136
136
if( !in_vcard ) {
137
137
// look for vcard beginning
138
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD.*$" ) ) {
138
if( line.matches( "^BEGIN:VCARD" ) ) {
143
else if( line.matches( "^END[ \t]*:[ \t]*VCARD.*$" ) )
143
else if( line.matches( "^END:VCARD" ) )
144
144
in_vcard = false;
176
176
importVCardFileContent( content, file.getName() );
178
catch( OutOfMemoryError e ) {
179
showError( R.string.error_outofmemory );
181
178
catch( FileNotFoundException e ) {
182
179
showError( getText( R.string.error_filenotfound ) +
183
180
file.getName() );
196
193
ContentLineIterator cli = new ContentLineIterator( content );
197
194
while( cli.hasNext() )
199
ContentLine content_line = cli.next();
196
ByteBuffer buffer = cli.next();
201
// get a US-ASCII version of the string, for processing
202
String line = content_line.getUsAsciiLine();
198
// get a US-ASCII version of the line for processing
201
line = new String( buffer.array(), buffer.position(),
202
buffer.limit() - buffer.position(), "US-ASCII" );
204
catch( UnsupportedEncodingException e ) {
205
// we know US-ASCII is supported, so appease the compiler...
204
209
if( vcard == null ) {
205
210
// look for vcard beginning
206
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD.*$" ) ) {
211
if( line.matches( "^BEGIN:VCARD" ) ) {
207
212
setProgress( _progress++ );
208
213
vcard = new Vcard();
209
214
vcard_start_line = cli.getLineNumber();
255
260
// try giving the line to the vcard
257
vcard.parseLine( content_line );
262
vcard.parseLine( buffer, line,
263
cli.doesNextLineLookFolded() );
259
265
catch( Vcard.ParseException e ) {
288
private ByteBuffer _buffer;
289
private boolean _folded_next;
290
private String _line;
292
public ContentLine( ByteBuffer buffer, boolean folded_next )
295
_folded_next = folded_next;
299
public ByteBuffer getBuffer()
304
public boolean doesNextLineLookFolded()
309
public String getUsAsciiLine()
311
// generated line and cache it
312
if( _line == null ) {
314
_line = new String( _buffer.array(), _buffer.position(),
315
_buffer.limit() - _buffer.position(), "US-ASCII" );
317
catch( UnsupportedEncodingException e ) {
318
// we know US-ASCII *is* supported, so appease the
323
// return cached line
328
class ContentLineIterator implements Iterator< ContentLine >
292
class ContentLineIterator implements Iterator< ByteBuffer >
330
294
protected byte[] _content = null;
331
295
protected int _pos = 0;
356
320
_pos > initial_pos )? _pos - 1 : _pos;
359
return new ContentLine(
360
ByteBuffer.wrap( _content, initial_pos,
362
doesNextLineLookFolded() );
323
return ByteBuffer.wrap( _content, initial_pos,
365
327
// we didn't find one, but were there bytes left?
370
return new ContentLine(
371
ByteBuffer.wrap( _content, initial_pos,
373
doesNextLineLookFolded() );
332
return ByteBuffer.wrap( _content, initial_pos,
388
348
* onto the end of this one?
391
private boolean doesNextLineLookFolded()
351
public boolean doesNextLineLookFolded()
393
353
return _pos > 0 && _pos < _content.length &&
394
_content[ _pos - 1 ] == '\n' &&
395
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
354
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
398
357
public int getLineNumber()
410
369
private final static int MULTILINE_NONE = 0;
411
370
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
412
371
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
413
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
372
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
415
374
private String _version = null;
416
private Vector< ContentLine > _content_lines = null;
375
private Vector< ByteBuffer > _buffers = null;
417
376
private int _name_level = NAMELEVEL_NONE;
418
377
private int _parser_multiline_state = MULTILINE_NONE;
419
378
private String _parser_current_name_and_params = null;
462
421
@SuppressWarnings("serial")
463
422
protected class SkipImportException extends Exception { }
465
private String extractCollonPartFromLine( ContentLine content_line,
424
private String extractCollonPartFromLine( ByteBuffer buffer,
425
String line, boolean former )
468
427
String ret = null;
429
// get a US-ASCII version of the line for processing, unless we were
433
line = new String( buffer.array(), buffer.position(),
434
buffer.limit() - buffer.position(), "US-ASCII" );
436
catch( UnsupportedEncodingException e ) {
437
// we know US-ASCII is supported, so appease the compiler...
