270
class ContentLineIterator implements Iterator< ByteBuffer >
258
private ByteBuffer[] getLinesFromContent( byte[] content )
272
protected byte[] _content = null;
273
protected int _pos = 0;
275
public ContentLineIterator( byte[] content )
281
public boolean hasNext()
283
return _pos < _content.length;
287
public ByteBuffer next()
289
int initial_pos = _pos;
292
for( ; _pos < _content.length; _pos++ )
293
if( _content[ _pos ] == '\n' )
295
// adjust for a \r preceding the \n
296
int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
297
_pos > initial_pos )? _pos - 1 : _pos;
299
return ByteBuffer.wrap( _content, initial_pos,
303
// we didn't find one, but were there bytes left?
304
if( _pos != initial_pos ) {
307
return ByteBuffer.wrap( _content, initial_pos,
260
// count lines in data
262
for( int a = 0; a < content.length; a++ )
263
if( content[ a ] == '\n' )
266
// get lines, removing \r's and \n's as we go
267
ByteBuffer lines[] = new ByteBuffer[ num_lines ];
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 );
312
throw new NoSuchElementException();
318
throw new UnsupportedOperationException();
322
* Does the next line, if there is one, look like it should be folded
323
* onto the end of this one?
326
public boolean doesNextLineLookFolded()
328
return _pos > 0 && _pos < _content.length &&
329
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
276
lines[ lines.length - 1 ] = ByteBuffer.wrap( content, last,
277
content.length - last );
333
282
private class VCard extends ContactData
335
284
private final static int NAMELEVEL_NONE = 0;
336
private final static int NAMELEVEL_FN = 1;
337
private final static int NAMELEVEL_N = 2;
339
private final static int MULTILINE_NONE = 0;
340
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
341
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
342
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
285
private final static int NAMELEVEL_ORG = 1;
286
private final static int NAMELEVEL_FN = 2;
287
private final static int NAMELEVEL_N = 3;
344
289
private String _version = null;
345
290
private Vector< ByteBuffer > _buffers = null;
346
291
private int _name_level = NAMELEVEL_NONE;
347
private int _parser_multiline_state = MULTILINE_NONE;
292
private boolean _parser_in_multiline = false;
348
293
private String _parser_current_name_and_params = null;
349
294
private String _parser_buffered_value_so_far = "";
350
private String _cached_organisation = null;
351
private String _cached_title = null;
353
296
protected class UnencodeResult
391
334
@SuppressWarnings("serial")
392
335
protected class SkipContactException extends Exception { }
394
private String extractCollonPartFromLine( ByteBuffer buffer,
395
String line, boolean former )
399
// get a US-ASCII version of the line for processing, unless we were
403
line = new String( buffer.array(), buffer.position(),
404
buffer.limit() - buffer.position(), "US-ASCII" );
406
catch( UnsupportedEncodingException e ) {
407
// we know US-ASCII is supported, so appease the compiler...
412
// split line into name and value parts and check to make sure we
413
// only got 2 parts and that the first part is not zero in length
414
String[] parts = line.split( ":", 2 );
415
if( parts.length == 2 && parts[ 0 ].length() > 0 )
416
ret = parts[ former? 0 : 1 ];
421
private String extractNameAndParamsFromLine( ByteBuffer buffer,
424
return extractCollonPartFromLine( buffer, line, true );
427
private String extractValueFromLine( ByteBuffer buffer, String line )
429
return extractCollonPartFromLine( buffer, line, false );
432
public void parseLine( ByteBuffer buffer, String line,
433
boolean next_line_looks_folded )
434
throws ParseException, SkipContactException,
437
// do we have a version yet?
337
public void parseLine( ByteBuffer buffer )
338
throws ParseException, SkipContactException,
341
// get a US-ASCII version of the line for processing
344
line = new String( buffer.array(), buffer.position(),
345
buffer.limit() - buffer.position(), "US-ASCII" );
347
catch( UnsupportedEncodingException e ) {
348
// we know US-ASCII is supported, so appease the compiler...
