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
21
21
* along with this program. If not, see <http://www.gnu.org/licenses/>.
24
package org.waxworlds.edam.importcontacts;
24
package am.ed.importcontacts;
26
26
import java.io.BufferedReader;
27
27
import java.io.File;
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() );
135
136
if( !in_vcard ) {
136
137
// look for vcard beginning
137
if( line.matches( "^BEGIN:VCARD" ) ) {
138
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD.*$" ) ) {
142
else if( line.matches( "^END:VCARD" ) )
143
else if( line.matches( "^END[ \t]*:[ \t]*VCARD.*$" ) )
143
144
in_vcard = false;
175
176
importVCardFileContent( content, file.getName() );
178
catch( OutOfMemoryError e ) {
179
showError( R.string.error_outofmemory );
177
181
catch( FileNotFoundException e ) {
178
182
showError( getText( R.string.error_filenotfound ) +
179
183
file.getName() );
192
196
ContentLineIterator cli = new ContentLineIterator( content );
193
197
while( cli.hasNext() )
195
ByteBuffer buffer = cli.next();
199
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...
201
// get a US-ASCII version of the string, for processing
202
String line = content_line.getUsAsciiLine();
208
204
if( vcard == null ) {
209
205
// look for vcard beginning
210
if( line.matches( "^BEGIN:VCARD" ) ) {
206
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD.*$" ) ) {
211
207
setProgress( _progress++ );
212
208
vcard = new Vcard();
213
209
vcard_start_line = cli.getLineNumber();
259
255
// try giving the line to the vcard
261
vcard.parseLine( buffer, line,
262
cli.doesNextLineLookFolded() );
257
vcard.parseLine( content_line );
264
259
catch( Vcard.ParseException e ) {
291
class ContentLineIterator implements Iterator< ByteBuffer >
288
private ByteBuffer _buffer;
289
private boolean _folded_next;
290
private String _line;
292
public ContentLine( ByteBuffer buffer, boolean folded_next )
295
_folded_next = folded_next;
299
public ByteBuffer getBuffer()
304
public boolean doesNextLineLookFolded()
309
public String getUsAsciiLine()
311
// generated line and cache it
312
if( _line == null ) {
314
_line = new String( _buffer.array(), _buffer.position(),
315
_buffer.limit() - _buffer.position(), "US-ASCII" );
317
catch( UnsupportedEncodingException e ) {
318
// we know US-ASCII *is* supported, so appease the
323
// return cached line
328
class ContentLineIterator implements Iterator< ContentLine >
293
330
protected byte[] _content = null;
294
331
protected int _pos = 0;
319
356
_pos > initial_pos )? _pos - 1 : _pos;
322
return ByteBuffer.wrap( _content, initial_pos,
359
return new ContentLine(
360
ByteBuffer.wrap( _content, initial_pos,
362
doesNextLineLookFolded() );
326
365
// we didn't find one, but were there bytes left?
331
return ByteBuffer.wrap( _content, initial_pos,
370
return new ContentLine(
371
ByteBuffer.wrap( _content, initial_pos,
373
doesNextLineLookFolded() );
347
388
* onto the end of this one?
350
public boolean doesNextLineLookFolded()
391
private boolean doesNextLineLookFolded()
352
393
return _pos > 0 && _pos < _content.length &&
353
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
394
_content[ _pos - 1 ] == '\n' &&
395
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
356
398
public int getLineNumber()
368
410
private final static int MULTILINE_NONE = 0;
369
411
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
370
412
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
371
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
413
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
373
415
private String _version = null;
374
private Vector< ByteBuffer > _buffers = null;
416
private Vector< ContentLine > _content_lines = null;
375
417
private int _name_level = NAMELEVEL_NONE;
376
418
private int _parser_multiline_state = MULTILINE_NONE;
377
419
private String _parser_current_name_and_params = null;
420
462
@SuppressWarnings("serial")
421
463
protected class SkipImportException extends Exception { }
423
private String extractCollonPartFromLine( ByteBuffer buffer,
424
String line, boolean former )
465
private String extractCollonPartFromLine( ContentLine content_line,
426
468
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
470
// split line into name and value parts and check to make sure we
442
471
// only got 2 parts and that the first part is not zero in length
443
String[] parts = line.split( ":", 2 );
472
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
444
473
if( parts.length == 2 && parts[ 0 ].length() > 0 )
445
474
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 )
479
private String extractNameAndParamsFromLine( ContentLine content_line )
481
return extractCollonPartFromLine( content_line, true ).trim();
484
private String extractValueFromLine( ContentLine content_line )
486
return extractCollonPartFromLine( content_line, false );
489
public void parseLine( ContentLine content_line )
463
490
throws ParseException, SkipImportException,
464
491
AbortImportException
469
496
// tentatively get name and params from line
470
497
String name_and_params =
471
extractNameAndParamsFromLine( buffer, line );
498
extractNameAndParamsFromLine( content_line );
473
500
// is it a version line?
