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
* to as "this program"). For more information, see
8
* http://ed.am/dev/android/import-contacts
7
* to as "this program"). For more information, see
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() );
133
132
boolean in_vcard = false;
134
133
while( ( line = reader.readLine() ) != null )
138
136
// look for vcard beginning
139
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
137
if( line.matches( "^BEGIN:VCARD" ) ) {
143
// check for vMsg files
144
else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
145
showError( getText( R.string.error_vcf_vmsgfile )
149
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
142
else if( line.matches( "^END:VCARD" ) )
150
143
in_vcard = false;
182
175
importVCardFileContent( content, file.getName() );
184
catch( OutOfMemoryError e ) {
185
showError( R.string.error_outofmemory );
187
177
catch( FileNotFoundException e ) {
188
178
showError( getText( R.string.error_filenotfound ) +
189
179
file.getName() );
202
192
ContentLineIterator cli = new ContentLineIterator( content );
203
193
while( cli.hasNext() )
205
ContentLine content_line = cli.next();
195
ByteBuffer buffer = cli.next();
207
// get a US-ASCII version of the string, for processing
208
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...
210
208
if( vcard == null ) {
211
209
// look for vcard beginning
212
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
210
if( line.matches( "^BEGIN:VCARD" ) ) {
213
211
setProgress( _progress++ );
214
212
vcard = new Vcard();
215
213
vcard_start_line = cli.getLineNumber();
261
259
// try giving the line to the vcard
263
vcard.parseLine( content_line );
261
vcard.parseLine( buffer, line,
262
cli.doesNextLineLookFolded() );
265
264
catch( Vcard.ParseException e ) {
273
272
finish( ACTION_ABORT );
276
// Although we're continuing, we still need to abort
277
// this vCard. Further lines will be ignored until we
275
// although we're continuing, we still need to abort
276
// this vCard. Further lines will be ignored until we
278
277
// get to another BEGIN:VCARD line.
281
280
catch( Vcard.SkipImportException e ) {
283
// Abort this vCard. Further lines will be ignored until
282
// abort this vCard. Further lines will be ignored until
284
283
// we get to another BEGIN:VCARD line.
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 >
291
class ContentLineIterator implements Iterator< ByteBuffer >
336
293
protected byte[] _content = null;
337
294
protected int _pos = 0;
362
319
_pos > initial_pos )? _pos - 1 : _pos;
365
return new ContentLine(
366
ByteBuffer.wrap( _content, initial_pos,
368
doesNextLineLookFolded() );
322
return ByteBuffer.wrap( _content, initial_pos,
371
326
// we didn't find one, but were there bytes left?
376
return new ContentLine(
377
ByteBuffer.wrap( _content, initial_pos,
379
doesNextLineLookFolded() );
331
return ByteBuffer.wrap( _content, initial_pos,
394
347
* onto the end of this one?
397
private boolean doesNextLineLookFolded()
350
public boolean doesNextLineLookFolded()
399
352
return _pos > 0 && _pos < _content.length &&
400
_content[ _pos - 1 ] == '\n' &&
401
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
353
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
404
356
public int getLineNumber()
416
368
private final static int MULTILINE_NONE = 0;
417
369
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
418
370
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
419
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
371
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
421
373
private String _version = null;
422
private Vector< ContentLine > _content_lines = null;
374
private Vector< ByteBuffer > _buffers = null;
423
375
private int _name_level = NAMELEVEL_NONE;
424
376
private int _parser_multiline_state = MULTILINE_NONE;
425
377
private String _parser_current_name_and_params = null;
468
420
@SuppressWarnings("serial")
469
421
protected class SkipImportException extends Exception { }
471
private String extractCollonPartFromLine( ContentLine content_line,
423
private String extractCollonPartFromLine( ByteBuffer buffer,
424
String line, boolean former )
474
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...
476
441
// split line into name and value parts and check to make sure we
477
442
// only got 2 parts and that the first part is not zero in length
478
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
443
String[] parts = line.split( ":", 2 );
479
444
if( parts.length == 2 && parts[ 0 ].length() > 0 )
480
445
ret = parts[ former? 0 : 1 ];
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 )
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 )
496
463
throws ParseException, SkipImportException,
497
464
AbortImportException
502
469
// tentatively get name and params from line
503
470
String name_and_params =
504
extractNameAndParamsFromLine( content_line );
471
extractNameAndParamsFromLine( buffer, line );
506
473
// is it a version line?