352
// ignore empty lines
353
if( line.trim() == "" ) return;
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;
365
String[] bits = line.split( ":", 2 );
366
if( bits.length == 2 ) {
367
name_and_params = bits[ 0 ].trim();
368
string_value = bits[ 1 ].trim();
369
if( name_and_params.length() == 0 )
370
throw new ParseException( R.string.error_vcf_malformed );
374
if( !_parser_in_multiline )
375
throw new ParseException( R.string.error_vcf_malformed );
376
name_and_params = null;
381
// if we haven't yet got a version, we won't be paring anything!
438
382
if( _version == null )
440
// tentatively get name and params from line
441
String name_and_params =
442
extractNameAndParamsFromLine( buffer, line );
444
// is it a version line?
445
if( name_and_params != null &&
446
name_and_params.equals( "VERSION" ) )
384
// is this a version?
385
if( name_and_params.equals( "VERSION" ) )
449
String value = extractValueFromLine( buffer, line );
450
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
387
// yes, check/store it
388
if( !string_value.equals( "2.1" ) &&
389
!string_value.equals( "3.0" ) )
451
390
throw new ParseException( R.string.error_vcf_version );
391
_version = string_value;
454
// parse any buffers we've been accumulating while we waited
393
// parse any other buffers we've accumulated so far
456
394
if( _buffers != null )
457
395
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() ) == ' ' );
396
parseLine( _buffers.get( i ) );
467
// no, so stash this line till we get a version
401
// no, so stash this buffer till we have a version
468
402
if( _buffers == null )
469
403
_buffers = new Vector< ByteBuffer >();
470
404
_buffers.add( buffer );
475
// name and params and the position in the buffer where the
476
// "value" part of the line start
477
String name_and_params;
409
// value bytes, for processing
480
if( _parser_multiline_state != MULTILINE_NONE )
412
if( _parser_in_multiline )
482
414
// if we're currently in a multi-line value, use the stored
483
415
// property name and parameters
484
416
name_and_params = _parser_current_name_and_params;
486
// skip some initial line characters, depending on the type
487
// of multi-line we're handling
488
pos = buffer.position();
489
switch( _parser_multiline_state )
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' ) )
491
case MULTILINE_FOLDED:
494
case MULTILINE_ENCODED:
495
while( pos < buffer.limit() && (
496
buffer.get( pos ) == ' ' ||
497
buffer.get( pos ) == '\t' ) )
506
// take us out of multi-line so that we can re-detect that
507
// this line is a multi-line or not
508
_parser_multiline_state = MULTILINE_NONE;
428
// get value from buffer
429
value = ByteBuffer.wrap( buffer.array(), pos,
430
buffer.limit() - pos );
512
// get name and params from line, and since we're not
513
// parsing a subsequent line in a multi-line, this should
514
// not fail, or it's an error
516
extractNameAndParamsFromLine( buffer, line );
517
if( name_and_params == null )
518
throw new ParseException(
519
R.string.error_vcf_malformed );
434
// ignore empty values
435
if( string_value.length() < 1 ) return;
521
437
// calculate how many chars to skip from beginning of line
522
438
// so we skip the property "name:" part
523
pos = buffer.position() + name_and_params.length() + 1;
439
int pos = buffer.position() + name_and_params.length() + 1;
441
// get value from buffer
442
value = ByteBuffer.wrap( buffer.array(), pos,
443
buffer.limit() - pos );
525
445
// reset the saved multi-line state
526
446
_parser_current_name_and_params = name_and_params;
527
447
_parser_buffered_value_so_far = "";
530
// get value from buffer, as raw bytes
532
value = ByteBuffer.wrap( buffer.array(), pos,
533
buffer.limit() - pos );
535
450
// get parameter parts
536
String[] name_param_parts = name_and_params.split( ";", -1 );
537
for( int i = 0; i < name_param_parts.length; i++ )
538
name_param_parts[ i ] = name_param_parts[ i ].trim();
451
String[] name_and_param_bits = name_and_params.split( ";" );
452
for( int i = 0; i < name_and_param_bits.length; i++ )
453
name_and_param_bits[ i ] = name_and_param_bits[ i ].trim();
540
455
// parse encoding parameter
541
String encoding = checkParam( name_param_parts, "ENCODING" );
456
String encoding = checkParam( name_and_param_bits, "ENCODING" );
542
457
if( encoding != null ) encoding = encoding.toUpperCase();
543
458
if( encoding != null && !encoding.equals( "8BIT" ) &&
544
459
!encoding.equals( "QUOTED-PRINTABLE" ) )
563
477
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
564
478
unencoding_result = unencodeQuotedPrintable( value );
565
479
// else if( encoding != null && encoding.equals( "BASE64" ) )
566
// unencoding_result = unencodeBase64( props[ 1 ], charset );
