4
* Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
4
* Copyright (C) 2009 to 2011 Tim Marston <tim@ed.am>
6
6
* This file is part of the Import Contacts program (hereafter referred
7
7
* to as "this program"). For more information, see
8
* http://www.waxworlds.org/edam/software/android/import-contacts
8
* http://ed.am/dev/android/import-contacts
10
10
* This program is free software: you can redistribute it and/or modify
11
11
* it under the terms of the GNU General Public License as published by
38
38
import java.util.HashSet;
39
39
import java.util.Iterator;
40
40
import java.util.List;
41
import java.util.Locale;
41
42
import java.util.NoSuchElementException;
42
43
import java.util.Set;
43
44
import java.util.Vector;
44
45
import java.util.regex.Matcher;
45
46
import java.util.regex.Pattern;
48
import android.annotation.SuppressLint;
47
49
import android.content.SharedPreferences;
48
import android.provider.Contacts;
49
import android.provider.Contacts.PhonesColumns;
51
51
public class VcardImporter extends Importer
362
364
private class Vcard extends ContactData
364
366
private final static int NAMELEVEL_NONE = 0;
365
private final static int NAMELEVEL_FN = 1;
366
private final static int NAMELEVEL_N = 2;
367
private final static int NAMELEVEL_N = 1;
368
private final static int NAMELEVEL_FN = 2;
368
370
private final static int MULTILINE_NONE = 0;
369
371
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
370
372
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
371
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
373
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
373
375
private String _version = null;
374
376
private Vector< ByteBuffer > _buffers = null;
569
571
// determine whether we care about this entry
570
572
final HashSet< String > interesting_fields =
571
new HashSet< String >( Arrays.asList( new String[]
572
{ "N", "FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR" }
573
new HashSet< String >( Arrays.asList( new String[] { "N",
574
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
574
576
boolean is_interesting_field =
575
577
interesting_fields.contains( name_param_parts[ 0 ] );
577
579
// parse encoding parameter
578
580
String encoding = checkParam( name_param_parts, "ENCODING" );
579
if( encoding != null ) encoding = encoding.toUpperCase();
581
if( encoding != null )
582
encoding = encoding.toUpperCase( Locale.US );
580
583
if( is_interesting_field && encoding != null &&
581
!encoding.equals( "8BIT" ) &&
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
583
//&& !encoding.equals( "BASE64" ) )
584
!encoding.equalsIgnoreCase( "8BIT" ) &&
585
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
586
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
585
588
throw new ParseException( R.string.error_vcf_encoding );
588
591
// parse charset parameter
589
592
String charset = checkParam( name_param_parts, "CHARSET" );
590
if( charset != null ) charset = charset.toUpperCase();
593
if( charset != null )
594
charset = charset.toUpperCase( Locale.US );
591
595
if( charset != null &&
592
!charset.equals( "US-ASCII" ) &&
593
!charset.equals( "ASCII" ) &&
594
!charset.equals( "UTF-8" ) )
596
!charset.equalsIgnoreCase( "US-ASCII" ) &&
597
!charset.equalsIgnoreCase( "ASCII" ) &&
598
!charset.equalsIgnoreCase( "UTF-8" ) )
596
600
throw new ParseException( R.string.error_vcf_charset );
599
603
// do unencoding (or default to a fake unencoding result with
600
604
// the raw string)
601
605
UnencodeResult unencoding_result = null;
602
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
606
if( encoding != null &&
607
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
603
609
unencoding_result = unencodeQuotedPrintable( value );
604
// else if( encoding != null && encoding.equals( "BASE64" ) )
611
// else if( encoding != null &&
612
// encoding.equalsIgnoreCase( "BASE64" ) )
605
614
// unencoding_result = unencodeBase64( props[ 1 ], charset );
606
616
if( unencoding_result != null ) {
607
617
value = unencoding_result.getBuffer();
608
618
if( unencoding_result.isAnotherLineRequired() )
609
619
_parser_multiline_state = MULTILINE_ENCODED;
612
// convert 8-bit ASCII charset to US-ASCII
613
if( charset == null || charset.equals( "ASCII" ) ) {
622
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
623
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
624
if( ( charset == null && _version.equals( "2.1" ) ) ||
625
( charset != null && (
626
charset.equalsIgnoreCase( "ASCII" ) ||
627
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
614
629
value = transcodeAsciiToUtf8( value );
