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
201
202
buffer.limit() - buffer.position(), "US-ASCII" );
203
204
catch( UnsupportedEncodingException e ) {
204
// we know US-ASCII is supported, so appease the compiler...
205
// we know US-ASCII *is* supported, so appease the compiler...
208
209
if( vcard == null ) {
209
210
// look for vcard beginning
210
if( line.matches( "^BEGIN:VCARD" ) ) {
211
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD" ) ) {
211
212
setProgress( _progress++ );
212
213
vcard = new Vcard();
213
214
vcard_start_line = cli.getLineNumber();
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
574
// determine whether we care about this entry
570
575
final HashSet< String > interesting_fields =
571
new HashSet< String >( Arrays.asList( new String[]
572
{ "N", "FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR" }
576
new HashSet< String >( Arrays.asList( new String[] { "N",
577
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
574
579
boolean is_interesting_field =
575
interesting_fields.contains( name_param_parts[ 0 ] );
580
interesting_fields.contains(
581
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
577
583
// parse encoding parameter
578
584
String encoding = checkParam( name_param_parts, "ENCODING" );
579
if( encoding != null ) encoding = encoding.toUpperCase();
585
if( encoding != null )
586
encoding = encoding.toUpperCase( Locale.US );
580
587
if( is_interesting_field && encoding != null &&
581
!encoding.equals( "8BIT" ) &&
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
583
//&& !encoding.equals( "BASE64" ) )
588
!encoding.equalsIgnoreCase( "8BIT" ) &&
589
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
590
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
585
592
throw new ParseException( R.string.error_vcf_encoding );
588
595
// parse charset parameter
589
596
String charset = checkParam( name_param_parts, "CHARSET" );
590
if( charset != null ) charset = charset.toUpperCase();
597
if( charset != null )
598
charset = charset.toUpperCase( Locale.US );
591
599
if( charset != null &&
592
!charset.equals( "US-ASCII" ) &&
593
!charset.equals( "ASCII" ) &&
594
!charset.equals( "UTF-8" ) )
600
!charset.equalsIgnoreCase( "US-ASCII" ) &&
601
!charset.equalsIgnoreCase( "ASCII" ) &&
602
!charset.equalsIgnoreCase( "UTF-8" ) )
596
604
throw new ParseException( R.string.error_vcf_charset );
599
607
// do unencoding (or default to a fake unencoding result with
600
608
// the raw string)
601
609
UnencodeResult unencoding_result = null;
602
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
610
if( encoding != null &&
611
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
603
613
unencoding_result = unencodeQuotedPrintable( value );
604
// else if( encoding != null && encoding.equals( "BASE64" ) )
615
// else if( encoding != null &&
616
// encoding.equalsIgnoreCase( "BASE64" ) )
605
618
// unencoding_result = unencodeBase64( props[ 1 ], charset );
606
620
if( unencoding_result != null ) {
607
621
value = unencoding_result.getBuffer();
608
622
if( unencoding_result.isAnotherLineRequired() )
609
623
_parser_multiline_state = MULTILINE_ENCODED;
612
// convert 8-bit ASCII charset to US-ASCII
613
if( charset == null || charset.equals( "ASCII" ) ) {
626
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
627
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
628
if( ( charset == null && _version.equals( "2.1" ) ) ||
629
( charset != null && (
630
charset.equalsIgnoreCase( "ASCII" ) ||
631
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
614
633
value = transcodeAsciiToUtf8( value );
636
// process charset (value is now in UTF-8)
619
637
String string_value;
621
639
string_value = new String( value.array(), value.position(),
622
value.limit() - value.position(), charset );
640
value.limit() - value.position(), "UTF-8" );
623
641
} catch( UnsupportedEncodingException e ) {
624
642
throw new ParseException( R.string.error_vcf_charset );
627
645
// for some entries that have semicolon-separated value parts,
628
646
// check to see if the value ends in an escape character, which
629
647
// 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" ) ) &&
648
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
649
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
650
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
633
651
doesStringEndInAnEscapeChar( string_value ) )
635
653
_parser_multiline_state = MULTILINE_ESCAPED;
637
655
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
658
// if we know we're not in an encoding-based multi-line, check
659
// to see if we're in a folded multi-line
642
660
if( _parser_multiline_state == MULTILINE_NONE &&
643
