4
* Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
4
* Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
6
6
* This file is part of the Import Contacts program (hereafter referred
7
* to as "this program"). For more information, see
8
* http://www.waxworlds.org/edam/software/android/import-contacts
7
* to as "this program"). For more information, see
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
132
133
boolean in_vcard = false;
133
134
while( ( line = reader.readLine() ) != null )
136
138
// look for vcard beginning
137
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 )
142
else if( line.matches( "^END:VCARD" ) )
149
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
143
150
in_vcard = false;
192
202
ContentLineIterator cli = new ContentLineIterator( content );
193
203
while( cli.hasNext() )
195
ByteBuffer buffer = cli.next();
205
ContentLine content_line = cli.next();
197
// get a US-ASCII version of the line for processing
200
line = new String( buffer.array(), buffer.position(),
201
buffer.limit() - buffer.position(), "US-ASCII" );
203
catch( UnsupportedEncodingException e ) {
204
// 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();
208
210
if( vcard == null ) {
209
211
// look for vcard beginning
210
if( line.matches( "^BEGIN:VCARD" ) ) {
212
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
211
213
setProgress( _progress++ );
212
214
vcard = new Vcard();
213
215
vcard_start_line = cli.getLineNumber();
272
273
finish( ACTION_ABORT );
275
// although we're continuing, we still need to abort
276
// this vCard. Further lines will be ignored until we
276
// Although we're continuing, we still need to abort
277
// this vCard. Further lines will be ignored until we
277
278
// get to another BEGIN:VCARD line.
280
281
catch( Vcard.SkipImportException e ) {
282
// abort this vCard. Further lines will be ignored until
283
// Abort this vCard. Further lines will be ignored until
283
284
// we get to another BEGIN:VCARD line.
291
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 >
293
336
protected byte[] _content = null;
294
337
protected int _pos = 0;
347
394
* onto the end of this one?
350
public boolean doesNextLineLookFolded()
397
private boolean doesNextLineLookFolded()
352
399
return _pos > 0 && _pos < _content.length &&
353
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
400
_content[ _pos - 1 ] == '\n' &&
401
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
356
404
public int getLineNumber()
368
416
private final static int MULTILINE_NONE = 0;
369
417
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
370
418
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
371
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
419
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
373
421
private String _version = null;
374
private Vector< ByteBuffer > _buffers = null;
422
private Vector< ContentLine > _content_lines = null;
375
423
private int _name_level = NAMELEVEL_NONE;
376
424
private int _parser_multiline_state = MULTILINE_NONE;
377
425
private String _parser_current_name_and_params = null;
420
468
@SuppressWarnings("serial")
421
469
protected class SkipImportException extends Exception { }
423
private String extractCollonPartFromLine( ByteBuffer buffer,
424
String line, boolean former )
471
private String extractCollonPartFromLine( ContentLine content_line,
426
474
String ret = null;
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
476
// split line into name and value parts and check to make sure we
442
477
// only got 2 parts and that the first part is not zero in length
443
String[] parts = line.split( ":", 2 );
478
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
444
479
if( parts.length == 2 && parts[ 0 ].length() > 0 )
445
480
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 )
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 )
463
496
throws ParseException, SkipImportException,
464
497
AbortImportException
469
502
// tentatively get name and params from line
470
503
String name_and_params =
471
extractNameAndParamsFromLine( buffer, line );
504
extractNameAndParamsFromLine( content_line );
473
506
// is it a version line?
