133
133
boolean in_vcard = false;
134
134
while( ( line = reader.readLine() ) != null )
137
138
// look for vcard beginning
138
if( line.matches( "^BEGIN:VCARD" ) ) {
139
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
143
// check for vMsg files
144
else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
145
showError( getText( R.string.error_vcf_vmsgfile )
143
else if( line.matches( "^END:VCARD" ) )
149
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
144
150
in_vcard = false;
176
182
importVCardFileContent( content, file.getName() );
184
catch( OutOfMemoryError e ) {
185
showError( R.string.error_outofmemory );
178
187
catch( FileNotFoundException e ) {
179
188
showError( getText( R.string.error_filenotfound ) +
180
189
file.getName() );
193
202
ContentLineIterator cli = new ContentLineIterator( content );
194
203
while( cli.hasNext() )
196
ByteBuffer buffer = cli.next();
205
ContentLine content_line = cli.next();
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...
207
// get a US-ASCII version of the string, for processing
208
String line = content_line.getUsAsciiLine();
209
210
if( vcard == null ) {
210
211
// look for vcard beginning
211
if( line.matches( "^BEGIN:VCARD" ) ) {
212
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
212
213
setProgress( _progress++ );
213
214
vcard = new Vcard();
214
215
vcard_start_line = cli.getLineNumber();
260
261
// try giving the line to the vcard
262
vcard.parseLine( buffer, line,
263
cli.doesNextLineLookFolded() );
263
vcard.parseLine( content_line );
265
265
catch( Vcard.ParseException e ) {
292
class ContentLineIterator implements Iterator< ByteBuffer >
294
private ByteBuffer _buffer;
295
private boolean _folded_next;
296
private String _line;
298
public ContentLine( ByteBuffer buffer, boolean folded_next )
301
_folded_next = folded_next;
305
public ByteBuffer getBuffer()
310
public boolean doesNextLineLookFolded()
315
public String getUsAsciiLine()
317
// generated line and cache it
318
if( _line == null ) {
320
_line = new String( _buffer.array(), _buffer.position(),
321
_buffer.limit() - _buffer.position(), "US-ASCII" );
323
catch( UnsupportedEncodingException e ) {
324
// we know US-ASCII *is* supported, so appease the
329
// return cached line
334
class ContentLineIterator implements Iterator< ContentLine >
294
336
protected byte[] _content = null;
295
337
protected int _pos = 0;
320
362
_pos > initial_pos )? _pos - 1 : _pos;
323
return ByteBuffer.wrap( _content, initial_pos,
365
return new ContentLine(
366
ByteBuffer.wrap( _content, initial_pos,
368
doesNextLineLookFolded() );
327
371
// we didn't find one, but were there bytes left?
332
return ByteBuffer.wrap( _content, initial_pos,
376
return new ContentLine(
377
ByteBuffer.wrap( _content, initial_pos,
379
doesNextLineLookFolded() );
348
394
* onto the end of this one?
351
public boolean doesNextLineLookFolded()
397
private boolean doesNextLineLookFolded()
353
399
return _pos > 0 && _pos < _content.length &&
354
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
400
_content[ _pos - 1 ] == '\n' &&
401
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
357
404
public int getLineNumber()
369
416
private final static int MULTILINE_NONE = 0;
370
417
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
371
418
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
372
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
419
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
374
421
private String _version = null;
375
private Vector< ByteBuffer > _buffers = null;
422
private Vector< ContentLine > _content_lines = null;
376
423
private int _name_level = NAMELEVEL_NONE;
377
424
private int _parser_multiline_state = MULTILINE_NONE;
378
425
private String _parser_current_name_and_params = null;
421
468
@SuppressWarnings("serial")
422
469
protected class SkipImportException extends Exception { }
424
private String extractCollonPartFromLine( ByteBuffer buffer,
425
String line, boolean former )
471
private String extractCollonPartFromLine( ContentLine content_line,
427
474
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...
