4
* Copyright (C) 2009 to 2011 Tim Marston <tim@ed.am>
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
7
* to as "this program"). For more information, see
8
8
* http://ed.am/dev/android/import-contacts
10
10
* This program is free software: you can redistribute it and/or modify
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
83
84
class VCardFilter implements FilenameFilter {
84
85
public boolean accept( File dir, String name ) {
85
return name.toLowerCase().endsWith( ".vcf" );
86
return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
88
89
files = file.listFiles( new VCardFilter() );
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;
175
182
importVCardFileContent( content, file.getName() );
184
catch( OutOfMemoryError e ) {
185
showError( R.string.error_outofmemory );
177
187
catch( FileNotFoundException e ) {
178
188
showError( getText( R.string.error_filenotfound ) +
179
189
file.getName() );
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();
259
261
// try giving the line to the vcard
261
vcard.parseLine( buffer, line,
262
cli.doesNextLineLookFolded() );
263
vcard.parseLine( content_line );
264
265
catch( Vcard.ParseException e ) {
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;
319
362
_pos > initial_pos )? _pos - 1 : _pos;
322
return ByteBuffer.wrap( _content, initial_pos,
365
return new ContentLine(
366
ByteBuffer.wrap( _content, initial_pos,
368
doesNextLineLookFolded() );
326
371
// we didn't find one, but were there bytes left?
331
return ByteBuffer.wrap( _content, initial_pos,
376
return new ContentLine(
377
ByteBuffer.wrap( _content, initial_pos,
379
doesNextLineLookFolded() );
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,
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
474
// split line into name and value parts and check to make sure we
442
475
// only got 2 parts and that the first part is not zero in length
443
String[] parts = line.split( ":", 2 );
476
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
444
477
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 )
478
return parts[ former? 0 : 1 ].trim();
483
private String extractNameAndParamsFromLine( ContentLine content_line )
485
return extractCollonPartFromLine( content_line, true );
488
private String extractValueFromLine( ContentLine content_line )
490
return extractCollonPartFromLine( content_line, false );
493
public void parseLine( ContentLine content_line )
463
494
throws ParseException, SkipImportException,
464
495
AbortImportException
469
500
// tentatively get name and params from line
470
501
String name_and_params =
471
extractNameAndParamsFromLine( buffer, line );
502
extractNameAndParamsFromLine( content_line );
473
504
// is it a version line?
474
505
if( name_and_params != null &&
475
name_and_params.equals( "VERSION" ) )
506
name_and_params.equalsIgnoreCase( "VERSION" ) )
478
String value = extractValueFromLine( buffer, line );
479
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
509
String value = extractValueFromLine( content_line );
510
if( value == null || (
511
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
480
513
throw new ParseException( R.string.error_vcf_version );
481
515
_version = value;
483
517
// 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() ) == ' ' );
519
if( _content_lines != null )
520
for( int i = 0; i < _content_lines.size(); i++ )
521
parseLine( _content_lines.get( i ) );
522
_content_lines = null;
496
526
// no, so stash this line till we get a version
497
if( _buffers == null )
498
_buffers = new Vector< ByteBuffer >();
499
_buffers.add( buffer );
527
if( _content_lines == null )
528
_content_lines = new Vector< ContentLine >();
529
_content_lines.add( content_line );
504
534
// name and params and the position in the buffer where the
505
// "value" part of the line start
535
// "value" part of the line starts
506
536
String name_and_params;
515
545
// skip some initial line characters, depending on the type
516
546
// of multi-line we're handling
517
pos = buffer.position();
547
pos = content_line.getBuffer().position();