507
474
if( name_and_params != null &&
508
name_and_params.equalsIgnoreCase( "VERSION" ) )
475
name_and_params.equals( "VERSION" ) )
511
String value = extractValueFromLine( content_line ).trim();
478
String value = extractValueFromLine( buffer, line );
512
479
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
480
throw new ParseException( R.string.error_vcf_version );
514
481
_version = value;
516
483
// parse any buffers we've been accumulating while we waited
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;
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() ) == ' ' );
525
496
// no, so stash this line till we get a version
526
if( _content_lines == null )
527
_content_lines = new Vector< ContentLine >();
528
_content_lines.add( content_line );
497
if( _buffers == null )
498
_buffers = new Vector< ByteBuffer >();
499
_buffers.add( buffer );
533
504
// name and params and the position in the buffer where the
534
// "value" part of the line starts
505
// "value" part of the line start
535
506
String name_and_params;
544
515
// skip some initial line characters, depending on the type
545
516
// of multi-line we're handling
546
pos = content_line.getBuffer().position();
517
pos = buffer.position();
547
518
switch( _parser_multiline_state )
549
520
case MULTILINE_FOLDED:
552
523
case MULTILINE_ENCODED:
553
while( pos < content_line.getBuffer().limit() && (
554
content_line.getBuffer().get( pos ) == ' ' ||
555
content_line.getBuffer().get( pos ) == '\t' ) )
524
while( pos < buffer.limit() && (
525
buffer.get( pos ) == ' ' ||
526
buffer.get( pos ) == '\t' ) )
571
if( content_line.getUsAsciiLine().trim().length() == 0 )
574
541
// get name and params from line, and since we're not
575
542
// parsing a subsequent line in a multi-line, this should
576
543
// not fail, or it's an error
577
544
name_and_params =
578
extractNameAndParamsFromLine( content_line );
545
extractNameAndParamsFromLine( buffer, line );
579
546
if( name_and_params == null )
580
547
throw new ParseException(
581
548
R.string.error_vcf_malformed );
583
550
// calculate how many chars to skip from beginning of line
584
551
// so we skip the property "name:" part
585
pos = content_line.getBuffer().position() +
586
name_and_params.length() + 1;
552
pos = buffer.position() + name_and_params.length() + 1;
588
554
// reset the saved multi-line state
589
555
_parser_current_name_and_params = name_and_params;
593
559
// get value from buffer, as raw bytes
594
560
ByteBuffer value;
595
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
596
content_line.getBuffer().limit() - pos );
561
value = ByteBuffer.wrap( buffer.array(), pos,
562
buffer.limit() - pos );
598
564
// get parameter parts
599
565
String[] name_param_parts = name_and_params.split( ";", -1 );
606
572
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
608
574
boolean is_interesting_field =
609
interesting_fields.contains(
610
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
575
interesting_fields.contains( name_param_parts[ 0 ] );
612
577
// parse encoding parameter
613
578
String encoding = checkParam( name_param_parts, "ENCODING" );
614
if( encoding != null )
615
encoding = encoding.toUpperCase( Locale.US );
579
if( encoding != null ) encoding = encoding.toUpperCase();
616
580
if( is_interesting_field && encoding != null &&
617
!encoding.equalsIgnoreCase( "8BIT" ) &&
618
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
619
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
581
!encoding.equals( "8BIT" ) &&
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
583
//&& !encoding.equals( "BASE64" ) )
621
585
throw new ParseException( R.string.error_vcf_encoding );
624
588
// parse charset parameter
625
589
String charset = checkParam( name_param_parts, "CHARSET" );
626
if( charset != null )
627
charset = charset.toUpperCase( Locale.US );
590
if( charset != null ) charset = charset.toUpperCase();
628
591
if( charset != null &&
629
!charset.equalsIgnoreCase( "US-ASCII" ) &&
630
!charset.equalsIgnoreCase( "ASCII" ) &&
631
!charset.equalsIgnoreCase( "UTF-8" ) )
592
!charset.equals( "US-ASCII" ) &&
593
!charset.equals( "ASCII" ) &&
594
!charset.equals( "UTF-8" ) )
633
596