442
476
// split line into name and value parts and check to make sure we
443
477
// only got 2 parts and that the first part is not zero in length
444
String[] parts = line.split( ":", 2 );
478
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
445
479
if( parts.length == 2 && parts[ 0 ].length() > 0 )
446
480
ret = parts[ former? 0 : 1 ];
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 )
485
private String extractNameAndParamsFromLine( ContentLine content_line )
487
return extractCollonPartFromLine( content_line, true ).trim();
490
private String extractValueFromLine( ContentLine content_line )
492
return extractCollonPartFromLine( content_line, false );
495
public void parseLine( ContentLine content_line )
464
496
throws ParseException, SkipImportException,
465
497
AbortImportException
470
502
// tentatively get name and params from line
471
503
String name_and_params =
472
extractNameAndParamsFromLine( buffer, line );
504
extractNameAndParamsFromLine( content_line );
474
506
// is it a version line?
475
507
if( name_and_params != null &&
476
name_and_params.equals( "VERSION" ) )
508
name_and_params.equalsIgnoreCase( "VERSION" ) )
479
String value = extractValueFromLine( buffer, line );
511
String value = extractValueFromLine( content_line ).trim();
480
512
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
481
513
throw new ParseException( R.string.error_vcf_version );
482
514
_version = value;
484
516
// parse any buffers we've been accumulating while we waited
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() ) == ' ' );
518
if( _content_lines != null )
519
for( int i = 0; i < _content_lines.size(); i++ )
520
parseLine( _content_lines.get( i ) );
521
_content_lines = null;
497
525
// no, so stash this line till we get a version
498
if( _buffers == null )
499
_buffers = new Vector< ByteBuffer >();
500
_buffers.add( buffer );
526
if( _content_lines == null )
527
_content_lines = new Vector< ContentLine >();
528
_content_lines.add( content_line );
505
533
// name and params and the position in the buffer where the
506
// "value" part of the line start
534
// "value" part of the line starts
507
535
String name_and_params;
516
544
// skip some initial line characters, depending on the type
517
545
// of multi-line we're handling
518
pos = buffer.position();
546
pos = content_line.getBuffer().position();
519
547
switch( _parser_multiline_state )
521
549
case MULTILINE_FOLDED:
524
552
case MULTILINE_ENCODED:
525
while( pos < buffer.limit() && (
526
buffer.get( pos ) == ' ' ||
527
buffer.get( pos ) == '\t' ) )
553
while( pos < content_line.getBuffer().limit() && (
554
content_line.getBuffer().get( pos ) == ' ' ||
555
content_line.getBuffer().get( pos ) == '\t' ) )
571
if( content_line.getUsAsciiLine().trim().length() == 0 )
542
574
// get name and params from line, and since we're not
543
575
// parsing a subsequent line in a multi-line, this should
544
576
// not fail, or it's an error
545
577
name_and_params =
546
extractNameAndParamsFromLine( buffer, line );
578
extractNameAndParamsFromLine( content_line );
547
579
if( name_and_params == null )
548
580
throw new ParseException(
549
581
R.string.error_vcf_malformed );
551
583
// calculate how many chars to skip from beginning of line
552
584
// so we skip the property "name:" part
553
pos = buffer.position() + name_and_params.length() + 1;
585
pos = content_line.getBuffer().position() +
586
name_and_params.length() + 1;
555
588
// reset the saved multi-line state
556
589
_parser_current_name_and_params = name_and_params;
560
593
// get value from buffer, as raw bytes
561
594
ByteBuffer value;
562
value = ByteBuffer.wrap( buffer.array(), pos,
563
buffer.limit() - pos );
595
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
596
content_line.getBuffer().limit() - pos );
565
598
// get parameter parts
566
599
String[] name_param_parts = name_and_params.split( ";", -1 );
573
606
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
575
608
boolean is_interesting_field =
576
interesting_fields.contains( name_param_parts[ 0 ] );