470
442
// split line into name and value parts and check to make sure we
471
443
// only got 2 parts and that the first part is not zero in length
472
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
444
String[] parts = line.split( ":", 2 );
473
445
if( parts.length == 2 && parts[ 0 ].length() > 0 )
474
446
ret = parts[ former? 0 : 1 ];
479
private String extractNameAndParamsFromLine( ContentLine content_line )
481
return extractCollonPartFromLine( content_line, true ).trim();
484
private String extractValueFromLine( ContentLine content_line )
486
return extractCollonPartFromLine( content_line, false );
489
public void parseLine( ContentLine content_line )
451
private String extractNameAndParamsFromLine( ByteBuffer buffer,
454
return extractCollonPartFromLine( buffer, line, true );
457
private String extractValueFromLine( ByteBuffer buffer, String line )
459
return extractCollonPartFromLine( buffer, line, false );
462
public void parseLine( ByteBuffer buffer, String line,
463
boolean next_line_looks_folded )
490
464
throws ParseException, SkipImportException,
491
465
AbortImportException
496
470
// tentatively get name and params from line
497
471
String name_and_params =
498
extractNameAndParamsFromLine( content_line );
472
extractNameAndParamsFromLine( buffer, line );
500
474
// is it a version line?
501
475
if( name_and_params != null &&
502
name_and_params.equalsIgnoreCase( "VERSION" ) )
476
name_and_params.equals( "VERSION" ) )
505
String value = extractValueFromLine( content_line ).trim();
479
String value = extractValueFromLine( buffer, line );
506
480
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
507
481
throw new ParseException( R.string.error_vcf_version );
508
482
_version = value;
510
484
// parse any buffers we've been accumulating while we waited
512
if( _content_lines != null )
513
for( int i = 0; i < _content_lines.size(); i++ )
514
parseLine( _content_lines.get( i ) );
515
_content_lines = null;
486
if( _buffers != null )
487
for( int i = 0; i < _buffers.size(); i++ )
488
parseLine( _buffers.get( i ), null,
489
i + 1 < _buffers.size() &&
490
_buffers.get( i + 1 ).hasRemaining() &&
491
_buffers.get( i + 1 ).get(
492
_buffers.get( i + 1 ).position() ) == ' ' );
519
497
// no, so stash this line till we get a version
520
if( _content_lines == null )
521
_content_lines = new Vector< ContentLine >();
522
_content_lines.add( content_line );
498
if( _buffers == null )
499
_buffers = new Vector< ByteBuffer >();
500
_buffers.add( buffer );
527
505
// name and params and the position in the buffer where the
528
// "value" part of the line starts
506
// "value" part of the line start
529
507
String name_and_params;
538
516
// skip some initial line characters, depending on the type
539
517
// of multi-line we're handling
540
pos = content_line.getBuffer().position();
518
pos = buffer.position();
541
519
switch( _parser_multiline_state )
543
521
case MULTILINE_FOLDED:
546
524
case MULTILINE_ENCODED:
547
while( pos < content_line.getBuffer().limit() && (
548
content_line.getBuffer().get( pos ) == ' ' ||
549
content_line.getBuffer().get( pos ) == '\t' ) )
525
while( pos < buffer.limit() && (
526
buffer.get( pos ) == ' ' ||
527
buffer.get( pos ) == '\t' ) )
565
if( content_line.getUsAsciiLine().trim().length() == 0 )
568
542
// get name and params from line, and since we're not
569
543
// parsing a subsequent line in a multi-line, this should
570
544
// not fail, or it's an error
571
545
name_and_params =
572
extractNameAndParamsFromLine( content_line );
546
extractNameAndParamsFromLine( buffer, line );
573
547
if( name_and_params == null )
574
548
throw new ParseException(
575
549
R.string.error_vcf_malformed );
577
551
// calculate how many chars to skip from beginning of line
578
552
// so we skip the property "name:" part
579
pos = content_line.getBuffer().position() +
580
name_and_params.length() + 1;
553
pos = buffer.position() + name_and_params.length() + 1;
582
555
// reset the saved multi-line state
583
556
_parser_current_name_and_params = name_and_params;
587
560
// get value from buffer, as raw bytes
588
561
ByteBuffer value;
589
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
590
content_line.getBuffer().limit() - pos );
562
value = ByteBuffer.wrap( buffer.array(), pos,
563