474
501
if( name_and_params != null &&
475
name_and_params.equals( "VERSION" ) )
502
name_and_params.equalsIgnoreCase( "VERSION" ) )
478
String value = extractValueFromLine( buffer, line );
505
String value = extractValueFromLine( content_line ).trim();
479
506
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
480
507
throw new ParseException( R.string.error_vcf_version );
481
508
_version = value;
483
510
// 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() ) == ' ' );
512
if( _content_lines != null )
513
for( int i = 0; i < _content_lines.size(); i++ )
514
parseLine( _content_lines.get( i ) );
515
_content_lines = null;
496
519
// no, so stash this line till we get a version
497
if( _buffers == null )
498
_buffers = new Vector< ByteBuffer >();
499
_buffers.add( buffer );
520
if( _content_lines == null )
521
_content_lines = new Vector< ContentLine >();
522
_content_lines.add( content_line );
504
527
// name and params and the position in the buffer where the
505
// "value" part of the line start
528
// "value" part of the line starts
506
529
String name_and_params;
515
538
// skip some initial line characters, depending on the type
516
539
// of multi-line we're handling
517
pos = buffer.position();
540
pos = content_line.getBuffer().position();
518
541
switch( _parser_multiline_state )
520
543
case MULTILINE_FOLDED:
523
546
case MULTILINE_ENCODED:
524
while( pos < buffer.limit() && (
525
buffer.get( pos ) == ' ' ||
526
buffer.get( pos ) == '\t' ) )
547
while( pos < content_line.getBuffer().limit() && (
548
content_line.getBuffer().get( pos ) == ' ' ||
549
content_line.getBuffer().get( pos ) == '\t' ) )
565
if( content_line.getUsAsciiLine().trim().length() == 0 )
541
568
// get name and params from line, and since we're not
542
569
// parsing a subsequent line in a multi-line, this should
543
570
// not fail, or it's an error
544
571
name_and_params =
545
extractNameAndParamsFromLine( buffer, line );
572
extractNameAndParamsFromLine( content_line );
546
573
if( name_and_params == null )
547
574
throw new ParseException(
548
575
R.string.error_vcf_malformed );
550
577
// calculate how many chars to skip from beginning of line
551
578
// so we skip the property "name:" part
552
pos = buffer.position() + name_and_params.length() + 1;
579
pos = content_line.getBuffer().position() +
580
name_and_params.length() + 1;
554
582
// reset the saved multi-line state
555
583
_parser_current_name_and_params = name_and_params;
559
587
// get value from buffer, as raw bytes
560
588
ByteBuffer value;
561
value = ByteBuffer.wrap( buffer.array(), pos,
562
buffer.limit() - pos );
589
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
590
content_line.getBuffer().limit() - pos );
564
592
// get parameter parts
565
593
String[] name_param_parts = name_and_params.split( ";", -1 );
572
600
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
574
602
boolean is_interesting_field =
575
interesting_fields.contains( name_param_parts[ 0 ] );
603
interesting_fields.contains(
604
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
577
606
// parse encoding parameter
578
607
String encoding = checkParam( name_param_parts, "ENCODING" );
579
if( encoding != null ) encoding = encoding.toUpperCase();
608
if( encoding != null )
609
encoding = encoding.toUpperCase( Locale.US );
580
610
if( is_interesting_field && encoding != null &&
581
!encoding.equals( "8BIT" ) &&
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
583
//&& !encoding.equals( "BASE64" ) )
611
!encoding.equalsIgnoreCase( "8BIT" ) &&
612
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
613
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
585
615
throw new ParseException( R.string.error_vcf_encoding );
588
618
// parse charset parameter
589
619
String charset = checkParam( name_param_parts, "CHARSET" );
590
if( charset != null ) charset = charset.toUpperCase();
620
if( charset != null )
621