609
interesting_fields.contains(
610
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
578
612
// parse encoding parameter
579
613
String encoding = checkParam( name_param_parts, "ENCODING" );
580
614
if( encoding != null )
581
615
encoding = encoding.toUpperCase( Locale.US );
582
616
if( is_interesting_field && encoding != null &&
583
!encoding.equals( "8BIT" ) &&
584
!encoding.equals( "QUOTED-PRINTABLE" ) )
585
//&& !encoding.equals( "BASE64" ) )
617
!encoding.equalsIgnoreCase( "8BIT" ) &&
618
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
619
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
587
621
throw new ParseException( R.string.error_vcf_encoding );
592
626
if( charset != null )
593
627
charset = charset.toUpperCase( Locale.US );
594
628
if( charset != null &&
595
!charset.equals( "US-ASCII" ) &&
596
!charset.equals( "ASCII" ) &&
597
!charset.equals( "UTF-8" ) )
629
!charset.equalsIgnoreCase( "US-ASCII" ) &&
630
!charset.equalsIgnoreCase( "ASCII" ) &&
631
!charset.equalsIgnoreCase( "UTF-8" ) )
599
633
throw new ParseException( R.string.error_vcf_charset );
602
636
// do unencoding (or default to a fake unencoding result with
603
637
// the raw string)
604
638
UnencodeResult unencoding_result = null;
605
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
639
if( encoding != null &&
640
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
606
642
unencoding_result = unencodeQuotedPrintable( value );
607
// else if( encoding != null && encoding.equals( "BASE64" ) )
644
// else if( encoding != null &&
645
// encoding.equalsIgnoreCase( "BASE64" ) )
608
647
// unencoding_result = unencodeBase64( props[ 1 ], charset );
609
649
if( unencoding_result != null ) {
610
650
value = unencoding_result.getBuffer();
611
651
if( unencoding_result.isAnotherLineRequired() )
616
656
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
617
657
if( ( charset == null && _version.equals( "2.1" ) ) ||
618
658
( charset != null && (
619
charset.equals( "ASCII" ) ||
620
charset.equals( "US-ASCII" ) ) ) )
659
charset.equalsIgnoreCase( "ASCII" ) ||
660
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
622
662
value = transcodeAsciiToUtf8( value );
634
674
// for some entries that have semicolon-separated value parts,
635
675
// check to see if the value ends in an escape character, which
636
676
// indicates that we have a multi-line value
637
if( ( name_param_parts[ 0 ].equals( "N" ) ||
638
name_param_parts[ 0 ].equals( "ORG" ) ||
639
name_param_parts[ 0 ].equals( "ADR" ) ) &&
677
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
678
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
679
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
640
680
doesStringEndInAnEscapeChar( string_value ) )
642
682
_parser_multiline_state = MULTILINE_ESCAPED;
644
684
string_value.length() - 1 );
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
687
// if we know we're not in an encoding-based multi-line, check
688
// to see if we're in a folded multi-line
649
689
if( _parser_multiline_state == MULTILINE_NONE &&
650
_version.equals( "3.0" ) && next_line_looks_folded )
690
content_line.doesNextLineLookFolded() )
652
692
_parser_multiline_state = MULTILINE_FOLDED;
665
705
if( complete_value.length() < 1 ) return;
667
707
// parse some properties
668
if( name_param_parts[ 0 ].equals( "N" ) )
708
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
669
709
parseN( name_param_parts, complete_value );
670
else if( name_param_parts[ 0 ].equals( "FN" ) )
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
671
711
parseFN( name_param_parts, complete_value );
672
else if( name_param_parts[ 0 ].equals( "ORG" ) )
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
673
713
parseORG( name_param_parts, complete_value );
674
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
675
715
parseTITLE( name_param_parts, complete_value );
676
else if( name_param_parts[ 0 ].equals( "TEL" ) )
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
677
717
parseTEL( name_param_parts, complete_value );