474
507
if( name_and_params != null &&
475
name_and_params.equals( "VERSION" ) )
508
name_and_params.equalsIgnoreCase( "VERSION" ) )
478
String value = extractValueFromLine( buffer, line );
511
String value = extractValueFromLine( content_line ).trim();
479
512
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
480
513
throw new ParseException( R.string.error_vcf_version );
481
514
_version = value;
483
516
// 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() ) == ' ' );
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;
496
525
// no, so stash this line till we get a version
497
if( _buffers == null )
498
_buffers = new Vector< ByteBuffer >();
499
_buffers.add( buffer );
526
if( _content_lines == null )
527
_content_lines = new Vector< ContentLine >();
528
_content_lines.add( content_line );
504
533
// name and params and the position in the buffer where the
505
// "value" part of the line start
534
// "value" part of the line starts
506
535
String name_and_params;
515
544
// skip some initial line characters, depending on the type
516
545
// of multi-line we're handling
517
pos = buffer.position();
546
pos = content_line.getBuffer().position();
518
547
switch( _parser_multiline_state )
520
549
case MULTILINE_FOLDED:
523
552
case MULTILINE_ENCODED:
524
while( pos < buffer.limit() && (
525
buffer.get( pos ) == ' ' ||
526
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 )
541
574
// get name and params from line, and since we're not
542
575
// parsing a subsequent line in a multi-line, this should
543
576
// not fail, or it's an error
544
577
name_and_params =
545
extractNameAndParamsFromLine( buffer, line );
578
extractNameAndParamsFromLine( content_line );
546
579
if( name_and_params == null )
547
580
throw new ParseException(
548
581
R.string.error_vcf_malformed );
550
583
// calculate how many chars to skip from beginning of line
551
584
// so we skip the property "name:" part
552
pos = buffer.position() + name_and_params.length() + 1;
585
pos = content_line.getBuffer().position() +
586
name_and_params.length() + 1;
554
588
// reset the saved multi-line state
555
589
_parser_current_name_and_params = name_and_params;
559
593
// get value from buffer, as raw bytes
560
594
ByteBuffer value;
561
value = ByteBuffer.wrap( buffer.array(), pos,
562
buffer.limit() - pos );
595
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
596
content_line.getBuffer().limit() - pos );
564
598
// get parameter parts
565
599
String[] name_param_parts = name_and_params.split( ";", -1 );
569
603
// determine whether we care about this entry
570
604
final HashSet< String > interesting_fields =
571
new HashSet< String >( Arrays.asList( new String[]
572
{ "N", "FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR" }
605
new HashSet< String >( Arrays.asList( new String[] { "N",
606
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
574
608
boolean is_interesting_field =
575
interesting_fields.contains( name_param_parts[ 0 ] );
609
interesting_fields.contains(
610
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
577
612
// parse encoding parameter
578
613
String encoding = checkParam( name_param_parts, "ENCODING" );
579
if( encoding != null ) encoding = encoding.toUpperCase();
614
if( encoding != null )
615
encoding = encoding.toUpperCase( Locale.US );
580
616
if( is_interesting_field && encoding != null &&
581
!encoding.equals( "8BIT" ) &&
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
583
//&& !encoding.equals( "BASE64" ) )
617
!encoding.equalsIgnoreCase( "8BIT" ) &&
618
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
619
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
585
621
throw new ParseException( R.string.error_vcf_encoding );
588
624
// parse charset parameter
589
625
String charset = checkParam( name_param_parts, "CHARSET" );
590
if( charset != null ) charset = charset.toUpperCase();
626
if( charset != null )
627
charset = charset.toUpperCase( Locale.US );
591
628
if( charset != null &&
592
!charset.equals( "US-ASCII" ) &&
593
!charset.equals( "ASCII" ) &&
594
!charset.equals( "UTF-8" ) )
629
!charset.equalsIgnoreCase( "US-ASCII" ) &&
630
!charset.equalsIgnoreCase( "ASCII" ) &&
631
!charset.equalsIgnoreCase( "UTF-8" ) )
596
633
throw new ParseException( R.string.error_vcf_charset );
