException,
427
AbortImportException
428
{
429
// do we have a version yet?
382
430
if( _version == null )
383
431
{
384
// is this a version?
385
if( name_and_params.equals( "VERSION" ) )
432
// tentatively get name and params from line
433
String name_and_params =
434
extractNameAndParamsFromLine( buffer, line );
435
436
// is it a version line?
437
if( name_and_params != null &&
438
name_and_params.equals( "VERSION" ) )
386
439
{
387
// yes, check/store it
388
if( !string_value.equals( "2.1" ) &&
389
!string_value.equals( "3.0" ) )
440
// yes, get it!
441
String value = extractValueFromLine( buffer, line );
442
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
390
443
throw new ParseException( R.string.error_vcf_version );
391
_version = string_value;
444
_version = value;
392
445
393
// parse any other buffers we've accumulated so far
446
// parse any buffers we've been accumulating while we waited
447
// for a version
394
448
if( _buffers != null )
395
449
for( int i = 0; i < _buffers.size(); i++ )
396
parseLine( _buffers.get( i ) );
450
parseLine( _buffers.get( i ), null,
451
i + 1 < _buffers.size() &&
452
_buffers.get( i + 1 ).hasRemaining() &&
453
_buffers.get( i + 1 ).get(
454
_buffers.get( i + 1 ).position() ) == ' ' );
397
455
_buffers = null;
398
456
}
399
457
else
400
458
{
401
// no, so stash this buffer till we have a version
459
// no, so stash this line till we get a version
402
460
if( _buffers == null )
403
461
_buffers = new Vector< ByteBuffer >();
404
462
_buffers.add( buffer );
406
464
}
407
465
else
408
466
{
409
// value bytes, for processing
410
ByteBuffer value;
467
// name and params and the position in the buffer where the
468
// "value" part of the line start
469
String name_and_params;
470
int pos;
411
471
412
if( _parser_in_multiline )
472
if( _parser_in_encoded_multiline ||
473
_parser_in_folded_multiline )
413
474
{
414
475
// if we're currently in a multi-line value, use the stored
415
476
// property name and parameters
416
477
name_and_params = _parser_current_name_and_params;
417
478
418
// find start of string (skip spaces/tabs)
419
int pos = buffer.position();
420
byte[] buffer_array = buffer.array();
421
while( pos < buffer.limit() && (
422
buffer_array[ pos ] == ' ' ||
423
buffer_array[ pos ] == '\t' ) )
424
{
479
pos = buffer.position();
480
481
// for folded multi-lines, skip the single space at the
482
// start of the next line
483
if( _parser_in_folded_multiline )
425
484
pos++;
426
}
427
485
428
// get value from buffer
429
value = ByteBuffer.wrap( buffer.array(), pos,
430
buffer.limit() - pos );
486
// else, this must be an encoded multi-line, so skip any
487
// whitespace we find at the start of the next line
488
else
489
while( pos < buffer.limit() && (
490
buffer.get( pos ) == ' ' ||
491
buffer.get( pos ) == '\t' ) )
492
{
493
pos++;
494
}
431
495
}
432
496
else
433
497
{
498
// get name and params from line, and since we're not
499
// parsing a subsequent line in a multi-line, this should
500
// not fail, or it's an error
501
name_and_params =
502
extractNameAndParamsFromLine( buffer, line );
503
if( name_and_params == null )
504
throw new ParseException(
505
R.string.error_vcf_malformed );
506
434
507
// calculate how many chars to skip from beginning of line
435
508
// so we skip the property "name:" part
436
int pos = buffer.position() + name_and_params.length() + 1;
437
438
// get value from buffer
439
value = ByteBuffer.wrap( buffer.array(), pos,
440
buffer.limit() - pos );
509
pos = buffer.position() + name_and_params.length() + 1;
441
510
442
511
// reset the saved multi-line state
443
512
_parser_current_name_and_params = name_and_params;
444
513
_parser_buffered_value_so_far = "";
445
514
}
446
515
516
// get value from buffer, as raw bytes
517
ByteBuffer value;
518
value = ByteBuffer.wrap( buffer.array(), pos,
519
buffer.limit() - pos );
520
447
521
// get parameter parts
448
522
String[] name_param_parts = name_and_params.split( ";", -1 );
449
523
for( int i = 0; i < name_param_parts.length; i++ )
463
537
String charset = checkParam( name_param_parts, "CHARSET" );
464
538
if( charset != null ) charset = charset.toUpperCase();
465
539
if( charset != null && !charset.equals( "US-ASCII" ) &&
466
!charset.equals( "ASCII" ) && !charset.equals( "UTF-8" ) )
540
!charset.equals( "ASCII" ) &&
541
!charset.equals( "UTF-8" ) )
467
542
{
468
543
throw new ParseException( R.string.error_vcf_charset );
469
544
}
477
552
// unencoding_result = unencodeBase64( props[ 1 ], charset );
478
553
if( unencoding_result != null ) {
479
554
value = unencoding_result.getBuffer();
