420
476
@SuppressWarnings("serial")
421
477
protected class SkipImportException extends Exception { }
423
private String extractCollonPartFromLine( ByteBuffer buffer,
424
String line, boolean former )
479
private String extractCollonPartFromLine( ContentLine content_line,
428
// get a US-ASCII version of the line for processing, unless we were
432
line = new String( buffer.array(), buffer.position(),
433
buffer.limit() - buffer.position(), "US-ASCII" );
435
catch( UnsupportedEncodingException e ) {
436
// we know US-ASCII is supported, so appease the compiler...
441
482
// split line into name and value parts and check to make sure we
442
483
// only got 2 parts and that the first part is not zero in length
443
String[] parts = line.split( ":", 2 );
484
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
444
485
if( parts.length == 2 && parts[ 0 ].length() > 0 )
445
ret = parts[ former? 0 : 1 ];
450
private String extractNameAndParamsFromLine( ByteBuffer buffer,
453
return extractCollonPartFromLine( buffer, line, true );
456
private String extractValueFromLine( ByteBuffer buffer, String line )
458
return extractCollonPartFromLine( buffer, line, false );
461
public void parseLine( ByteBuffer buffer, String line,
462
boolean next_line_looks_folded )
486
return parts[ former? 0 : 1 ].trim();
491
private String extractNameAndParamsFromLine( ContentLine content_line )
493
return extractCollonPartFromLine( content_line, true );
496
private String extractValueFromLine( ContentLine content_line )
498
return extractCollonPartFromLine( content_line, false );
501
public void parseLine( ContentLine content_line )
463
502
throws ParseException, SkipImportException,
464
503
AbortImportException
469
508
// tentatively get name and params from line
470
509
String name_and_params =
471
extractNameAndParamsFromLine( buffer, line );
510
extractNameAndParamsFromLine( content_line );
473
512
// is it a version line?
474
513
if( name_and_params != null &&
475
name_and_params.equals( "VERSION" ) )
514
name_and_params.equalsIgnoreCase( "VERSION" ) )
478
String value = extractValueFromLine( buffer, line );
479
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
517
String value = extractValueFromLine( content_line );
518
if( value == null || (
519
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
480
521
throw new ParseException( R.string.error_vcf_version );
481
523
_version = value;
483
525
// parse any buffers we've been accumulating while we waited
485
if( _buffers != null )
486
for( int i = 0; i < _buffers.size(); i++ )
487
parseLine( _buffers.get( i ), null,
488
i + 1 < _buffers.size() &&
489
_buffers.get( i + 1 ).hasRemaining() &&
490
_buffers.get( i + 1 ).get(
491
_buffers.get( i + 1 ).position() ) == ' ' );
527
if( _content_lines != null )
528
for( int i = 0; i < _content_lines.size(); i++ )
529
parseLine( _content_lines.get( i ) );
530
_content_lines = null;
496
534
// no, so stash this line till we get a version
497
if( _buffers == null )
498
_buffers = new Vector< ByteBuffer >();
499
_buffers.add( buffer );
535
if( _content_lines == null )
536
_content_lines = new Vector< ContentLine >();
537
_content_lines.add( content_line );
504
542
// name and params and the position in the buffer where the
505
// "value" part of the line start
543
// "value" part of the line starts
506
544
String name_and_params;
569
612
// determine whether we care about this entry
570
613
final HashSet< String > interesting_fields =
571
new HashSet< String >( Arrays.asList( new String[]
572
{ "N", "FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR" }
614
new HashSet< String >( Arrays.asList( new String[] { "N",
615
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
574
617
boolean is_interesting_field =
575
interesting_fields.contains( name_param_parts[ 0 ] );
618
interesting_fields.contains(
619
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
577
621
// parse encoding parameter
578
622
String encoding = checkParam( name_param_parts, "ENCODING" );
579
if( encoding != null ) encoding = encoding.toUpperCase();
623
if( encoding != null )
624