632
// process charset (value is now in UTF-8)
619
633
String string_value;
621
635
string_value = new String( value.array(), value.position(),
622
value.limit() - value.position(), charset );
636
value.limit() - value.position(), "UTF-8" );
623
637
} catch( UnsupportedEncodingException e ) {
624
638
throw new ParseException( R.string.error_vcf_charset );
627
641
// for some entries that have semicolon-separated value parts,
628
642
// check to see if the value ends in an escape character, which
629
643
// indicates that we have a multi-line value
630
if( ( name_param_parts[ 0 ].equals( "N" ) ||
631
name_param_parts[ 0 ].equals( "ORG" ) ||
632
name_param_parts[ 0 ].equals( "ADR" ) ) &&
644
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
645
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
646
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
633
647
doesStringEndInAnEscapeChar( string_value ) )
635
649
_parser_multiline_state = MULTILINE_ESCAPED;
637
651
string_value.length() - 1 );
640
// now we know whether we're in an encoding multi-line,
641
// determine if we're in a v3 folded multi-line or not
654
// if we know we're not in an encoding-based multi-line, check
655
// to see if we're in a folded multi-line
642
656
if( _parser_multiline_state == MULTILINE_NONE &&
643
_version.equals( "3.0" ) && next_line_looks_folded )
657
next_line_looks_folded )
645
659
_parser_multiline_state = MULTILINE_FOLDED;
658
672
if( complete_value.length() < 1 ) return;
660
674
// parse some properties
661
if( name_param_parts[ 0 ].equals( "N" ) )
675
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
662
676
parseN( name_param_parts, complete_value );
663
else if( name_param_parts[ 0 ].equals( "FN" ) )
677
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
664
678
parseFN( name_param_parts, complete_value );
665
else if( name_param_parts[ 0 ].equals( "ORG" ) )
679
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
666
680
parseORG( name_param_parts, complete_value );
667
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
681
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
668
682
parseTITLE( name_param_parts, complete_value );
669
else if( name_param_parts[ 0 ].equals( "TEL" ) )
683
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
670
684
parseTEL( name_param_parts, complete_value );
671
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
685
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
672
686
parseEMAIL( name_param_parts, complete_value );
673
else if( name_param_parts[ 0 ].equals( "ADR" ) )
687
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
674
688
parseADR( name_param_parts, complete_value );
689
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
690
parseLABEL( name_param_parts, complete_value );
691
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
692
parseNOTE( name_param_parts, complete_value );
690
708
return ( count & 1 ) == 1;
693
private String[] splitValueBySemicolon( String value )
711
private String[] splitValueByCharacter( String value, char character )
695
// split string in to parts by semicolon
713
// split string in to parts by specified character
696
714
ArrayList< String > parts = new ArrayList< String >(
697
Arrays.asList( value.split( ";" ) ) );
715
Arrays.asList( value.split( "" + character ) ) );
699
717
// go through parts
700
718
for( int a = 0; a < parts.size(); a++ )
708
726
if( a < parts.size() - 1 &&
709
727
doesStringEndInAnEscapeChar( str ) )
711
// join the next part to this part and remove the next part
729
// append the escaped character, join the next part to this
730
// part and remove the next part
712
731
parts.set( a, str.substring( 0, str.length() - 1 ) +
713
';' + parts.get( a + 1 ) );
732
character + parts.get( a + 1 ) );
714
733
parts.remove( a + 1 );
716
735
// re-visit this part
727
746
return parts.toArray( ret );
749
private String unescapeValue( String value )
751
StringBuilder ret = new StringBuilder( value.length() );
752
boolean in_escape = false;
753
for( int a = 0; a < value.length(); a++ )
755
int c = value.codePointAt( a );
757
// process a normal character
762
ret.append( Character.toChars( c ) );
766
// process an escape sequence
772
// add tab (invalid/non-standard, but accepted)
783
// add escaped character
784
ret.append( Character.toChars( c ) );
787
// unknown escape sequence, so add it unescaped
788
// (invalid/non-standard, but accepted)
790
ret.append( Character.toChars( c ) );
795
return ret.toString();
730
798
private void parseN( String[] params, String value )
732
800
// already got a better name?