buffer.limit() - pos );
592
565
// get parameter parts
593
566
String[] name_param_parts = name_and_params.split( ";", -1 );
600
573
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
602
575
boolean is_interesting_field =
603
interesting_fields.contains(
604
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
576
interesting_fields.contains( name_param_parts[ 0 ] );
606
578
// parse encoding parameter
607
579
String encoding = checkParam( name_param_parts, "ENCODING" );
608
580
if( encoding != null )
609
581
encoding = encoding.toUpperCase( Locale.US );
610
582
if( is_interesting_field && encoding != null &&
611
!encoding.equalsIgnoreCase( "8BIT" ) &&
612
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
613
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
583
!encoding.equals( "8BIT" ) &&
584
!encoding.equals( "QUOTED-PRINTABLE" ) )
585
//&& !encoding.equals( "BASE64" ) )
615
587
throw new ParseException( R.string.error_vcf_encoding );
620
592
if( charset != null )
621
593
charset = charset.toUpperCase( Locale.US );
622
594
if( charset != null &&
623
!charset.equalsIgnoreCase( "US-ASCII" ) &&
624
!charset.equalsIgnoreCase( "ASCII" ) &&
625
!charset.equalsIgnoreCase( "UTF-8" ) )
595
!charset.equals( "US-ASCII" ) &&
596
!charset.equals( "ASCII" ) &&
597
!charset.equals( "UTF-8" ) )
627
599
throw new ParseException( R.string.error_vcf_charset );
630
602
// do unencoding (or default to a fake unencoding result with
631
603
// the raw string)
632
604
UnencodeResult unencoding_result = null;
633
if( encoding != null &&
634
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
605
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
636
606
unencoding_result = unencodeQuotedPrintable( value );
638
// else if( encoding != null &&
639
// encoding.equalsIgnoreCase( "BASE64" ) )
607
// else if( encoding != null && encoding.equals( "BASE64" ) )
641
608
// unencoding_result = unencodeBase64( props[ 1 ], charset );
643
609
if( unencoding_result != null ) {
644
610
value = unencoding_result.getBuffer();
645
611
if( unencoding_result.isAnotherLineRequired() )
650
616
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
651
617
if( ( charset == null && _version.equals( "2.1" ) ) ||
652
618
( charset != null && (
653
charset.equalsIgnoreCase( "ASCII" ) ||
654
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
619
charset.equals( "ASCII" ) ||
620
charset.equals( "US-ASCII" ) ) ) )
656
622
value = transcodeAsciiToUtf8( value );
668
634
// for some entries that have semicolon-separated value parts,
669
635
// check to see if the value ends in an escape character, which
670
636
// indicates that we have a multi-line value
671
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
672
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
673
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
637
if( ( name_param_parts[ 0 ].equals( "N" ) ||
638
name_param_parts[ 0 ].equals( "ORG" ) ||
639
name_param_parts[ 0 ].equals( "ADR" ) ) &&
674
640
doesStringEndInAnEscapeChar( string_value ) )
676
642
_parser_multiline_state = MULTILINE_ESCAPED;
678
644
string_value.length() - 1 );
681
// if we know we're not in an encoding-based multi-line, check
682
// to see if we're in a folded multi-line
647
// now we know whether we're in an encoding multi-line,
648
// determine if we're in a v3 folded multi-line or not
683
649
if( _parser_multiline_state == MULTILINE_NONE &&
684
content_line.doesNextLineLookFolded() )
650
_version.equals( "3.0" ) && next_line_looks_folded )
686
652
_parser_multiline_state = MULTILINE_FOLDED;
699
665
if( complete_value.length() < 1 ) return;
701
667
// parse some properties
702
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
668
if( name_param_parts[ 0 ].equals( "N" ) )
703
669
parseN( name_param_parts, complete_value );
704
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
670
else if( name_param_parts[ 0 ].equals( "FN" ) )
705
671
parseFN( name_param_parts, complete_value );
706
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
672
else if( name_param_parts[ 0 ].equals( "ORG" ) )
707
673
parseORG( name_param_parts, complete_value );
708
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
674
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
709
675
parseTITLE( name_param_parts, complete_value );