480
// result = unencodeBase64( props[ 1 ], charset );
567
481
if( unencoding_result != null ) {
568
482
value = unencoding_result.getBuffer();
569
if( unencoding_result.isAnotherLineRequired() )
570
_parser_multiline_state = MULTILINE_ENCODED;
483
_parser_in_multiline =
484
unencoding_result.isAnotherLineRequired();
573
487
// convert 8-bit ASCII charset to US-ASCII
574
if( charset == null || charset.equals( "ASCII" ) ) {
488
if( charset == null || charset == "ASCII" ) {
575
489
value = transcodeAsciiToUtf8( value );
576
490
charset = "UTF-8";
579
493
// process charset
582
string_value = new String( value.array(), value.position(),
583
value.limit() - value.position(), charset );
496
new String( value.array(), value.position(),
497
value.limit() - value.position(), charset );
584
498
} catch( UnsupportedEncodingException e ) {
585
499
throw new ParseException( R.string.error_vcf_charset );
588
// for some entries that have semicolon-separated value parts,
589
// check to see if the value ends in an escape character, which
590
// 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" ) ) &&
594
doesStringEndInAnEscapeChar( string_value ) )
596
_parser_multiline_state = MULTILINE_ESCAPED;
597
string_value = string_value.substring( 0,
598
string_value.length() - 1 );
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
603
if( _parser_multiline_state == MULTILINE_NONE &&
604
_version.equals( "3.0" ) && next_line_looks_folded )
606
_parser_multiline_state = MULTILINE_FOLDED;
609
// handle multi-lines by buffering them and parsing them when we
610
// are processing the last line in a multi-line sequence
611
if( _parser_multiline_state != MULTILINE_NONE ) {
502
// handle multi-line requests
503
if( _parser_in_multiline ) {
612
504
_parser_buffered_value_so_far += string_value;
508
// add on buffered multi-line content
615
509
String complete_value =
616
( _parser_buffered_value_so_far + string_value ).trim();
618
// ignore empty values
619
if( complete_value.length() < 1 ) return;
510
_parser_buffered_value_so_far + string_value;
621
512
// parse some properties
622
if( name_param_parts[ 0 ].equals( "N" ) )
623
parseN( name_param_parts, complete_value );
624
else if( name_param_parts[ 0 ].equals( "FN" ) )
625
parseFN( name_param_parts, complete_value );
626
else if( name_param_parts[ 0 ].equals( "ORG" ) )
627
parseORG( name_param_parts, complete_value );
628
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
629
parseTITLE( name_param_parts, complete_value );
630
else if( name_param_parts[ 0 ].equals( "TEL" ) )
631
parseTEL( name_param_parts, complete_value );
632
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
633
parseEMAIL( name_param_parts, complete_value );
634
else if( name_param_parts[ 0 ].equals( "ADR" ) )
635
parseADR( name_param_parts, complete_value );
639
private boolean doesStringEndInAnEscapeChar( String string )
641
// count the number of backslashes at the end of the string
643
for( int a = string.length() - 1; a >= 0; a-- )
644
if( string.charAt( a ) == '\\' )
649
// if there are an even number of backslashes then the final one
651
return ( count & 1 ) == 1;
654
private String[] splitValueBySemicolon( String value )
656
// split string in to parts by semicolon
657
ArrayList< String > parts = new ArrayList< String >(
658
Arrays.asList( value.split( ";" ) ) );
661
for( int a = 0; a < parts.size(); a++ )
663
String str = parts.get( a );
665
// look for parts that end in an escape character, but ignore
666
// the final part. We've already detected escape chars at the
667
// end of the final part in parseLine() and handled multi-lines
669
if( a < parts.size() - 1 &&
670
doesStringEndInAnEscapeChar( str ) )
672
// join the next part to this part and remove the next part
673
parts.set( a, str.substring( 0, str.length() - 1 ) +
674
';' + parts.get( a + 1 ) );
675
parts.remove( a + 1 );
677
// re-visit this part
682
// trim and replace string
687
String[] ret = new String[ parts.size() ];
688
return parts.toArray( ret );
513
if( name_and_param_bits[ 0 ].equals( "N" ) )
514
parseN( name_and_param_bits, complete_value );
515
else if( name_and_param_bits[ 0 ].equals( "FN" ) )
516
parseFN( name_and_param_bits, complete_value );
517
else if( name_and_param_bits[ 0 ].equals( "ORG" ) )
518
parseORG( name_and_param_bits, complete_value );
519
else if( name_and_param_bits[ 0 ].equals( "TEL" ) )
520
parseTEL( name_and_param_bits, complete_value );
521
else if( name_and_param_bits[ 0 ].equals( "EMAIL" ) )
522
parseEMAIL( name_and_param_bits, complete_value );
691
526
private void parseN( String[] params, String value )
527
throws ParseException, SkipContactException,
693
530
// already got a better name?
