4
* Copyright (C) 2009 to 2011 Tim Marston <tim@ed.am>
4
* Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
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://ed.am/dev/android/import-contacts
8
* http://www.waxworlds.org/edam/software/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 am.ed.importcontacts;
24
package org.waxworlds.edam.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;
42
41
import java.util.NoSuchElementException;
43
42
import java.util.Set;
44
43
import java.util.Vector;
45
44
import java.util.regex.Matcher;
46
45
import java.util.regex.Pattern;
48
import android.annotation.SuppressLint;
49
47
import android.content.SharedPreferences;
48
import android.provider.Contacts;
49
import android.provider.Contacts.PhonesColumns;
51
51
public class VcardImporter extends Importer
84
83
class VCardFilter implements FilenameFilter {
85
84
public boolean accept( File dir, String name ) {
86
return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
85
return name.toLowerCase().endsWith( ".vcf" );
89
88
files = file.listFiles( new VCardFilter() );
176
175
importVCardFileContent( content, file.getName() );
178
catch( OutOfMemoryError e ) {
179
showError( R.string.error_outofmemory );
181
177
catch( FileNotFoundException e ) {
182
178
showError( getText( R.string.error_filenotfound ) +
183
179
file.getName() );
196
192
ContentLineIterator cli = new ContentLineIterator( content );
197
193
while( cli.hasNext() )
199
ContentLine content_line = cli.next();
195
ByteBuffer buffer = cli.next();
201
// get a US-ASCII version of the string, for processing
202
String line = content_line.getUsAsciiLine();
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...
204
208
if( vcard == null ) {
205
209
// look for vcard beginning
206
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD" ) ) {
210
if( line.matches( "^BEGIN:VCARD" ) ) {
207
211
setProgress( _progress++ );
208
212
vcard = new Vcard();
209
213
vcard_start_line = cli.getLineNumber();
255
259
// try giving the line to the vcard
257
vcard.parseLine( content_line );
261
vcard.parseLine( buffer, line,
262
cli.doesNextLineLookFolded() );
259
264
catch( Vcard.ParseException e ) {
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 >
291
class ContentLineIterator implements Iterator< ByteBuffer >
330
293
protected byte[] _content = null;
331
294
protected int _pos = 0;
356
319
_pos > initial_pos )? _pos - 1 : _pos;
359
return new ContentLine(
360
ByteBuffer.wrap( _content, initial_pos,
362
doesNextLineLookFolded() );
322
return ByteBuffer.wrap( _content, initial_pos,
365
326
// we didn't find one, but were there bytes left?
370
return new ContentLine(
371
ByteBuffer.wrap( _content, initial_pos,
373
doesNextLineLookFolded() );
331
return ByteBuffer.wrap( _content, initial_pos,
388
347
* onto the end of this one?
391
private boolean doesNextLineLookFolded()
350
public boolean doesNextLineLookFolded()
393
352
return _pos > 0 && _pos < _content.length &&
394
_content[ _pos - 1 ] == '\n' &&
395
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
353
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
398
356
public int getLineNumber()
410
368
private final static int MULTILINE_NONE = 0;
411
369
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
412
370
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
413
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
371
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
415
373
private String _version = null;
416
private Vector< ContentLine > _content_lines = null;
374
private Vector< ByteBuffer > _buffers = null;
417
375
private int _name_level = NAMELEVEL_NONE;
418
376
private int _parser_multiline_state = MULTILINE_NONE;
419
377
private String _parser_current_name_and_params = null;
462
420
@SuppressWarnings("serial")
463
421
protected class SkipImportException extends Exception { }
465
private String extractCollonPartFromLine( ContentLine content_line,
423
private String extractCollonPartFromLine( ByteBuffer buffer,
424
String line, boolean former )
468
426
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...
470
441
// split line into name and value parts and check to make sure we
471
442
// only got 2 parts and that the first part is not zero in length
472
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
443
String[] parts = line.split( ":", 2 );
473
444
if( parts.length == 2 && parts[ 0 ].length() > 0 )
474
445
ret = parts[ former? 0 : 1 ];
479
private String extractNameAndParamsFromLine( ContentLine content_line )
481
return extractCollonPartFromLine( content_line, true );
484
private String extractValueFromLine( ContentLine content_line )
486
return extractCollonPartFromLine( content_line, false );
489
public void parseLine( ContentLine content_line )
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 )
490
463
throws ParseException, SkipImportException,
491
464
AbortImportException
496
469
// tentatively get name and params from line
497
470
String name_and_params =
498
extractNameAndParamsFromLine( content_line );
471
extractNameAndParamsFromLine( buffer, line );
500
473
// is it a version line?