480
_parser_in_multiline =
555
_parser_in_encoded_multiline =
481
556
unencoding_result.isAnotherLineRequired();
482
557
}
483
558
488
563
}
489
564
490
565
// process charset
566
String string_value;
491
567
try {
492
string_value =
493
new String( value.array(), value.position(),
494
value.limit() - value.position(), charset );
568
string_value = new String( value.array(), value.position(),
569
value.limit() - value.position(), charset );
495
570
} catch( UnsupportedEncodingException e ) {
496
571
throw new ParseException( R.string.error_vcf_charset );
497
572
}
498
573
574
// now we know whether we're in an encoding multi-line,
575
// determine if we're in a v3 folded multi-line or not
576
_parser_in_folded_multiline = !_parser_in_encoded_multiline &&
577
_version.equals( "3.0" ) && next_line_looks_folded;
578
499
579
// handle multi-line requests
500
if( _parser_in_multiline ) {
580
if( _parser_in_encoded_multiline ||
581
_parser_in_folded_multiline )
582
{
501
583
_parser_buffered_value_so_far += string_value;
502
584
return;
503
585
}
524
606
}
525
607
526
608
private void parseN( String[] params, String value )
527
throws ParseException, SkipContactException,
528
AbortImportException
609
throws ParseException, SkipContactException,
610
AbortImportException
529
611
{
530
612
// already got a better name?
531
613
if( _name_level >= NAMELEVEL_N ) return;
553
635
}
554
636
555
637
private void parseFN( String[] params, String value )
556
throws ParseException, SkipContactException
638
throws ParseException, SkipContactException
557
639
{
558
640
// already got a better name?
559
641
if( _name_level >= NAMELEVEL_FN ) return;
564
646
}
565
647
566
648
private void parseORG( String[] params, String value )
567
throws ParseException, SkipContactException
649
throws ParseException, SkipContactException
568
650
{
569
651
// already got a better name?
570
652
if( _name_level >= NAMELEVEL_ORG ) return;
586
668
}
587
669
588
670
private void parseTEL( String[] params, String value )
589
throws ParseException
671
throws ParseException
590
672
{
591
673
if( value.length() == 0 ) return;
592
674
593
675
Set< String > types = extractTypes( params, Arrays.asList(
594
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
595
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
676
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
677
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
596
678
597
679
// here's the logic...
598
680
boolean preferred = types.contains( "PREF" );
617
699
}
618
700
619
701
public void parseEMAIL( String[] params, String value )
620
throws ParseException
702
throws ParseException
621
703
{
622
704
if( value.length() == 0 ) return;
623
705
624
706
Set< String > types = extractTypes( params, Arrays.asList(
625
"PREF", "WORK", "HOME", "INTERNET" ) );
707
"PREF", "WORK", "HOME", "INTERNET" ) );
626
708
627
709
// here's the logic...
628
710
boolean preferred = types.contains( "PREF" );
633
715
}
634
716
635
717
public void finaliseParsing()
636
throws ParseException, SkipContactException,
637
AbortImportException
718
throws ParseException, SkipContactException,
719
AbortImportException
638
720
{
639
721
// missing version (and data is present)
640
722
if( _version == null && _buffers != null )
654
736
private String checkParam( String[] params, String name )
655
737
{
656
738
Pattern p = Pattern.compile(
657
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
739
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
658
740
for( int i = 0; i < params.length; i++ ) {
659
741
Matcher m = p.matcher( params[ i ] );
660
742
if( m.matches() )
664
746
}
665
747
666
748
private Set< String > extractTypes( String[] params,
667
List< String > valid_types )
749
List< String > valid_types )
668
750
{
669
751
HashSet< String > types = new HashSet< String >();
670
752
691
773
{
692
774
boolean another = false;
693
775
694
// unencode quoted-pritable encoding, as per RFC1521 section 5.1
776
// unencode quoted-printable encoding, as per RFC1521 section 5.1
695
777
byte[] out = new byte[ in.limit() - in.position() ];
696
778
int j = 0;
697
779
for( int i = in.position(); i < in.limit(); i++ )
702
784
{
703
785
// we found a =XX format byte, add it
704
786
out[ j ] = (byte)(
705
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
706
Character.digit( in.array()[ i + 2 ], 16 ) );
787
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
788
Character.digit( in.array()[ i + 2 ], 16 ) );
707
789
i += 2;
708
790
}
709
791
else if( ch == '=' && i == in.limit() - 1 )
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480