694
531
if( _name_level >= NAMELEVEL_N ) return;
696
533
// get name parts
697
String[] name_parts = splitValueBySemicolon( value );
534
String[] nameparts = value.split( ";" );
535
for( int i = 0; i < nameparts.length; i++ )
536
nameparts[ i ] = nameparts[ i ].trim();
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 ];
540
if( nameparts.length > 1 && nameparts[ 1 ].length() > 0 )
541
value += nameparts[ 1 ];
542
if( nameparts[ 0 ].length() > 0 )
543
value += ( value.length() == 0? "" : " " ) + nameparts[ 0 ];
707
546
setName( value );
708
547
_name_level = NAMELEVEL_N;
549
// check now to see if we need to import this contact (to avoid
550
// parsing the rest of the vCard unnecessarily)
551
if( !isImportRequired( getName() ) )
552
throw new SkipContactException();
711
555
private void parseFN( String[] params, String value )
556
throws ParseException, SkipContactException
713
558
// already got a better name?
714
559
if( _name_level >= NAMELEVEL_FN ) return;
721
566
private void parseORG( String[] params, String value )
567
throws ParseException, SkipContactException
569
// already got a better name?
570
if( _name_level >= NAMELEVEL_ORG ) return;
724
String[] org_parts = splitValueBySemicolon( value );
725
if( org_parts == null || org_parts.length < 1 ) return;
727
// build organisation name
728
StringBuilder builder = new StringBuilder(
729
String.valueOf( org_parts[ 0 ] ) );
730
for( int a = 1; a < org_parts.length; a++ )
731
builder.append( ", " ).append( org_parts[ a ] );
732
String organisation = builder.toString();
734
// set organisation name (using a title we've previously found)
735
addOrganisation( organisation, _cached_title, true );
737
// if we've not previously found a title, store this organisation
738
// name (we'll need it when we find a title to update the
739
// organisation, by name), else if we *have* previously found a
740
// title, clear it (since we just used it)
741
if( _cached_title == null )
742
_cached_organisation = organisation;
744
_cached_title = null;
747
private void parseTITLE( String[] params, String value )
749
// if we previously had an organisation, look it up and append this
751
if( _cached_organisation != null && hasOrganisations() ) {
752
HashMap< String, ExtraDetail > datas = getOrganisations();
753
ExtraDetail detail = datas.get( _cached_organisation );
755
detail.setExtra( value );
758
// same as when handling organisation, if we've not previously found
759
// an organisation we store this title, else we clear it (since we
760
// just appended this title to it)
761
if( _cached_organisation == null )
762
_cached_title = value;
764
_cached_organisation = null;
573
String[] orgparts = value.split( ";" );
574
for( int i = 0; i < orgparts.length; i++ )
575
orgparts[ i ] = orgparts[ i ].trim();
578
if( orgparts[ 0 ].length() == 0 && orgparts.length > 1 )
579
value = orgparts[ 1 ];
581
value = orgparts[ 0 ];
585
_name_level = NAMELEVEL_ORG;
767
588
private void parseTEL( String[] params, String value )
589
throws ParseException
769
591
if( value.length() == 0 ) return;
771
593
Set< String > types = extractTypes( params, Arrays.asList(
772
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
773
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
594
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
595
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
775
597
// here's the logic...