501
474
if( name_and_params != null &&
502
name_and_params.equalsIgnoreCase( "VERSION" ) )
475
name_and_params.equals( "VERSION" ) )
505
String value = extractValueFromLine( content_line );
478
String value = extractValueFromLine( buffer, line );
506
479
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
507
480
throw new ParseException( R.string.error_vcf_version );
508
481
_version = value;
510
483
// parse any buffers we've been accumulating while we waited
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;
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
496
// no, so stash this line till we get a version
520
if( _content_lines == null )
521
_content_lines = new Vector< ContentLine >();
522
_content_lines.add( content_line );
497
if( _buffers == null )
498
_buffers = new Vector< ByteBuffer >();
499
_buffers.add( buffer );
527
504
// name and params and the position in the buffer where the
528
// "value" part of the line starts
505
// "value" part of the line start
529
506
String name_and_params;
538
515
// skip some initial line characters, depending on the type
539
516
// of multi-line we're handling
540
pos = content_line.getBuffer().position();
517
pos = buffer.position();
541
518
switch( _parser_multiline_state )
543
520
case MULTILINE_FOLDED:
546
523
case MULTILINE_ENCODED:
547
while( pos < content_line.getBuffer().limit() && (
548
content_line.getBuffer().get( pos ) == ' ' ||
549
content_line.getBuffer().get( pos ) == '\t' ) )
524
while( pos < buffer.limit() && (
525
buffer.get( pos ) == ' ' ||
526
buffer.get( pos ) == '\t' ) )
565
if( content_line.getUsAsciiLine().trim().length() == 0 )
568
541
// get name and params from line, and since we're not
569
542
// parsing a subsequent line in a multi-line, this should
570
543
// not fail, or it's an error
571
544
name_and_params =
572
extractNameAndParamsFromLine( content_line );
545
extractNameAndParamsFromLine( buffer, line );
573
546
if( name_and_params == null )
574
547
throw new ParseException(
575
548
R.string.error_vcf_malformed );
577
550
// calculate how many chars to skip from beginning of line
578
551
// so we skip the property "name:" part
579
pos = content_line.getBuffer().position() +
580
name_and_params.length() + 1;
552
pos = buffer.position() + name_and_params.length() + 1;
582
554
// reset the saved multi-line state
583
555
_parser_current_name_and_params = name_and_params;
587
559
// get value from buffer, as raw bytes
588
560
ByteBuffer value;
589
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
590
content_line.getBuffer().limit() - pos );
561
value = ByteBuffer.wrap( buffer.array(), pos,
562
buffer.limit() - pos );
592
564
// get parameter parts
593
565
String[] name_param_parts = name_and_params.split( ";", -1 );
600
572
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
602
574
boolean is_interesting_field =
603
interesting_fields.contains(
604
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
575
interesting_fields.contains( name_param_parts[ 0 ] );
606
577
// parse encoding parameter
607
578
String encoding = checkParam( name_param_parts, "ENCODING" );
608
if( encoding != null )
609
encoding = encoding.toUpperCase( Locale.US );
579
if( encoding != null ) encoding = encoding.toUpperCase();
610
580
if( is_interesting_field && encoding != null &&
611
!encoding.equalsIgnoreCase( "8BIT" ) &&
612
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
613
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
581
!encoding.equals( "8BIT" ) &&
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
583
//&& !encoding.equals( "BASE64" ) )
615
585
throw new ParseException( R.string.error_vcf_encoding );
618
588
// parse charset parameter
619
589
String charset = checkParam( name_param_parts, "CHARSET" );
620
if( charset != null )
621
charset = charset.toUpperCase( Locale.US );
590
if( charset != null ) charset = charset.toUpperCase();
622
591
if( charset != null &&
623
!charset.equalsIgnoreCase( "US-ASCII" ) &&
624
!charset.equalsIgnoreCase( "ASCII" ) &&
625
!charset.equalsIgnoreCase( "UTF-8" ) )
592
!charset.equals( "US-ASCII" ) &&
593
!charset.equals( "ASCII" ) &&
594
!charset.equals( "UTF-8" ) )
627
596
throw new ParseException( R.string.error_vcf_charset );
630
599
// do unencoding (or default to a fake unencoding result with
631
600
// the raw string)
632
601
UnencodeResult unencoding_result = null;
633
if( encoding != null &&
634
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
602
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
636
603
unencoding_result = unencodeQuotedPrintable( value );
638
// else if( encoding != null &&
639
// encoding.equalsIgnoreCase( "BASE64" ) )
604
// else if( encoding != null && encoding.equals( "BASE64" ) )
641
605
// unencoding_result = unencodeBase64( props[ 1 ], charset );
643
606
if( unencoding_result != null ) {
644
607
value = unencoding_result.getBuffer();
645
608
if( unencoding_result.isAnotherLineRequired() )
650
613
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
651
614
if( ( charset == null && _version.equals( "2.1" ) ) ||
652
615
( charset != null && (
653
charset.equalsIgnoreCase( "ASCII" ) ||
654
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
616
charset.equals( "ASCII" ) ||
617
charset.equals( "US-ASCII" ) ) ) )
656
619
value = transcodeAsciiToUtf8( value );
668
631
// for some entries that have semicolon-separated value parts,
669
632
// check to see if the value ends in an escape character, which
670
633
// indicates that we have a multi-line value