charset = charset.toUpperCase( Locale.US );
591
622
if( charset != null &&
592
!charset.equals( "US-ASCII" ) &&
593
!charset.equals( "ASCII" ) &&
594
!charset.equals( "UTF-8" ) )
623
!charset.equalsIgnoreCase( "US-ASCII" ) &&
624
!charset.equalsIgnoreCase( "ASCII" ) &&
625
!charset.equalsIgnoreCase( "UTF-8" ) )
596
627
throw new ParseException( R.string.error_vcf_charset );
599
630
// do unencoding (or default to a fake unencoding result with
600
631
// the raw string)
601
632
UnencodeResult unencoding_result = null;
602
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
633
if( encoding != null &&
634
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
603
636
unencoding_result = unencodeQuotedPrintable( value );
604
// else if( encoding != null && encoding.equals( "BASE64" ) )
638
// else if( encoding != null &&
639
// encoding.equalsIgnoreCase( "BASE64" ) )
605
641
// unencoding_result = unencodeBase64( props[ 1 ], charset );
606
643
if( unencoding_result != null ) {
607
644
value = unencoding_result.getBuffer();
608
645
if( unencoding_result.isAnotherLineRequired() )
613
650
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
614
651
if( ( charset == null && _version.equals( "2.1" ) ) ||
615
652
( charset != null && (
616
charset.equals( "ASCII" ) ||
617
charset.equals( "US-ASCII" ) ) ) )
653
charset.equalsIgnoreCase( "ASCII" ) ||
654
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
619
656
value = transcodeAsciiToUtf8( value );
631
668
// for some entries that have semicolon-separated value parts,
632
669
// check to see if the value ends in an escape character, which
633
670
// 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" ) ) &&
671
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
672
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
673
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
637
674
doesStringEndInAnEscapeChar( string_value ) )
639
676
_parser_multiline_state = MULTILINE_ESCAPED;
641
678
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
681
// if we know we're not in an encoding-based multi-line, check
682
// to see if we're in a folded multi-line
646
683
if( _parser_multiline_state == MULTILINE_NONE &&
647
_version.equals( "3.0" ) && next_line_looks_folded )
684
content_line.doesNextLineLookFolded() )
649
686
_parser_multiline_state = MULTILINE_FOLDED;
662
699
if( complete_value.length() < 1 ) return;
664
701
// parse some properties
665
if( name_param_parts[ 0 ].equals( "N" ) )
702
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
666
703
parseN( name_param_parts, complete_value );
667
else if( name_param_parts[ 0 ].equals( "FN" ) )
704
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
668
705
parseFN( name_param_parts, complete_value );
669
else if( name_param_parts[ 0 ].equals( "ORG" ) )
706
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
670
707
parseORG( name_param_parts, complete_value );
671
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
708
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
672
709
parseTITLE( name_param_parts, complete_value );
673
else if( name_param_parts[ 0 ].equals( "TEL" ) )
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
674
711
parseTEL( name_param_parts, complete_value );
675
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
676
713
parseEMAIL( name_param_parts, complete_value );
677
else if( name_param_parts[ 0 ].equals( "ADR" ) )
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
678
715
parseADR( name_param_parts, complete_value );
679
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
680
717
parseLABEL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
719
parseNOTE( name_param_parts, complete_value );
798
843
for( int b = 0; b < name_part_parts.length; b++ )
799
844
if( name_part_parts[ b ].length() > 0 )
801
if( value.length() == 0 ) value += " ";
846
if( value.length() > 0 ) value += " ";
802
847
value += name_part_parts[ b ];