599
636
// do unencoding (or default to a fake unencoding result with
600
637
// the raw string)
601
638
UnencodeResult unencoding_result = null;
602
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
639
if( encoding != null &&
640
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
603
642
unencoding_result = unencodeQuotedPrintable( value );
604
// else if( encoding != null && encoding.equals( "BASE64" ) )
644
// else if( encoding != null &&
645
// encoding.equalsIgnoreCase( "BASE64" ) )
605
647
// unencoding_result = unencodeBase64( props[ 1 ], charset );
606
649
if( unencoding_result != null ) {
607
650
value = unencoding_result.getBuffer();
608
651
if( unencoding_result.isAnotherLineRequired() )
613
656
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
614
657
if( ( charset == null && _version.equals( "2.1" ) ) ||
615
658
( charset != null && (
616
charset.equals( "ASCII" ) ||
617
charset.equals( "US-ASCII" ) ) ) )
659
charset.equalsIgnoreCase( "ASCII" ) ||
660
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
619
662
value = transcodeAsciiToUtf8( value );
631
674
// for some entries that have semicolon-separated value parts,
632
675
// check to see if the value ends in an escape character, which
633
676
// indicates that we have a multi-line value
634
if( ( name_param_parts[ 0 ].equals( "N" ) ||
635
name_param_parts[ 0 ].equals( "ORG" ) ||
636
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" ) ) &&
637
680
doesStringEndInAnEscapeChar( string_value ) )
639
682
_parser_multiline_state = MULTILINE_ESCAPED;
641
684
string_value.length() - 1 );
644
// now we know whether we're in an encoding multi-line,
645
// 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
646
689
if( _parser_multiline_state == MULTILINE_NONE &&
647
_version.equals( "3.0" ) && next_line_looks_folded )
690
content_line.doesNextLineLookFolded() )
649
692
_parser_multiline_state = MULTILINE_FOLDED;
662
705
if( complete_value.length() < 1 ) return;
664
707
// parse some properties
665
if( name_param_parts[ 0 ].equals( "N" ) )
708
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
666
709
parseN( name_param_parts, complete_value );
667
else if( name_param_parts[ 0 ].equals( "FN" ) )
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
668
711
parseFN( name_param_parts, complete_value );
669
else if( name_param_parts[ 0 ].equals( "ORG" ) )
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
670
713
parseORG( name_param_parts, complete_value );
671
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
672
715
parseTITLE( name_param_parts, complete_value );
673
else if( name_param_parts[ 0 ].equals( "TEL" ) )
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
674
717
parseTEL( name_param_parts, complete_value );
675
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
676
719
parseEMAIL( name_param_parts, complete_value );
677
else if( name_param_parts[ 0 ].equals( "ADR" ) )
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
678
721
parseADR( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
723
parseLABEL( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
725
parseNOTE( name_param_parts, complete_value );
726
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
727
parseBDAY( name_param_parts, complete_value );
706
755
String str = parts.get( a );
708
// look for parts that end in an escape character, but ignore
709
// the final part. We've already detected escape chars at the
757
// Look for parts that end in an escape character, but ignore
758
// the final part. We've already detected escape chars at the
710
759
// end of the final part in parseLine() and handled multi-lines
712
761
if( a < parts.size() - 1 &&
878
933
if( types.contains( "FAX" ) )
879
934
if( types.contains( "HOME" ) )
880
type = PhonesColumns.TYPE_FAX_HOME;
935
type = TYPE_FAX_HOME;
882
type = PhonesColumns.TYPE_FAX_WORK;
937
type = TYPE_FAX_WORK;
883
938
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
884
type = PhonesColumns.TYPE_MOBILE;
885
940
else if( types.contains( "PAGER" ) )
886
type = PhonesColumns.TYPE_PAGER;
887
942
else if( types.contains( "WORK" ) )
888
type = PhonesColumns.TYPE_WORK;
890
type = PhonesColumns.TYPE_HOME;
892
947
// add phone number
893
948
addNumber( value, type, is_preferred );