678
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
679
719
parseEMAIL( name_param_parts, complete_value );
680
else if( name_param_parts[ 0 ].equals( "ADR" ) )
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
681
721
parseADR( name_param_parts, complete_value );
682
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
683
723
parseLABEL( name_param_parts, complete_value );
684
else if( name_param_parts[ 0 ].equals( "NOTE" ) )
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
685
725
parseNOTE( name_param_parts, complete_value );
809
849
for( int b = 0; b < name_part_parts.length; b++ )
810
850
if( name_part_parts[ b ].length() > 0 )
812
if( value.length() == 0 ) value += " ";
852
if( value.length() > 0 ) value += " ";
813
853
value += name_part_parts[ b ];
934
974
for( int a = 0; a < adr_parts.length; a++ )
935
975
if( adr_parts[ a ].length() > 0 )
937
// split this part in to it's comma-separated bits
938
String[] adr_part_parts =
939
splitValueByCharacter( adr_parts[ a ], ',' );
940
for( int b = 0; b < adr_part_parts.length; b++ )
941
if( adr_part_parts[ b ].length() > 0 )
943
if( value.length() > 0 ) value += "\n";
944
value += adr_part_parts[ b ];
977
// version 3.0 vCards allow further splitting by comma
978
if( _version.equals( "3.0" ) )
980
// split this part in to it's comma-separated bits and
981
// add them on individual lines
982
String[] adr_part_parts =
983
splitValueByCharacter( adr_parts[ a ], ',' );
984
for( int b = 0; b < adr_part_parts.length; b++ )
985
if( adr_part_parts[ b ].length() > 0 )
987
if( value.length() > 0 ) value += "\n";
988
value += adr_part_parts[ b ];
993
// add this part on an individual line
994
if( value.length() > 0 ) value += "\n";
995
value += adr_parts[ a ];
948
999
Set< String > types = extractTypes( params, Arrays.asList(
982
1033
throws ParseException, ContactNotIdentifiableException
984
1035
// missing version (and data is present)
985
if( _version == null && _buffers != null )
1036
if( _version == null && _content_lines != null )
986
1037
throw new ParseException( R.string.error_vcf_malformed );
988
1039
// finalise the parent class
1013
1064
HashSet< String > ret = new HashSet< String >();
1015
1066
Pattern p = Pattern.compile(
1016
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1067
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1068
Pattern.CASE_INSENSITIVE );
1017
1069
for( int i = 0; i < params.length; i++ ) {
1018
1070
Matcher m = p.matcher( params[ i ] );
1019
1071
if( m.matches() )
1027
1079
* Amongst the params, return any type values present. For v2.1 vCards,
1028
1080
* those types are just parameters. For v3.0, they are prefixed with
1029
1081
* "TYPE=". There may also be multiple type parameters.
1031
* @param a list of type values to look for
1082
* @param params an array of params to look for types in
1083
* @param valid_types an list of upper-case type values to look for
1032
1084
* @return a set of present type values
1034
1086
private Set< String > extractTypes( String[] params,
1040
1092
String type_params[] = checkParams( params, "TYPE" );
1041
1093
for( int a = 0; a < type_params.length; a++ )
1043
// check for a comma-separated list of types (why? this isn't in
1095
// check for a comma-separated list of types (why? I don't think
1096
// this is in the specs!)
1045
1097
String[] parts = type_params[ a ].split( "," );
1046
for( int i = 0; i < parts.length; i++ )
1047
if( valid_types.contains( parts[ i ] ) )
1048
types.add( parts[ i ] );
1098
for( int i = 0; i < parts.length; i++ ) {
1099
String ucpart = parts[ i ].toUpperCase( Locale.US );
1100
if( valid_types.contains( ucpart ) )
1101
types.add( ucpart );
1051
1105
// get 2.1-style type param
1052
1106
if( _version.equals( "2.1" ) ) {
1053
for( int i = 1; i < params.length; i++ )
1054
if( valid_types.contains( params[ i ] ) )
1055
types.add( params[ i ] );
1107
for( int i = 1; i < params.length; i++ ) {
1108
String ucparam = params[ i ].toUpperCase( Locale.US );
1109
if( valid_types.contains( ucparam ) )
1110
types.add( ucparam );