776
598
boolean preferred = types.contains( "PREF" );
599
int type = PhonesColumns.TYPE_MOBILE;
600
if( types.contains( "VOICE" ) )
601
if( types.contains( "WORK" ) )
602
type = PhonesColumns.TYPE_WORK;
604
type = PhonesColumns.TYPE_HOME;
605
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
606
type = PhonesColumns.TYPE_MOBILE;
778
607
if( types.contains( "FAX" ) )
779
608
if( types.contains( "HOME" ) )
780
609
type = PhonesColumns.TYPE_FAX_HOME;
782
611
type = PhonesColumns.TYPE_FAX_WORK;
783
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
784
type = PhonesColumns.TYPE_MOBILE;
785
else if( types.contains( "PAGER" ) )
612
if( types.contains( "PAGER" ) )
786
613
type = PhonesColumns.TYPE_PAGER;
787
else if( types.contains( "WORK" ) )
788
type = PhonesColumns.TYPE_WORK;
790
type = PhonesColumns.TYPE_HOME;
792
615
// add phone number
793
addNumber( value, type, preferred );
616
addPhone( value, type, preferred );
796
619
public void parseEMAIL( String[] params, String value )
620
throws ParseException
798
622
if( value.length() == 0 ) return;
800
624
Set< String > types = extractTypes( params, Arrays.asList(
801
"PREF", "WORK", "HOME", "INTERNET" ) );
625
"PREF", "WORK", "HOME", "INTERNET" ) );
627
// here's the logic...
804
628
boolean preferred = types.contains( "PREF" );
806
if( types.contains( "WORK" ) )
807
type = Contacts.ContactMethods.TYPE_WORK;
809
type = Contacts.ContactMethods.TYPE_HOME;
811
addEmail( value, type, preferred );
814
private void parseADR( String[] params, String value )
817
String[] adr_parts = splitValueBySemicolon( value );
821
for( int a = 0; a < adr_parts.length; a++ ) {
822
if( value.length() > 0 ) value += "\n";
823
value += adr_parts[ a ].trim();
826
Set< String > types = extractTypes( params, Arrays.asList(
827
"PREF", "WORK", "HOME", "INTERNET" ) );
831
if( types.contains( "WORK" ) )
832
type = Contacts.ContactMethods.TYPE_WORK;
834
type = Contacts.ContactMethods.TYPE_HOME;
836
addAddress( value, type );
629
if( types.contains( "WORK" ) )
630
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
632
addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
839
635
public void finaliseParsing()
840
throws ParseException, SkipContactException,
636
throws ParseException, SkipContactException,
843
639
// missing version (and data is present)
844
640
if( _version == null && _buffers != null )
845
641
throw new ParseException( R.string.error_vcf_malformed );
847
// check if we should import this contact
849
if( !isImportRequired( this ) )
850
throw new SkipContactException();
852
catch( ContactNeedsMoreInfoException e ) {
853
throw new ParseException( R.string.error_vcf_notenoughinfo );
643
// missing name properties?
644
if( _name_level == NAMELEVEL_NONE )
645
throw new ParseException( R.string.error_vcf_noname );
647
// check if we should import this one? If we've already got an 'N'-
648
// type name, this will already have been done by parseN() so we
649
// mustn't do this here (or it could prompt twice!)
650
if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
651
throw new SkipContactException();
857
654
private String checkParam( String[] params, String name )
859
Pattern p = Pattern.compile(
860
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
656
Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
861
657
for( int i = 0; i < params.length; i++ ) {
862
658
Matcher m = p.matcher( params[ i ] );
863
659
if( m.matches() )
869
665
private Set< String > extractTypes( String[] params,
870
List< String > valid_types )
666
List< String > validTypes )
872
668
HashSet< String > types = new HashSet< String >();
874
670
// get 3.0-style 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 ] );
672
if( ( typeParam = checkParam( params, "TYPE" ) ) != null ) {
673
String[] bits = typeParam.split( "," );
674
for( int i = 0; i < bits.length; i++ )
675
if( validTypes.contains( bits[ i ] ) )
676
types.add( bits[ i ] );
883
679
// get 2.1-style type param
884
680
if( _version.equals( "2.1" ) ) {
885
681
for( int i = 1; i < params.length; i++ )
886
if( valid_types.contains( params[ i ] ) )
682
if( validTypes.contains( params[ i ] ) )
887
683
types.add( params[ i ] );