921
976
for( int a = 0; a < adr_parts.length; a++ )
922
977
if( adr_parts[ a ].length() > 0 )
924
// split this part in to it's comma-separated bits
925
String[] adr_part_parts =
926
splitValueByCharacter( adr_parts[ a ], ',' );
927
for( int b = 0; b < adr_part_parts.length; b++ )
928
if( adr_part_parts[ b ].length() > 0 )
930
if( value.length() > 0 ) value += "\n";
931
value += adr_part_parts[ b ];
979
// version 3.0 vCards allow further splitting by comma
980
if( _version.equals( "3.0" ) )
982
// split this part in to it's comma-separated bits and
983
// add them on individual lines
984
String[] adr_part_parts =
985
splitValueByCharacter( adr_parts[ a ], ',' );
986
for( int b = 0; b < adr_part_parts.length; b++ )
987
if( adr_part_parts[ b ].length() > 0 )
989
if( value.length() > 0 ) value += "\n";
990
value += adr_part_parts[ b ];
995
// add this part on an individual line
996
if( value.length() > 0 ) value += "\n";
997
value += adr_parts[ a ];
935
1001
Set< String > types = extractTypes( params, Arrays.asList(
936
"PREF", "WORK", "HOME", "INTERNET" ) );
940
if( types.contains( "WORK" ) )
941
type = Contacts.ContactMethods.TYPE_WORK;
943
type = Contacts.ContactMethods.TYPE_HOME;
945
addAddress( unescapeValue( value ), type );
1002
"PREF", "WORK", "HOME" ) );
1006
if( types.contains( "WORK" ) )
1011
addAddress( unescapeValue( value ), type );
1014
private void parseLABEL( String[] params, String value )
1016
Set< String > types = extractTypes( params, Arrays.asList(
1017
"PREF", "WORK", "HOME" ) );
1021
if( types.contains( "WORK" ) )
1026
addAddress( unescapeValue( value ), type );
1029
private void parseNOTE( String[] params, String value )
1031
addNote( unescapeValue( value ) );
1034
private void parseBDAY( String[] params, String value )
1036
setBirthday( value );
948
1039
public void finaliseVcard()
949
1040
throws ParseException, ContactNotIdentifiableException
951
1042
// missing version (and data is present)
952
if( _version == null && _buffers != null )
1043
if( _version == null && _content_lines != null )
953
1044
throw new ParseException( R.string.error_vcf_malformed );
955
1046
// finalise the parent class
980
1073
HashSet< String > ret = new HashSet< String >();
982
1075
Pattern p = Pattern.compile(
983
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1076
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1077
Pattern.CASE_INSENSITIVE );
984
1078
for( int i = 0; i < params.length; i++ ) {
985
1079
Matcher m = p.matcher( params[ i ] );
986
1080
if( m.matches() )
994
* Amongst the params, return any type values present. For v2.1 vCards,
995
* those types are just parameters. For v3.0, they are prefixed with
996
* "TYPE=". There may also be multiple type parameters.
998
* @param a list of type values to look for
1088
* Amongst the params, return any type values present. For v2.1 vCards,
1089
* those types are just parameters. For v3.0, they are prefixed with
1090
* "TYPE=". There may also be multiple type parameters.
1092
* @param params an array of params to look for types in
1093
* @param valid_types an list of upper-case type values to look for
999
1094
* @return a set of present type values
1001
1096
private Set< String > extractTypes( String[] params,
1007
1102
String type_params[] = checkParams( params, "TYPE" );
1008
1103
for( int a = 0; a < type_params.length; a++ )
1010
// check for a comma-separated list of types (why? this isn't in
1105
// check for a comma-separated list of types (why? I don't think
1106
// this is in the specs!)
1012
1107
String[] parts = type_params[ a ].split( "," );
1013
for( int i = 0; i < parts.length; i++ )
1014
if( valid_types.contains( parts[ i ] ) )
1015
types.add( parts[ i ] );
1108
for( int i = 0; i < parts.length; i++ ) {
1109
String ucpart = parts[ i ].toUpperCase( Locale.US );
1110
if( valid_types.contains( ucpart ) )
1111
types.add( ucpart );
1018
1115
// get 2.1-style type param
1019
1116
if( _version.equals( "2.1" ) ) {
1020
for( int i = 1; i < params.length; i++ )
1021
if( valid_types.contains( params[ i ] ) )
1022
types.add( params[ i ] );
1117
for( int i = 1; i < params.length; i++ ) {
1118
String ucparam = params[ i ].toUpperCase( Locale.US );
1119
if( valid_types.contains( ucparam ) )
1120
types.add( ucparam );