4
* Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
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
* 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
133
133
boolean in_vcard = false;
134
134
while( ( line = reader.readLine() ) != null )
138
137
// look for vcard beginning
139
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
138
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.*" ) )
143
else if( line.matches( "^END:VCARD" ) )
150
144
in_vcard = false;
155
148
catch( FileNotFoundException e ) {
177
170
FileInputStream istream = new FileInputStream( file );
178
171
byte[] content = new byte[ (int)file.length() ];
179
172
istream.read( content );
183
176
importVCardFileContent( content, file.getName() );
185
catch( OutOfMemoryError e ) {
186
showError( R.string.error_outofmemory );
188
178
catch( FileNotFoundException e ) {
189
179
showError( getText( R.string.error_filenotfound ) +
190
180
file.getName() );
203
193
ContentLineIterator cli = new ContentLineIterator( content );
204
194
while( cli.hasNext() )
206
ContentLine content_line = cli.next();
196
ByteBuffer buffer = cli.next();
208
// get a US-ASCII version of the string, for processing
209
String line = content_line.getUsAsciiLine();
198
// get a US-ASCII version of the line for processing
201
line = new String( buffer.array(), buffer.position(),
202
buffer.limit() - buffer.position(), "US-ASCII" );
204
catch( UnsupportedEncodingException e ) {
205
// we know US-ASCII is supported, so appease the compiler...
211
209
if( vcard == null ) {
212
210
// look for vcard beginning
213
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
211
if( line.matches( "^BEGIN:VCARD" ) ) {
214
212
setProgress( _progress++ );
215
213
vcard = new Vcard();
216
214
vcard_start_line = cli.getLineNumber();
262
260
// try giving the line to the vcard
264
vcard.parseLine( content_line );
262
vcard.parseLine( buffer, line,
263
cli.doesNextLineLookFolded() );
266
265
catch( Vcard.ParseException e ) {
274
273
finish( ACTION_ABORT );
277
// Although we're continuing, we still need to abort
278
// 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
279
278
// get to another BEGIN:VCARD line.
282
281
catch( Vcard.SkipImportException e ) {
284
// Abort this vCard. Further lines will be ignored until
283
// abort this vCard. Further lines will be ignored until
285
284
// we get to another BEGIN:VCARD line.
295
private ByteBuffer _buffer;
296
private boolean _folded_next;
297
private String _line;
299
public ContentLine( ByteBuffer buffer, boolean folded_next )
302
_folded_next = folded_next;
306
public ByteBuffer getBuffer()
311
public boolean doesNextLineLookFolded()
316
public String getUsAsciiLine()
318
// generated line and cache it
319
if( _line == null ) {
321
_line = new String( _buffer.array(), _buffer.position(),
322
_buffer.limit() - _buffer.position(), "US-ASCII" );
324
catch( UnsupportedEncodingException e ) {
325
// we know US-ASCII *is* supported, so appease the
330
// return cached line
335
class ContentLineIterator implements Iterator< ContentLine >
292
class ContentLineIterator implements Iterator< ByteBuffer >
337
294
protected byte[] _content = null;
338
295
protected int _pos = 0;
363
320
_pos > initial_pos )? _pos - 1 : _pos;
366
return new ContentLine(
367
ByteBuffer.wrap( _content, initial_pos,
369
doesNextLineLookFolded() );
323
return ByteBuffer.wrap( _content, initial_pos,
372
327
// we didn't find one, but were there bytes left?
377
return new ContentLine(
378
ByteBuffer.wrap( _content, initial_pos,
380
doesNextLineLookFolded() );
332
return ByteBuffer.wrap( _content, initial_pos,
395
348
* onto the end of this one?