encoding = encoding.toUpperCase( Locale.US );
580
625
if( is_interesting_field && encoding != null &&
581
!encoding.equals( "8BIT" ) &&
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
583
//&& !encoding.equals( "BASE64" ) )
626
!encoding.equalsIgnoreCase( "8BIT" ) &&
627
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
628
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
585
630
throw new ParseException( R.string.error_vcf_encoding );
588
633
// parse charset parameter
589
634
String charset = checkParam( name_param_parts, "CHARSET" );
590
if( charset != null ) charset = charset.toUpperCase();
635
if( charset != null )
636
charset = charset.toUpperCase( Locale.US );
591
637
if( charset != null &&
592
!charset.equals( "US-ASCII" ) &&
593
!charset.equals( "ASCII" ) &&
594
!charset.equals( "UTF-8" ) )
638
!charset.equalsIgnoreCase( "US-ASCII" ) &&
639
!charset.equalsIgnoreCase( "ASCII" ) &&
640
!charset.equalsIgnoreCase( "UTF-8" ) )
596
642
throw new ParseException( R.string.error_vcf_charset );
599
645
// do unencoding (or default to a fake unencoding result with
600
646
// the raw string)
601
647
UnencodeResult unencoding_result = null;
602
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
648
if( encoding != null &&
649
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
603
651
unencoding_result = unencodeQuotedPrintable( value );
604
// else if( encoding != null && encoding.equals( "BASE64" ) )
653
// else if( encoding != null &&
654
// encoding.equalsIgnoreCase( "BASE64" ) )
605
656
// unencoding_result = unencodeBase64( props[ 1 ], charset );
606
658
if( unencoding_result != null ) {
607
659
value = unencoding_result.getBuffer();
608
660
if( unencoding_result.isAnotherLineRequired() )
609
661
_parser_multiline_state = MULTILINE_ENCODED;
612
// convert 8-bit ASCII charset to US-ASCII
613
if( charset == null || charset.equals( "ASCII" ) ) {
664
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
665
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
666
if( ( charset == null && _version.equals( "2.1" ) ) ||
667
( charset != null && (
668
charset.equalsIgnoreCase( "ASCII" ) ||
669
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
614
671
value = transcodeAsciiToUtf8( value );
674
// process charset (value is now in UTF-8)
619
675
String string_value;
621
677
string_value = new String( value.array(), value.position(),
622
value.limit() - value.position(), charset );
678
value.limit() - value.position(), "UTF-8" );
623
679
} catch( UnsupportedEncodingException e ) {
624
680
throw new ParseException( R.string.error_vcf_charset );
658
714
if( complete_value.length() < 1 ) return;
660
716
// parse some properties
661
if( name_param_parts[ 0 ].equals( "N" ) )
717
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
662
718
parseN( name_param_parts, complete_value );
663
else if( name_param_parts[ 0 ].equals( "FN" ) )
719
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
664
720
parseFN( name_param_parts, complete_value );
665
else if( name_param_parts[ 0 ].equals( "ORG" ) )
721
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
666
722
parseORG( name_param_parts, complete_value );
667
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
723
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
668
724
parseTITLE( name_param_parts, complete_value );
669
else if( name_param_parts[ 0 ].equals( "TEL" ) )
725
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
670
726
parseTEL( name_param_parts, complete_value );
671
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
727
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
672
728
parseEMAIL( name_param_parts, complete_value );
673
else if( name_param_parts[ 0 ].equals( "ADR" ) )
729
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
674
730
parseADR( name_param_parts, complete_value );
731
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
732
parseLABEL( name_param_parts, complete_value );
733
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
734
parseNOTE( name_param_parts, complete_value );
735
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