518
548
switch( _parser_multiline_state )
520
550
case MULTILINE_FOLDED:
523
553
case MULTILINE_ENCODED:
524
while( pos < buffer.limit() && (
525
buffer.get( pos ) == ' ' ||
526
buffer.get( pos ) == '\t' ) )
554
while( pos < content_line.getBuffer().limit() && (
555
content_line.getBuffer().get( pos ) == ' ' ||
556
content_line.getBuffer().get( pos ) == '\t' ) )
572
if( content_line.getUsAsciiLine().trim().length() == 0 )
541
575
// get name and params from line, and since we're not
542
576
// parsing a subsequent line in a multi-line, this should
543
577
// not fail, or it's an error
544
578
name_and_params =
545
extractNameAndParamsFromLine( buffer, line );
579
extractNameAndParamsFromLine( content_line );
546
580
if( name_and_params == null )
547
581
throw new ParseException(
548
582
R.string.error_vcf_malformed );
550
584
// calculate how many chars to skip from beginning of line
551
585
// so we skip the property "name:" part
552
pos = buffer.position() + name_and_params.length() + 1;
586
pos = content_line.getBuffer().position() +
587
name_and_params.length() + 1;
554
589
// reset the saved multi-line state
555
590
_parser_current_name_and_params = name_and_params;
559
594
// get value from buffer, as raw bytes
560
595
ByteBuffer value;
561
value = ByteBuffer.wrap( buffer.array(), pos,
562
buffer.limit() - pos );
596
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
597
content_line.getBuffer().limit() - pos );
564
599
// get parameter parts
565
600
String[] name_param_parts = name_and_params.split( ";", -1 );
572
607
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
574
609
boolean is_interesting_field =
575
interesting_fields.contains( name_param_parts[ 0 ] );
610
interesting_fields.contains(
611
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
577
613
// parse encoding parameter
578
614
String encoding = checkParam( name_param_parts, "ENCODING" );
579
if( encoding != null ) encoding = encoding.toUpperCase();
615
if( encoding != null )
616
encoding = encoding.toUpperCase( Locale.US );
580
617
if( is_interesting_field && encoding != null &&
581
!encoding.equals( "8BIT" ) &&
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
583
//&& !encoding.equals( "BASE64" ) )
618
!encoding.equalsIgnoreCase( "8BIT" ) &&
619
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
620
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
585
622
throw new ParseException( R.string.error_vcf_encoding );
588
625
// parse charset parameter
589
626
String charset = checkParam( name_param_parts, "CHARSET" );
590
if( charset != null ) charset = charset.toUpperCase();
627
if( charset != null )
628
charset = charset.toUpperCase( Locale.US );
591
629
if( charset != null &&
592
!charset.equals( "US-ASCII" ) &&
593
!charset.equals( "ASCII" ) &&
594
!charset.equals( "UTF-8" ) )
630
!charset.equalsIgnoreCase( "US-ASCII" ) &&
631
!charset.equalsIgnoreCase( "ASCII" ) &&
632
!charset.equalsIgnoreCase( "UTF-8" ) )
596
634
throw new ParseException( R.string.error_vcf_charset );
599
637
// do unencoding (or default to a fake unencoding result with
600
638
// the raw string)
601
639
UnencodeResult unencoding_result = null;
602
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
640
if( encoding != null &&
641
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
603
643
unencoding_result = unencodeQuotedPrintable( value );
604
// else if( encoding != null && encoding.equals( "BASE64" ) )
645
// else if( encoding != null &&
646
// encoding.equalsIgnoreCase( "BASE64" ) )
605
648
// unencoding_result = unencodeBase64( props[ 1 ], charset );
606
650
if( unencoding_result != null ) {
607
651
value = unencoding_result.getBuffer();
608
652
if( unencoding_result.isAnotherLineRequired() )
613
657
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
614
658
if( ( charset == null && _version.equals( "2.1" ) ) ||
615
659
( charset != null && (
616
charset.equals( "ASCII" ) ||
617
charset.equals( "US-ASCII" ) ) ) )
660
charset.equalsIgnoreCase( "ASCII" ) ||
661
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