671
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
672
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
673
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
634
if( ( name_param_parts[ 0 ].equals( "N" ) ||
635
name_param_parts[ 0 ].equals( "ORG" ) ||
636
name_param_parts[ 0 ].equals( "ADR" ) ) &&
674
637
doesStringEndInAnEscapeChar( string_value ) )
676
639
_parser_multiline_state = MULTILINE_ESCAPED;
678
641
string_value.length() - 1 );
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
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
683
646
if( _parser_multiline_state == MULTILINE_NONE &&
684
content_line.doesNextLineLookFolded() )
647
_version.equals( "3.0" ) && next_line_looks_folded )
686
649
_parser_multiline_state = MULTILINE_FOLDED;
699
662
if( complete_value.length() < 1 ) return;
701
664
// parse some properties
702
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
665
if( name_param_parts[ 0 ].equals( "N" ) )
703
666
parseN( name_param_parts, complete_value );
704
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
667
else if( name_param_parts[ 0 ].equals( "FN" ) )
705
668
parseFN( name_param_parts, complete_value );
706
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
669
else if( name_param_parts[ 0 ].equals( "ORG" ) )
707
670
parseORG( name_param_parts, complete_value );
708
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
671
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
709
672
parseTITLE( name_param_parts, complete_value );
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
673
else if( name_param_parts[ 0 ].equals( "TEL" ) )
711
674
parseTEL( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
675
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
713
676
parseEMAIL( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
677
else if( name_param_parts[ 0 ].equals( "ADR" ) )
715
678
parseADR( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
679
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
717
680
parseLABEL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
719
parseNOTE( name_param_parts, complete_value );
925
880
if( types.contains( "FAX" ) )
926
881
if( types.contains( "HOME" ) )
927
type = TYPE_FAX_HOME;
882
type = PhonesColumns.TYPE_FAX_HOME;
929
type = TYPE_FAX_WORK;
884
type = PhonesColumns.TYPE_FAX_WORK;
930
885
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
886
type = PhonesColumns.TYPE_MOBILE;
932
887
else if( types.contains( "PAGER" ) )
888
type = PhonesColumns.TYPE_PAGER;
934
889
else if( types.contains( "WORK" ) )
890
type = PhonesColumns.TYPE_WORK;
892
type = PhonesColumns.TYPE_HOME;
939
894
// add phone number
940
895
addNumber( value, type, is_preferred );
951
906
boolean is_preferred = types.contains( "PREF" );
953
908
if( types.contains( "WORK" ) )
909
type = Contacts.ContactMethods.TYPE_WORK;
911
type = Contacts.ContactMethods.TYPE_HOME;
958
913
addEmail( unescapeValue( value ), type, is_preferred );
968
923
for( int a = 0; a < adr_parts.length; a++ )
969
924
if( adr_parts[ a ].length() > 0 )
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 ];
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 ];
993
937
Set< String > types = extractTypes( params, Arrays.asList(
1013
957
if( types.contains( "WORK" ) )
958
type = Contacts.ContactMethods.TYPE_WORK;
960
type = Contacts.ContactMethods.TYPE_HOME;
1018
962
addAddress( unescapeValue( value ), type );
1021
private void parseNOTE( String[] params, String value )
1023
addNote( unescapeValue( value ) );
1026
965
public void finaliseVcard()
1027
966
throws ParseException, ContactNotIdentifiableException
1029
968
// missing version (and data is present)
1030
if( _version == null && _content_lines != null )
969
if( _version == null && _buffers != null )
1031
970
throw new ParseException( R.string.error_vcf_malformed );
1033
972
// finalise the parent class
1058
997
HashSet< String > ret = new HashSet< String >();
1060
999
Pattern p = Pattern.compile(
1061
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1062
Pattern.CASE_INSENSITIVE );
1000
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1063
1001
for( int i = 0; i < params.length; i++ ) {
1064
1002
Matcher m = p.matcher( params[ i ] );
1065
1003
if( m.matches() )
1073
1011
* Amongst the params, return any type values present. For v2.1 vCards,
1074
1012
* those types are just parameters. For v3.0, they are prefixed with
1075
1013
* "TYPE=". There may also be multiple type parameters.
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
1015
* @param a list of type values to look for
1078
1016
* @return a set of present type values
1080
1018
private Set< String > extractTypes( String[] params,
1086
1024
String type_params[] = checkParams( params, "TYPE" );
1087
1025
for( int a = 0; a < type_params.length; a++ )
1089
// check for a comma-separated list of types (why? I don't think
1090
// this is in the specs!)
1027
// check for a comma-separated list of types (why? this isn't in
1091
1029
String[] parts = type_params[ a ].split( "," );
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 );
1030
for( int i = 0; i < parts.length; i++ )
1031
if( valid_types.contains( parts[ i ] ) )
1032
types.add( parts[ i ] );
1099
1035
// get 2.1-style type param
1100
1036
if( _version.equals( "2.1" ) ) {
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 );
1037
for( int i = 1; i < params.length; i++ )
1038
if( valid_types.contains( params[ i ] ) )
1039
types.add( params[ i ] );