736
parseBDAY( name_param_parts, complete_value );
690
752
return ( count & 1 ) == 1;
693
private String[] splitValueBySemicolon( String value )
755
private String[] splitValueByCharacter( String value, char character )
695
// split string in to parts by semicolon
757
// split string in to parts by specified character
696
758
ArrayList< String > parts = new ArrayList< String >(
697
Arrays.asList( value.split( ";" ) ) );
759
Arrays.asList( value.split( "" + character ) ) );
699
761
// go through parts
700
762
for( int a = 0; a < parts.size(); a++ )
702
764
String str = parts.get( a );
704
// look for parts that end in an escape character, but ignore
705
// the final part. We've already detected escape chars at the
766
// Look for parts that end in an escape character, but ignore
767
// the final part. We've already detected escape chars at the
706
768
// end of the final part in parseLine() and handled multi-lines
708
770
if( a < parts.size() - 1 &&
709
771
doesStringEndInAnEscapeChar( str ) )
711
// join the next part to this part and remove the next part
773
// append the escaped character, join the next part to this
774
// part and remove the next part
712
775
parts.set( a, str.substring( 0, str.length() - 1 ) +
713
';' + parts.get( a + 1 ) );
776
character + parts.get( a + 1 ) );
714
777
parts.remove( a + 1 );
716
779
// re-visit this part
727
790
return parts.toArray( ret );
793
private String unescapeValue( String value )
795
StringBuilder ret = new StringBuilder( value.length() );
796
boolean in_escape = false;
797
for( int a = 0; a < value.length(); a++ )
799
int c = value.codePointAt( a );
801
// process a normal character
806
ret.append( Character.toChars( c ) );
810
// process an escape sequence
816
// add tab (invalid/non-standard, but accepted)
827
// add escaped character
828
ret.append( Character.toChars( c ) );
831
// unknown escape sequence, so add it unescaped
832
// (invalid/non-standard, but accepted)
834
ret.append( Character.toChars( c ) );
839
return ret.toString();
730
842
private void parseN( String[] params, String value )
732
844
// already got a better name?
733
845
if( _name_level >= NAMELEVEL_N ) return;
735
847
// get name parts
736
String[] name_parts = splitValueBySemicolon( value );
848
String[] name_parts = splitValueByCharacter( value, ';' );
740
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
741
value += name_parts[ 1 ];
742
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
743
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
852
final int[] part_order = { 3, 1, 2, 0, 4 };
853
for( int a = 0; a < part_order.length; a++ )
854
if( name_parts.length > part_order[ a ] &&
855
name_parts[ part_order[ a ] ].length() > 0 )
857
// split this part in to it's comma-separated bits
858
String[] name_part_parts = splitValueByCharacter(
859
name_parts[ part_order[ a ] ], ',' );
860
for( int b = 0; b < name_part_parts.length; b++ )
861
if( name_part_parts[ b ].length() > 0 )
863
if( value.length() > 0 ) value += " ";
864
value += name_part_parts[ b ];
869
setName( unescapeValue( value ) );
747
870
_name_level = NAMELEVEL_N;
843
968
boolean is_preferred = types.contains( "PREF" );
845
970
if( types.contains( "WORK" ) )
846
type = Contacts.ContactMethods.TYPE_WORK;
848
type = Contacts.ContactMethods.TYPE_HOME;
850
addEmail( value, type, is_preferred );
975
addEmail( unescapeValue( value ), type, is_preferred );
853
978
private void parseADR( String[] params, String value )
855
980
// get address parts
856
String[] adr_parts = splitValueBySemicolon( value );
981
String[] adr_parts = splitValueByCharacter( value, ';' );
860
for( int a = 0; a < adr_parts.length; a++ ) {
861
if( value.length() > 0 ) value += "\n";
862
value += adr_parts[ a ].trim();
865
Set< String > types = extractTypes( params, Arrays.asList(
866
"PREF", "WORK", "HOME", "INTERNET" ) );
870
if( types.contains( "WORK" ) )
871
type = Contacts.ContactMethods.TYPE_WORK;
873
type = Contacts.ContactMethods.TYPE_HOME;
875
addAddress( value, type );
985
for( int a = 0; a < adr_parts.length; a++ )
986
if( adr_parts[ a ].length() > 0 )