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
676
else if( name_param_parts[ 0 ].equals( "TEL" ) )
711
677
parseTEL( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
678
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
713
679
parseEMAIL( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
680
else if( name_param_parts[ 0 ].equals( "ADR" ) )
715
681
parseADR( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
682
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
717
683
parseLABEL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
719
parseNOTE( name_param_parts, complete_value );
843
801
for( int b = 0; b < name_part_parts.length; b++ )
844
802
if( name_part_parts[ b ].length() > 0 )
846
if( value.length() > 0 ) value += " ";
804
if( value.length() == 0 ) value += " ";
847
805
value += name_part_parts[ b ];
968
926
for( int a = 0; a < adr_parts.length; a++ )
969
927
if( adr_parts[ a ].length() > 0 )
971
// version 3.0 vCards allow further splitting by comma
972
if( _version.equals( "3.0" ) )
974
// split this part in to it's comma-separated bits and
975
// add them on individual lines
976
String[] adr_part_parts =
977
splitValueByCharacter( adr_parts[ a ], ',' );
978
for( int b = 0; b < adr_part_parts.length; b++ )
979
if( adr_part_parts[ b ].length() > 0 )
981
if( value.length() > 0 ) value += "\n";
982
value += adr_part_parts[ b ];
987
// add this part on an individual line
988
if( value.length() > 0 ) value += "\n";
989
value += adr_parts[ a ];
929
// split this part in to it's comma-separated bits
930
String[] adr_part_parts =
931
splitValueByCharacter( adr_parts[ a ], ',' );
932
for( int b = 0; b < adr_part_parts.length; b++ )
933
if( adr_part_parts[ b ].length() > 0 )
935
if( value.length() > 0 ) value += "\n";
936
value += adr_part_parts[ b ];
993
940
Set< String > types = extractTypes( params, Arrays.asList(
1018
965
addAddress( unescapeValue( value ), type );
1021
private void parseNOTE( String[] params, String value )
1023
addNote( unescapeValue( value ) );
1026
968
public void finaliseVcard()
1027
969
throws ParseException, ContactNotIdentifiableException
1029
971
// missing version (and data is present)
1030
if( _version == null && _content_lines != null )
972
if( _version == null && _buffers != null )
1031
973
throw new ParseException( R.string.error_vcf_malformed );
1033
975
// finalise the parent class
1058
1000
HashSet< String > ret = new HashSet< String >();
1060
1002
Pattern p = Pattern.compile(
1061
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1062
Pattern.CASE_INSENSITIVE );
1003
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1063
1004
for( int i = 0; i < params.length; i++ ) {
1064
1005
Matcher m = p.matcher( params[ i ] );
1065
1006
if( m.matches() )
1073
1014
* Amongst the params, return any type values present. For v2.1 vCards,
1074
1015
* those types are just parameters. For v3.0, they are prefixed with
1075
1016
* "TYPE=". There may also be multiple type parameters.
1076
* @param params an array of params to look for types in
1077
* @param valid_types an list of upper-case type values to look for
1018
* @param a list of type values to look for
1078
1019
* @return a set of present type values
1080
1021
private Set< String > extractTypes( String[] params,
1086
1027
String type_params[] = checkParams( params, "TYPE" );
1087
1028
for( int a = 0; a < type_params.length; a++ )
1089
// check for a comma-separated list of types (why? I don't think
1090
// this is in the specs!)
1030
// check for a comma-separated list of types (why? this isn't in
1091
1032
String[] parts = type_params[ a ].split( "," );
1092
for( int i = 0; i < parts.length; i++ ) {
1093
String ucpart = parts[ i ].toUpperCase( Locale.US );
1094
if( valid_types.contains( ucpart ) )
1095
types.add( ucpart );
1033
for( int i = 0; i < parts.length; i++ )
1034
if( valid_types.contains( parts[ i ] ) )
1035
types.add( parts[ i ] );
1099
1038
// get 2.1-style type param
1100
1039
if( _version.equals( "2.1" ) ) {
1101
for( int i = 1; i < params.length; i++ ) {
1102
String ucparam = params[ i ].toUpperCase( Locale.US );
1103
if( valid_types.contains( ucparam ) )
1104
types.add( ucparam );
1040
for( int i = 1; i < params.length; i++ )
1041
if( valid_types.contains( params[ i ] ) )
1042
types.add( params[ i ] );