_version.equals( "3.0" ) && next_line_looks_folded )
661
next_line_looks_folded )
645
663
_parser_multiline_state = MULTILINE_FOLDED;
658
676
if( complete_value.length() < 1 ) return;
660
678
// parse some properties
661
if( name_param_parts[ 0 ].equals( "N" ) )
679
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
662
680
parseN( name_param_parts, complete_value );
663
else if( name_param_parts[ 0 ].equals( "FN" ) )
681
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
664
682
parseFN( name_param_parts, complete_value );
665
else if( name_param_parts[ 0 ].equals( "ORG" ) )
683
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
666
684
parseORG( name_param_parts, complete_value );
667
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
685
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
668
686
parseTITLE( name_param_parts, complete_value );
669
else if( name_param_parts[ 0 ].equals( "TEL" ) )
687
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
670
688
parseTEL( name_param_parts, complete_value );
671
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
689
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
672
690
parseEMAIL( name_param_parts, complete_value );
673
else if( name_param_parts[ 0 ].equals( "ADR" ) )
691
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
674
692
parseADR( name_param_parts, complete_value );
693
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
694
parseLABEL( name_param_parts, complete_value );
695
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
696
parseNOTE( name_param_parts, complete_value );
690
712
return ( count & 1 ) == 1;
693
private String[] splitValueBySemicolon( String value )
715
private String[] splitValueByCharacter( String value, char character )
695
// split string in to parts by semicolon
717
// split string in to parts by specified character
696
718
ArrayList< String > parts = new ArrayList< String >(
697
Arrays.asList( value.split( ";" ) ) );
719
Arrays.asList( value.split( "" + character ) ) );
699
721
// go through parts
700
722
for( int a = 0; a < parts.size(); a++ )
708
730
if( a < parts.size() - 1 &&
709
731
doesStringEndInAnEscapeChar( str ) )
711
// join the next part to this part and remove the next part
733
// append the escaped character, join the next part to this
734
// part and remove the next part
712
735
parts.set( a, str.substring( 0, str.length() - 1 ) +
713
';' + parts.get( a + 1 ) );
736
character + parts.get( a + 1 ) );
714
737
parts.remove( a + 1 );
716
739
// re-visit this part
727
750
return parts.toArray( ret );
753
private String unescapeValue( String value )
755
StringBuilder ret = new StringBuilder( value.length() );
756
boolean in_escape = false;
757
for( int a = 0; a < value.length(); a++ )
759
int c = value.codePointAt( a );
761
// process a normal character
766
ret.append( Character.toChars( c ) );
770
// process an escape sequence
776
// add tab (invalid/non-standard, but accepted)
787
// add escaped character
788
ret.append( Character.toChars( c ) );
791
// unknown escape sequence, so add it unescaped
792
// (invalid/non-standard, but accepted)
794
ret.append( Character.toChars( c ) );
799
return ret.toString();
730
802
private void parseN( String[] params, String value )
732
804
// already got a better name?
733
805
if( _name_level >= NAMELEVEL_N ) return;
735
807
// get name parts
736
String[] name_parts = splitValueBySemicolon( value );
808
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 ];
812
final int[] part_order = { 3, 1, 2, 0, 4 };
813
for( int a = 0; a < part_order.length; a++ )
814
if( name_parts.length > part_order[ a ] &&
815
name_parts[ part_order[ a ] ].length() > 0 )
817
// split this part in to it's comma-separated bits
818
String[] name_part_parts = splitValueByCharacter(
819
name_parts[ part_order[ a ] ], ',' );
820
for( int b = 0; b < name_part_parts.length; b++ )
821
if( name_part_parts[ b ].length() > 0 )
823
if( value.length() == 0 ) value += " ";
824
value += name_part_parts[ b ];
829
setName( unescapeValue( value ) );
747
830
_name_level = NAMELEVEL_N;
753
836
if( _name_level >= NAMELEVEL_FN ) return;
839
setName( unescapeValue( value ) );
757
840
_name_level = NAMELEVEL_FN;
760
843
private void parseORG( String[] params, String value )
763
String[] org_parts = splitValueBySemicolon( value );
846
String[] org_parts = splitValueByCharacter( value, ';' );
764
847
if( org_parts == null || org_parts.length < 1 ) return;
766
849
// build organisation name
768
851
String.valueOf( org_parts[ 0 ] ) );
769
852
for( int a = 1; a < org_parts.length; a++ )
770
853
builder.append( ", " ).append( org_parts[ a ] );
771
String organisation = builder.toString();
854
String organisation = unescapeValue( builder.toString() );