880
925
if( types.contains( "FAX" ) )
881
926
if( types.contains( "HOME" ) )
882
type = PhonesColumns.TYPE_FAX_HOME;
927
type = TYPE_FAX_HOME;
884
type = PhonesColumns.TYPE_FAX_WORK;
929
type = TYPE_FAX_WORK;
885
930
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
886
type = PhonesColumns.TYPE_MOBILE;
887
932
else if( types.contains( "PAGER" ) )
888
type = PhonesColumns.TYPE_PAGER;
889
934
else if( types.contains( "WORK" ) )
890
type = PhonesColumns.TYPE_WORK;
892
type = PhonesColumns.TYPE_HOME;
894
939
// add phone number
895
940
addNumber( value, type, is_preferred );
906
951
boolean is_preferred = types.contains( "PREF" );
908
953
if( types.contains( "WORK" ) )
909
type = Contacts.ContactMethods.TYPE_WORK;
911
type = Contacts.ContactMethods.TYPE_HOME;
913
958
addEmail( unescapeValue( value ), type, is_preferred );
923
968
for( int a = 0; a < adr_parts.length; a++ )
924
969
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 ];
971
// version 3.0 vCards allow further splitting by comma
972
if( _version.equals( "3.0" ) )
974
// split this part in to it's comma-separated bits and
975
// add them on individual lines
976
String[] adr_part_parts =
977
splitValueByCharacter( adr_parts[ a ], ',' );
978
for( int b = 0; b < adr_part_parts.length; b++ )
979
if( adr_part_parts[ b ].length() > 0 )
981
if( value.length() > 0 ) value += "\n";
982
value += adr_part_parts[ b ];
987
// add this part on an individual line
988
if( value.length() > 0 ) value += "\n";
989
value += adr_parts[ a ];
937
993
Set< String > types = extractTypes( params, Arrays.asList(
957
1013
if( types.contains( "WORK" ) )
958
type = Contacts.ContactMethods.TYPE_WORK;
960
type = Contacts.ContactMethods.TYPE_HOME;
962
1018
addAddress( unescapeValue( value ), type );
1021
private void parseNOTE( String[] params, String value )
1023
addNote( unescapeValue( value ) );
965
1026
public void finaliseVcard()
966
1027
throws ParseException, ContactNotIdentifiableException
968
1029
// missing version (and data is present)
969
if( _version == null && _buffers != null )
1030
if( _version == null && _content_lines != null )
970
1031
throw new ParseException( R.string.error_vcf_malformed );
972
1033
// finalise the parent class
997
1058
HashSet< String > ret = new HashSet< String >();
999
1060
Pattern p = Pattern.compile(
1000
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1061
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1062
Pattern.CASE_INSENSITIVE );
1001
1063
for( int i = 0; i < params.length; i++ ) {
1002
1064
Matcher m = p.matcher( params[ i ] );
1003
1065
if( m.matches() )
1011
1073
* Amongst the params, return any type values present. For v2.1 vCards,
1012
1074
* those types are just parameters. For v3.0, they are prefixed with
1013
1075
* "TYPE=". There may also be multiple type parameters.
1015
* @param a list of type values to look for
1076
* @param params an array of params to look for types in
1077
* @param valid_types an list of upper-case type values to look for
1016
1078
* @return a set of present type values
1018
1080
private Set< String > extractTypes( String[] params,
1024
1086
String type_params[] = checkParams( params, "TYPE" );
1025
1087
for( int a = 0; a < type_params.length; a++ )
1027
// check for a comma-separated list of types (why? this isn't in
1089
// check for a comma-separated list of types (why? I don't think
1090
// this is in the specs!)
1029
1091
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 ] );
1092
for( int i = 0; i < parts.length; i++ ) {
1093
String ucpart = parts[ i ].toUpperCase( Locale.US );
1094
if( valid_types.contains( ucpart ) )
1095
types.add( ucpart );
1035
1099
// get 2.1-style type param
1036
1100
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 ] );
1101
for( int i = 1; i < params.length; i++ ) {
1102
String ucparam = params[ i ].toUpperCase( Locale.US );
1103
if( valid_types.contains( ucparam ) )
1104
types.add( ucparam );