throw new ParseException( R.string.error_vcf_charset );
636
599
// do unencoding (or default to a fake unencoding result with
637
600
// the raw string)
638
601
UnencodeResult unencoding_result = null;
639
if( encoding != null &&
640
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
602
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
642
603
unencoding_result = unencodeQuotedPrintable( value );
644
// else if( encoding != null &&
645
// encoding.equalsIgnoreCase( "BASE64" ) )
604
// else if( encoding != null && encoding.equals( "BASE64" ) )
647
605
// unencoding_result = unencodeBase64( props[ 1 ], charset );
649
606
if( unencoding_result != null ) {
650
607
value = unencoding_result.getBuffer();
651
608
if( unencoding_result.isAnotherLineRequired() )
656
613
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
657
614
if( ( charset == null && _version.equals( "2.1" ) ) ||
658
615
( charset != null && (
659
charset.equalsIgnoreCase( "ASCII" ) ||
660
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
616
charset.equals( "ASCII" ) ||
617
charset.equals( "US-ASCII" ) ) ) )
662
619
value = transcodeAsciiToUtf8( value );
674
631
// for some entries that have semicolon-separated value parts,
675
632
// check to see if the value ends in an escape character, which
676
633
// indicates that we have a multi-line value
677
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
678
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
679
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" ) ) &&
680
637
doesStringEndInAnEscapeChar( string_value ) )
682
639
_parser_multiline_state = MULTILINE_ESCAPED;
684
641
string_value.length() - 1 );
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
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
689
646
if( _parser_multiline_state == MULTILINE_NONE &&
690
content_line.doesNextLineLookFolded() )
647
_version.equals( "3.0" ) && next_line_looks_folded )
692
649
_parser_multiline_state = MULTILINE_FOLDED;
705
662
if( complete_value.length() < 1 ) return;
707
664
// parse some properties
708
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
665
if( name_param_parts[ 0 ].equals( "N" ) )
709
666
parseN( name_param_parts, complete_value );
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
667
else if( name_param_parts[ 0 ].equals( "FN" ) )
711
668
parseFN( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
669
else if( name_param_parts[ 0 ].equals( "ORG" ) )
713
670
parseORG( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
671
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
715
672
parseTITLE( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
673
else if( name_param_parts[ 0 ].equals( "TEL" ) )
717
674
parseTEL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
675
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
719
676
parseEMAIL( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
677
else if( name_param_parts[ 0 ].equals( "ADR" ) )
721
678
parseADR( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
679
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
723
680
parseLABEL( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
725
parseNOTE( name_param_parts, complete_value );
753
708
String str = parts.get( a );
755
// Look for parts that end in an escape character, but ignore
756
// the final part. We've already detected escape chars at the
710
// look for parts that end in an escape character, but ignore
711
// the final part. We've already detected escape chars at the
757
712
// end of the final part in parseLine() and handled multi-lines
759
714
if( a < parts.size() - 1 &&
849
798
for( int b = 0; b < name_part_parts.length; b++ )
850
799
if( name_part_parts[ b ].length() > 0 )
852
if( value.length() > 0 ) value += " ";
801
if( value.length() == 0 ) value += " ";
853
802
value += name_part_parts[ b ];
931
880
if( types.contains( "FAX" ) )
932
881
if( types.contains( "HOME" ) )
933
type = TYPE_FAX_HOME;
882
type = PhonesColumns.TYPE_FAX_HOME;
935
type = TYPE_FAX_WORK;
884
type = PhonesColumns.TYPE_FAX_WORK;
936
885
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
886
type = PhonesColumns.TYPE_MOBILE;
938
887
else if( types.contains( "PAGER" ) )
888