773
856
// set organisation name (using a title we've previously found)
774
857
addOrganisation( organisation, _cached_title, true );
817
902
if( types.contains( "FAX" ) )
818
903
if( types.contains( "HOME" ) )
819
type = PhonesColumns.TYPE_FAX_HOME;
904
type = TYPE_FAX_HOME;
821
type = PhonesColumns.TYPE_FAX_WORK;
906
type = TYPE_FAX_WORK;
822
907
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
823
type = PhonesColumns.TYPE_MOBILE;
824
909
else if( types.contains( "PAGER" ) )
825
type = PhonesColumns.TYPE_PAGER;
826
911
else if( types.contains( "WORK" ) )
827
type = PhonesColumns.TYPE_WORK;
829
type = PhonesColumns.TYPE_HOME;
831
916
// add phone number
832
917
addNumber( value, type, is_preferred );
843
928
boolean is_preferred = types.contains( "PREF" );
845
930
if( types.contains( "WORK" ) )
846
type = Contacts.ContactMethods.TYPE_WORK;
848
type = Contacts.ContactMethods.TYPE_HOME;
850
addEmail( value, type, is_preferred );
935
addEmail( unescapeValue( value ), type, is_preferred );
853
938
private void parseADR( String[] params, String value )
855
940
// get address parts
856
String[] adr_parts = splitValueBySemicolon( value );
941
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 );
945
for( int a = 0; a < adr_parts.length; a++ )
946
if( adr_parts[ a ].length() > 0 )
948
// version 3.0 vCards allow further splitting by comma
949
if( _version.equals( "3.0" ) )
951
// split this part in to it's comma-separated bits and
952
// add them on individual lines
953
String[] adr_part_parts =
954
splitValueByCharacter( adr_parts[ a ], ',' );
955
for( int b = 0; b < adr_part_parts.length; b++ )
956
if( adr_part_parts[ b ].length() > 0 )
958
if( value.length() > 0 ) value += "\n";
959
value += adr_part_parts[ b ];
964
// add this part on an individual line
965
if( value.length() > 0 ) value += "\n";
966
value += adr_parts[ a ];
970
Set< String > types = extractTypes( params, Arrays.asList(
971
"PREF", "WORK", "HOME" ) );
975
if( types.contains( "WORK" ) )
980
addAddress( unescapeValue( value ), type );
983
private void parseLABEL( String[] params, String value )
985
Set< String > types = extractTypes( params, Arrays.asList(
986
"PREF", "WORK", "HOME" ) );
990
if( types.contains( "WORK" ) )
995
addAddress( unescapeValue( value ), type );
998
private void parseNOTE( String[] params, String value )
1000
addNote( unescapeValue( value ) );
878
1003
public void finaliseVcard()
1015
* Amongst the params, find the value of the first, only, of any with
1016
* the specified name
1019
* @return a value, or null
889
1021
private String checkParam( String[] params, String name )
1023
String[] res = checkParams( params, name );
1024
return res.length > 0? res[ 0 ] : null;
1028
* Amongst the params, find the values of any with the specified name
1031
* @return an array of values, or null
1033
private String[] checkParams( String[] params, String name )
1035
HashSet< String > ret = new HashSet< String >();
891
1037
Pattern p = Pattern.compile(
892
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1038
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1039
Pattern.CASE_INSENSITIVE );
893
1040
for( int i = 0; i < params.length; i++ ) {
894
1041
Matcher m = p.matcher( params[ i ] );
895
1042
if( m.matches() )
1043
ret.add( m.group( 2 ) );
1046
return (String[]) ret.toArray( new String[ ret.size() ] );
1050
* Amongst the params, return any type values present. For v2.1 vCards,
1051
* those types are just parameters. For v3.0, they are prefixed with
1052
* "TYPE=". There may also be multiple type parameters.
1053
* @param params an array of params to look for types in
1054
* @param valid_types an list of upper-case type values to look for
1055
* @return a set of present type values
901
1057
private Set< String > extractTypes( String[] params,
902
1058
List< String > valid_types )
904
1060
HashSet< String > types = new HashSet< String >();
906
1062
// 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 ] );
1063
String type_params[] = checkParams( params, "TYPE" );
1064
for( int a = 0; a < type_params.length; a++ )
1066
// check for a comma-separated list of types (why? I don't think
1067
// this is in the specs!)
1068
String[] parts = type_params[ a ].split( "," );
1069
for( int i = 0; i < parts.length; i++ ) {
1070
String ucpart = parts[ i ].toUpperCase( Locale.US );
1071
if( valid_types.contains( ucpart ) )
1072
types.add( ucpart );
915
1076
// get 2.1-style type param
916
1077
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 ] );
1078
for( int i = 1; i < params.length; i++ ) {
1079
String ucparam = params[ i ].toUpperCase( Locale.US );
1080
if( valid_types.contains( ucparam ) )
1081
types.add( ucparam );