619
663
value = transcodeAsciiToUtf8( value );
631
675
// for some entries that have semicolon-separated value parts,
632
676
// check to see if the value ends in an escape character, which
633
677
// 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" ) ) &&
678
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
679
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
680
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
637
681
doesStringEndInAnEscapeChar( string_value ) )
639
683
_parser_multiline_state = MULTILINE_ESCAPED;
641
685
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
688
// if we know we're not in an encoding-based multi-line, check
689
// to see if we're in a folded multi-line
646
690
if( _parser_multiline_state == MULTILINE_NONE &&
647
_version.equals( "3.0" ) && next_line_looks_folded )
691
content_line.doesNextLineLookFolded() )
649
693
_parser_multiline_state = MULTILINE_FOLDED;
662
706
if( complete_value.length() < 1 ) return;
664
708
// parse some properties
665
if( name_param_parts[ 0 ].equals( "N" ) )
709
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
666
710
parseN( name_param_parts, complete_value );
667
else if( name_param_parts[ 0 ].equals( "FN" ) )
711
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
668
712
parseFN( name_param_parts, complete_value );
669
else if( name_param_parts[ 0 ].equals( "ORG" ) )
713
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
670
714
parseORG( name_param_parts, complete_value );
671
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
715
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
672
716
parseTITLE( name_param_parts, complete_value );
673
else if( name_param_parts[ 0 ].equals( "TEL" ) )
717
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
674
718
parseTEL( name_param_parts, complete_value );
675
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
719
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
676
720
parseEMAIL( name_param_parts, complete_value );
677
else if( name_param_parts[ 0 ].equals( "ADR" ) )
721
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
678
722
parseADR( name_param_parts, complete_value );
679
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
723
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
680
724
parseLABEL( name_param_parts, complete_value );
725
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
726
parseNOTE( name_param_parts, complete_value );
727
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
728
parseBDAY( name_param_parts, complete_value );
708
756
String str = parts.get( a );
710
// look for parts that end in an escape character, but ignore
711
// the final part. We've already detected escape chars at the
758
// Look for parts that end in an escape character, but ignore
759
// the final part. We've already detected escape chars at the
712
760
// end of the final part in parseLine() and handled multi-lines
714
762
if( a < parts.size() - 1 &&
798
852
for( int b = 0; b < name_part_parts.length; b++ )
799
853
if( name_part_parts[ b ].length() > 0 )
801
if( value.length() == 0 ) value += " ";
855
if( value.length() > 0 ) value += " ";
802
856
value += name_part_parts[ b ];
880
934
if( types.contains( "FAX" ) )
881
935
if( types.contains( "HOME" ) )
882
type = PhonesColumns.TYPE_FAX_HOME;
936
type = TYPE_FAX_HOME;
884
type = PhonesColumns.TYPE_FAX_WORK;
938
type = TYPE_FAX_WORK;
885
939
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
886
type = PhonesColumns.TYPE_MOBILE;
887
941
else if( types.contains( "PAGER" ) )
888
type = PhonesColumns.TYPE_PAGER;
889
943
else if( types.contains( "WORK" ) )
890
type = PhonesColumns.TYPE_WORK;
892
type = PhonesColumns.TYPE_HOME;
894
948
// add phone number
895
949
addNumber( value, type, is_preferred );
906
960
boolean is_preferred = types.contains( "PREF" );
908
962
if( types.contains( "WORK" ) )
909
type = Contacts.ContactMethods.TYPE_WORK;
911
type = Contacts.ContactMethods.TYPE_HOME;
913
967
addEmail( unescapeValue( value ), type, is_preferred );
923
977