988
// version 3.0 vCards allow further splitting by comma
989
if( _version.equals( "3.0" ) )
991
// split this part in to it's comma-separated bits and
992
// add them on individual lines
993
String[] adr_part_parts =
994
splitValueByCharacter( adr_parts[ a ], ',' );
995
for( int b = 0; b < adr_part_parts.length; b++ )
996
if( adr_part_parts[ b ].length() > 0 )
998
if( value.length() > 0 ) value += "\n";
999
value += adr_part_parts[ b ];
1004
// add this part on an individual line
1005
if( value.length() > 0 ) value += "\n";
1006
value += adr_parts[ a ];
1010
Set< String > types = extractTypes( params, Arrays.asList(
1011
"PREF", "WORK", "HOME" ) );
1015
if( types.contains( "WORK" ) )
1020
addAddress( unescapeValue( value ), type );
1023
private void parseLABEL( String[] params, String value )
1025
Set< String > types = extractTypes( params, Arrays.asList(
1026
"PREF", "WORK", "HOME" ) );
1030
if( types.contains( "WORK" ) )
1035
addAddress( unescapeValue( value ), type );
1038
private void parseNOTE( String[] params, String value )
1040
addNote( unescapeValue( value ) );
1043
private void parseBDAY( String[] params, String value )
1045
setBirthday( value );
878
1048
public void finaliseVcard()
879
1049
throws ParseException, ContactNotIdentifiableException
881
1051
// missing version (and data is present)
882
if( _version == null && _buffers != null )
1052
if( _version == null && _content_lines != null )
883
1053
throw new ParseException( R.string.error_vcf_malformed );
885
1055
// finalise the parent class
1060
* Amongst the params, find the value of the first, only, of any with
1061
* the specified name.
1065
* @return a value, or null
889
1067
private String checkParam( String[] params, String name )
1069
String[] res = checkParams( params, name );
1070
return res.length > 0? res[ 0 ] : null;
1074
* Amongst the params, find the values of any with the specified name.
1078
* @return an array of values, or null
1080
private String[] checkParams( String[] params, String name )
1082
HashSet< String > ret = new HashSet< String >();
891
1084
Pattern p = Pattern.compile(
892
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1085
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1086
Pattern.CASE_INSENSITIVE );
893
1087
for( int i = 0; i < params.length; i++ ) {
894
1088
Matcher m = p.matcher( params[ i ] );
895
1089
if( m.matches() )
1090
ret.add( m.group( 2 ) );
1093
return (String[]) ret.toArray( new String[ ret.size() ] );
1097
* Amongst the params, return any type values present. For v2.1 vCards,
1098
* those types are just parameters. For v3.0, they are prefixed with
1099
* "TYPE=". There may also be multiple type parameters.
1101
* @param params an array of params to look for types in
1102
* @param valid_types an list of upper-case type values to look for
1103
* @return a set of present type values
901
1105
private Set< String > extractTypes( String[] params,
902
1106
List< String > valid_types )
904
1108
HashSet< String > types = new HashSet< String >();
906
1110
// get 3.0-style TYPE= param
908
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
909
String[] parts = type_param.split( "," );
910
for( int i = 0; i < parts.length; i++ )
911
if( valid_types.contains( parts[ i ] ) )
912
types.add( parts[ i ] );
1111
String type_params[] = checkParams( params, "TYPE" );
1112
for( int a = 0; a < type_params.length; a++ )
1114
// check for a comma-separated list of types (why? I don't think
1115
// this is in the specs!)
1116
String[] parts = type_params[ a ].split( "," );
1117
for( int i = 0; i < parts.length; i++ ) {
1118
String ucpart = parts[ i ].toUpperCase( Locale.US );
1119
if( valid_types.contains( ucpart ) )
1120
types.add( ucpart );
915
1124
// get 2.1-style type param
916
1125
if( _version.equals( "2.1" ) ) {
917
for( int i = 1; i < params.length; i++ )
918
if( valid_types.contains( params[ i ] ) )
919
types.add( params[ i ] );
1126
for( int i = 1; i < params.length; i++ ) {
1127
String ucparam = params[ i ].toUpperCase( Locale.US );
1128
if( valid_types.contains( ucparam ) )
1129
types.add( ucparam );