398
private boolean doesNextLineLookFolded()
351
public boolean doesNextLineLookFolded()
400
353
return _pos > 0 && _pos < _content.length &&
401
_content[ _pos - 1 ] == '\n' &&
402
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
354
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
405
357
public int getLineNumber()
417
369
private final static int MULTILINE_NONE = 0;
418
370
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
419
371
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
420
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
372
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
422
374
private String _version = null;
423
private Vector< ContentLine > _content_lines = null;
375
private Vector< ByteBuffer > _buffers = null;
424
376
private int _name_level = NAMELEVEL_NONE;
425
377
private int _parser_multiline_state = MULTILINE_NONE;
426
378
private String _parser_current_name_and_params = null;
469
421
@SuppressWarnings("serial")
470
422
protected class SkipImportException extends Exception { }
472
private String extractCollonPartFromLine( ContentLine content_line,
424
private String extractCollonPartFromLine( ByteBuffer buffer,
425
String line, boolean former )
429
// get a US-ASCII version of the line for processing, unless we were
433
line = new String( buffer.array(), buffer.position(),
434
buffer.limit() - buffer.position(), "US-ASCII" );
436
catch( UnsupportedEncodingException e ) {
437
// we know US-ASCII is supported, so appease the compiler...
475
442
// split line into name and value parts and check to make sure we
476
443
// only got 2 parts and that the first part is not zero in length
477
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
444
String[] parts = line.split( ":", 2 );
478
445
if( parts.length == 2 && parts[ 0 ].length() > 0 )
479
return parts[ former? 0 : 1 ].trim();
484
private String extractNameAndParamsFromLine( ContentLine content_line )
486
return extractCollonPartFromLine( content_line, true );
489
private String extractValueFromLine( ContentLine content_line )
491
return extractCollonPartFromLine( content_line, false );
494
public void parseLine( ContentLine content_line )
446
ret = parts[ former? 0 : 1 ];
451
private String extractNameAndParamsFromLine( ByteBuffer buffer,
454
return extractCollonPartFromLine( buffer, line, true );
457
private String extractValueFromLine( ByteBuffer buffer, String line )
459
return extractCollonPartFromLine( buffer, line, false );
462
public void parseLine( ByteBuffer buffer, String line,
463
boolean next_line_looks_folded )
495
464
throws ParseException, SkipImportException,
496
465
AbortImportException
501
470
// tentatively get name and params from line
502
471
String name_and_params =
503
extractNameAndParamsFromLine( content_line );
472
extractNameAndParamsFromLine( buffer, line );
505
474
// is it a version line?
506
475
if( name_and_params != null &&
507
name_and_params.equalsIgnoreCase( "VERSION" ) )
476
name_and_params.equals( "VERSION" ) )
510
String value = extractValueFromLine( content_line );
511
if( value == null || (
512
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
479
String value = extractValueFromLine( buffer, line );
480
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
514
481
throw new ParseException( R.string.error_vcf_version );
516
482
_version = value;
518
484
// parse any buffers we've been accumulating while we waited
520
if( _content_lines != null )
521
for( int i = 0; i < _content_lines.size(); i++ )
522
parseLine( _content_lines.get( i ) );
523
_content_lines = null;
486
if( _buffers != null )
487
for( int i = 0; i < _buffers.size(); i++ )
488
parseLine( _buffers.get( i ), null,
489
i + 1 < _buffers.size() &&
490
_buffers.get( i + 1 ).hasRemaining() &&
491
_buffers.get( i + 1 ).get(
492
_buffers.get( i + 1 ).position() ) == ' ' );
527
497
// no, so stash this line till we get a version
528
if( _content_lines == null )
529
_content_lines = new Vector< ContentLine >();
530
_content_lines.add( content_line );
498
if( _buffers == null )
499
_buffers = new Vector< ByteBuffer >();
500
_buffers.add( buffer );
535
505
// name and params and the position in the buffer where the
536
// "value" part of the line starts
506
// "value" part of the line start