type = PhonesColumns.TYPE_PAGER;
940
889
else if( types.contains( "WORK" ) )
890
type = PhonesColumns.TYPE_WORK;
892
type = PhonesColumns.TYPE_HOME;
945
894
// add phone number
946
895
addNumber( value, type, is_preferred );
957
906
boolean is_preferred = types.contains( "PREF" );
959
908
if( types.contains( "WORK" ) )
909
type = Contacts.ContactMethods.TYPE_WORK;
911
type = Contacts.ContactMethods.TYPE_HOME;
964
913
addEmail( unescapeValue( value ), type, is_preferred );
974
923
for( int a = 0; a < adr_parts.length; a++ )
975
924
if( adr_parts[ a ].length() > 0 )
977
// version 3.0 vCards allow further splitting by comma
978
if( _version.equals( "3.0" ) )
980
// split this part in to it's comma-separated bits and
981
// add them on individual lines
982
String[] adr_part_parts =
983
splitValueByCharacter( adr_parts[ a ], ',' );
984
for( int b = 0; b < adr_part_parts.length; b++ )
985
if( adr_part_parts[ b ].length() > 0 )
987
if( value.length() > 0 ) value += "\n";
988
value += adr_part_parts[ b ];
993
// add this part on an individual line
994
if( value.length() > 0 ) value += "\n";
995
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 ];
999
937
Set< String > types = extractTypes( params, Arrays.asList(
1019
957
if( types.contains( "WORK" ) )
958
type = Contacts.ContactMethods.TYPE_WORK;
960
type = Contacts.ContactMethods.TYPE_HOME;
1024
962
addAddress( unescapeValue( value ), type );
1027
private void parseNOTE( String[] params, String value )
1029
addNote( unescapeValue( value ) );
1032
965
public void finaliseVcard()
1033
966
throws ParseException, ContactNotIdentifiableException
1035
968
// missing version (and data is present)
1036
if( _version == null && _content_lines != null )
969
if( _version == null && _buffers != null )
1037
970
throw new ParseException( R.string.error_vcf_malformed );
1039
972
// finalise the parent class
1058
* Amongst the params, find the values of any with the specified name.
990
* Amongst the params, find the values of any with the specified name
1062
993
* @return an array of values, or null
1066
997
HashSet< String > ret = new HashSet< String >();
1068
999
Pattern p = Pattern.compile(
1069
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1070
Pattern.CASE_INSENSITIVE );
1000
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1071
1001
for( int i = 0; i < params.length; i++ ) {
1072
1002
Matcher m = p.matcher( params[ i ] );
1073
1003
if( m.matches() )
1081
* Amongst the params, return any type values present. For v2.1 vCards,
1082
* those types are just parameters. For v3.0, they are prefixed with
1083
* "TYPE=". There may also be multiple type parameters.
1085
* @param params an array of params to look for types in
1086
* @param valid_types an list of upper-case type values to look for
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
1087
1016
* @return a set of present type values
1089
1018
private Set< String > extractTypes( String[] params,
1095
1024
String type_params[] = checkParams( params, "TYPE" );
1096
1025
for( int a = 0; a < type_params.length; a++ )
1098
// check for a comma-separated list of types (why? I don't think
1099
// this is in the specs!)
1027
// check for a comma-separated list of types (why? this isn't in
1100
1029
String[] parts = type_params[ a ].split( "," );
1101
for( int i = 0; i < parts.length; i++ ) {
1102
String ucpart = parts[ i ].toUpperCase( Locale.US );
1103
if( valid_types.contains( ucpart ) )
1104
types.add( ucpart );
1030
for( int i = 0; i < parts.length; i++ )
1031
if( valid_types.contains( parts[ i ] ) )
1032
types.add( parts[ i ] );
1108
1035
// get 2.1-style type param
1109
1036
if( _version.equals( "2.1" ) ) {
1110
for( int i = 1; i < params.length; i++ ) {
1111
String ucparam = params[ i ].toUpperCase( Locale.US );
1112
if( valid_types.contains( ucparam ) )
1113
types.add( ucparam );
1037
for( int i = 1; i < params.length; i++ )
1038
if( valid_types.contains( params[ i ] ) )
1039
types.add( params[ i ] );
1139
1064
else if( ch == '=' && i == in.limit() - 1 )
1141
1066
// we found a '=' at the end of a line signifying a multi-
1142
// line string, so we don't add it
1067
// line string, so we don't add it.
1143
1068
another = true;