733
801
if( _name_level >= NAMELEVEL_N ) return;
735
803
// get name parts
736
String[] name_parts = splitValueBySemicolon( value );
804
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 ];
808
final int[] part_order = { 3, 1, 2, 0, 4 };
809
for( int a = 0; a < part_order.length; a++ )
810
if( name_parts.length > part_order[ a ] &&
811
name_parts[ part_order[ a ] ].length() > 0 )
813
// split this part in to it's comma-separated bits
814
String[] name_part_parts = splitValueByCharacter(
815
name_parts[ part_order[ a ] ], ',' );
816
for( int b = 0; b < name_part_parts.length; b++ )
817
if( name_part_parts[ b ].length() > 0 )
819
if( value.length() == 0 ) value += " ";
820
value += name_part_parts[ b ];
825
setName( unescapeValue( value ) );
747
826
_name_level = NAMELEVEL_N;
753
832
if( _name_level >= NAMELEVEL_FN ) return;
835
setName( unescapeValue( value ) );
757
836
_name_level = NAMELEVEL_FN;
760
839
private void parseORG( String[] params, String value )
763
String[] org_parts = splitValueBySemicolon( value );
842
String[] org_parts = splitValueByCharacter( value, ';' );
764
843
if( org_parts == null || org_parts.length < 1 ) return;
766
845
// build organisation name
768
847
String.valueOf( org_parts[ 0 ] ) );
769
848
for( int a = 1; a < org_parts.length; a++ )
770
849
builder.append( ", " ).append( org_parts[ a ] );
771
String organisation = builder.toString();
850
String organisation = unescapeValue( builder.toString() );
773
852
// set organisation name (using a title we've previously found)
774
853
addOrganisation( organisation, _cached_title, true );
817
898
if( types.contains( "FAX" ) )
818
899
if( types.contains( "HOME" ) )
819
type = PhonesColumns.TYPE_FAX_HOME;
900
type = TYPE_FAX_HOME;
821
type = PhonesColumns.TYPE_FAX_WORK;
902
type = TYPE_FAX_WORK;
822
903
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
823
type = PhonesColumns.TYPE_MOBILE;
824
905
else if( types.contains( "PAGER" ) )
825
type = PhonesColumns.TYPE_PAGER;
826
907
else if( types.contains( "WORK" ) )
827
type = PhonesColumns.TYPE_WORK;
829
type = PhonesColumns.TYPE_HOME;
831
912
// add phone number
832
913
addNumber( value, type, is_preferred );
843
924
boolean is_preferred = types.contains( "PREF" );
845
926
if( types.contains( "WORK" ) )
846
type = Contacts.ContactMethods.TYPE_WORK;
848
type = Contacts.ContactMethods.TYPE_HOME;
850
addEmail( value, type, is_preferred );
931
addEmail( unescapeValue( value ), type, is_preferred );
853
934
private void parseADR( String[] params, String value )
855
936
// get address parts
856
String[] adr_parts = splitValueBySemicolon( value );
937
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 );
941
for( int a = 0; a < adr_parts.length; a++ )
942
if( adr_parts[ a ].length() > 0 )
944
// version 3.0 vCards allow further splitting by comma
945
if( _version.equals( "3.0" ) )
947
// split this part in to it's comma-separated bits and
948
// add them on individual lines
949
String[] adr_part_parts =
950
splitValueByCharacter( adr_parts[ a ], ',' );
951
for( int b = 0; b < adr_part_parts.length; b++ )
952
if( adr_part_parts[ b ].length() > 0 )
954
if( value.length() > 0 ) value += "\n";
955
value += adr_part_parts[ b ];
960
// add this part on an individual line
961
if( value.length() > 0 ) value += "\n";
962
value += adr_parts[ a ];
966
Set< String > types = extractTypes( params, Arrays.asList(
967
"PREF", "WORK", "HOME" ) );
971
if( types.contains( "WORK" ) )
976
addAddress( unescapeValue( value ), type );
979
private void parseLABEL( String[] params, String value )
981
Set< String > types = extractTypes( params, Arrays.asList(
982
"PREF", "WORK", "HOME" ) );
986
if( types.contains( "WORK" ) )
991
addAddress( unescapeValue( value ), type );
994
private void parseNOTE( String[] params, String value )
996
addNote( unescapeValue( value ) );
878
999
public void finaliseVcard()
1011
* Amongst the params, find the value of the first, only, of any with
1012
* the specified name
1015
* @return a value, or null
889
1017
private String checkParam( String[] params, String name )
1019
String[] res = checkParams( params, name );
1020
return res.length > 0? res[ 0 ] : null;
1024
* Amongst the params, find the values of any with the specified name
1027
* @return an array of values, or null
1029
private String[] checkParams( String[] params, String name )
1031
HashSet< String > ret = new HashSet< String >();
891
1033
Pattern p = Pattern.compile(
892
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1034
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1035
Pattern.CASE_INSENSITIVE );
893
1036
for( int i = 0; i < params.length; i++ ) {
894
1037
Matcher m = p.matcher( params[ i ] );
895
1038
if( m.matches() )
1039
ret.add( m.group( 2 ) );
1042
return (String[]) ret.toArray( new String[ ret.size() ] );
1046
* Amongst the params, return any type values present. For v2.1 vCards,
1047
* those types are just parameters. For v3.0, they are prefixed with
1048
* "TYPE=". There may also be multiple type parameters.
1049
* @param params an array of params to look for types in
1050
* @param valid_types an list of upper-case type values to look for
1051
* @return a set of present type values
901
1053
private Set< String > extractTypes( String[] params,
902
1054
List< String > valid_types )
904
1056
HashSet< String > types = new HashSet< String >();
906
1058
// get 3.0-style TYPE= param
908
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
909
String[] parts = type_param.split( "," );
1059
String type_params[] = checkParams( params, "TYPE" );
1060
for( int a = 0; a < type_params.length; a++ )
1062
// check for a comma-separated list of types (why? this isn't in
1064
String[] parts = type_params[ a ].split( "," );
1065
for( int i = 0; i < parts.length; i++ )
1066
parts[ i ] = parts[ i ].toUpperCase( Locale.US );
910
1067
for( int i = 0; i < parts.length; i++ )
911
1068
if( valid_types.contains( parts[ i ] ) )
912
1069
types.add( parts[ i ] );