for( int a = 0; a < adr_parts.length; a++ )
924
978
if( adr_parts[ a ].length() > 0 )
926
// split this part in to it's comma-separated bits
927
String[] adr_part_parts =
928
splitValueByCharacter( adr_parts[ a ], ',' );
929
for( int b = 0; b < adr_part_parts.length; b++ )
930
if( adr_part_parts[ b ].length() > 0 )
932
if( value.length() > 0 ) value += "\n";
933
value += adr_part_parts[ b ];
980
// version 3.0 vCards allow further splitting by comma
981
if( _version.equals( "3.0" ) )
983
// split this part in to it's comma-separated bits and
984
// add them on individual lines
985
String[] adr_part_parts =
986
splitValueByCharacter( adr_parts[ a ], ',' );
987
for( int b = 0; b < adr_part_parts.length; b++ )
988
if( adr_part_parts[ b ].length() > 0 )
990
if( value.length() > 0 ) value += "\n";
991
value += adr_part_parts[ b ];
996
// add this part on an individual line
997
if( value.length() > 0 ) value += "\n";
998
value += adr_parts[ a ];
937
1002
Set< String > types = extractTypes( params, Arrays.asList(
957
1022
if( types.contains( "WORK" ) )
958
type = Contacts.ContactMethods.TYPE_WORK;
960
type = Contacts.ContactMethods.TYPE_HOME;
962
1027
addAddress( unescapeValue( value ), type );
1030
private void parseNOTE( String[] params, String value )
1032
addNote( unescapeValue( value ) );
1035
private void parseBDAY( String[] params, String value )
1037
setBirthday( value );
965
1040
public void finaliseVcard()
966
1041
throws ParseException, ContactNotIdentifiableException
968
1043
// missing version (and data is present)
969
if( _version == null && _buffers != null )
1044
if( _version == null && _content_lines != null )
970
1045
throw new ParseException( R.string.error_vcf_malformed );
972
1047
// finalise the parent class
997
1074
HashSet< String > ret = new HashSet< String >();
999
1076
Pattern p = Pattern.compile(
1000
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1077
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1078
Pattern.CASE_INSENSITIVE );
1001
1079
for( int i = 0; i < params.length; i++ ) {
1002
1080
Matcher m = p.matcher( params[ i ] );
1003
1081
if( m.matches() )
1011
* Amongst the params, return any type values present. For v2.1 vCards,
1012
* those types are just parameters. For v3.0, they are prefixed with
1013
* "TYPE=". There may also be multiple type parameters.
1015
* @param a list of type values to look for
1089
* Amongst the params, return any type values present. For v2.1 vCards,
1090
* those types are just parameters. For v3.0, they are prefixed with
1091
* "TYPE=". There may also be multiple type parameters.
1093
* @param params an array of params to look for types in
1094
* @param valid_types an list of upper-case type values to look for
1016
1095
* @return a set of present type values
1018
1097
private Set< String > extractTypes( String[] params,
1024
1103
String type_params[] = checkParams( params, "TYPE" );
1025
1104
for( int a = 0; a < type_params.length; a++ )
1027
// check for a comma-separated list of types (why? this isn't in
1106
// check for a comma-separated list of types (why? I don't think
1107
// this is in the specs!)
1029
1108
String[] parts = type_params[ a ].split( "," );
1030
for( int i = 0; i < parts.length; i++ )
1031
if( valid_types.contains( parts[ i ] ) )
1032
types.add( parts[ i ] );
1109
for( int i = 0; i < parts.length; i++ ) {
1110
String ucpart = parts[ i ].toUpperCase( Locale.US );
1111
if( valid_types.contains( ucpart ) )
1112
types.add( ucpart );
1035
1116
// get 2.1-style type param
1036
1117
if( _version.equals( "2.1" ) ) {
1037
for( int i = 1; i < params.length; i++ )
1038
if( valid_types.contains( params[ i ] ) )
1039
types.add( params[ i ] );
1118
for( int i = 1; i < params.length; i++ ) {
1119
String ucparam = params[ i ].toUpperCase( Locale.US );
1120
if( valid_types.contains( ucparam ) )
1121
types.add( ucparam );
1064
1147
else if( ch == '=' && i == in.limit() - 1 )
1066
1149
// we found a '=' at the end of a line signifying a multi-
1067
// line string, so we don't add it.
1150
// line string, so we don't add it
1068
1151
another = true;