537
507
String name_and_params;
546
516
// skip some initial line characters, depending on the type
547
517
// of multi-line we're handling
548
pos = content_line.getBuffer().position();
518
pos = buffer.position();
549
519
switch( _parser_multiline_state )
551
521
case MULTILINE_FOLDED:
554
524
case MULTILINE_ENCODED:
555
while( pos < content_line.getBuffer().limit() && (
556
content_line.getBuffer().get( pos ) == ' ' ||
557
content_line.getBuffer().get( pos ) == '\t' ) )
525
while( pos < buffer.limit() && (
526
buffer.get( pos ) == ' ' ||
527
buffer.get( pos ) == '\t' ) )
573
if( content_line.getUsAsciiLine().trim().length() == 0 )
576
542
// get name and params from line, and since we're not
577
543
// parsing a subsequent line in a multi-line, this should
578
544
// not fail, or it's an error
579
545
name_and_params =
580
extractNameAndParamsFromLine( content_line );
546
extractNameAndParamsFromLine( buffer, line );
581
547
if( name_and_params == null )
582
548
throw new ParseException(
583
549
R.string.error_vcf_malformed );
585
551
// calculate how many chars to skip from beginning of line
586
552
// so we skip the property "name:" part
587
pos = content_line.getBuffer().position() +
588
name_and_params.length() + 1;
553
pos = buffer.position() + name_and_params.length() + 1;
590
555
// reset the saved multi-line state
591
556
_parser_current_name_and_params = name_and_params;
595
560
// get value from buffer, as raw bytes
596
561
ByteBuffer value;
597
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
598
content_line.getBuffer().limit() - pos );
562
value = ByteBuffer.wrap( buffer.array(), pos,
563
buffer.limit() - pos );
600
565
// get parameter parts
601
566
String[] name_param_parts = name_and_params.split( ";", -1 );
608
573
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
610
575
boolean is_interesting_field =
611
interesting_fields.contains(
612
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
576
interesting_fields.contains( name_param_parts[ 0 ] );
614
578
// parse encoding parameter
615
579
String encoding = checkParam( name_param_parts, "ENCODING" );
616
580
if( encoding != null )
617
581
encoding = encoding.toUpperCase( Locale.US );
618
582
if( is_interesting_field && encoding != null &&
619
!encoding.equalsIgnoreCase( "8BIT" ) &&
620
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
621
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
583
!encoding.equals( "8BIT" ) &&
584
!encoding.equals( "QUOTED-PRINTABLE" ) )
585
//&& !encoding.equals( "BASE64" ) )
623
587
throw new ParseException( R.string.error_vcf_encoding );
628
592
if( charset != null )
629
593
charset = charset.toUpperCase( Locale.US );
630
594
if( charset != null &&
631
!charset.equalsIgnoreCase( "US-ASCII" ) &&
632
!charset.equalsIgnoreCase( "ASCII" ) &&
633
!charset.equalsIgnoreCase( "UTF-8" ) )
595
!charset.equals( "US-ASCII" ) &&
596
!charset.equals( "ASCII" ) &&
597
!charset.equals( "UTF-8" ) )
635
599
throw new ParseException( R.string.error_vcf_charset );
638
602
// do unencoding (or default to a fake unencoding result with
639
603
// the raw string)
640
604
UnencodeResult unencoding_result = null;
641
if( encoding != null &&
642
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
605
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
644
606
unencoding_result = unencodeQuotedPrintable( value );
646
// else if( encoding != null &&
647
// encoding.equalsIgnoreCase( "BASE64" ) )
607
// else if( encoding != null && encoding.equals( "BASE64" ) )
649
608
// unencoding_result = unencodeBase64( props[ 1 ], charset );
651
609
if( unencoding_result != null ) {
652
610
value = unencoding_result.getBuffer();
653
611
if( unencoding_result.isAnotherLineRequired() )
658
616
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
659
617
if( ( charset == null && _version.equals( "2.1" ) ) ||
660
618
( charset != null && (
661
charset.equalsIgnoreCase( "ASCII" ) ||
662
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
619
charset.equals( "ASCII" ) ||
620
charset.equals( "US-ASCII" ) ) ) )
664
622
value = transcodeAsciiToUtf8( value );
676
634
// for some entries that have semicolon-separated value parts,
677
635
// check to see if the value ends in an escape character, which
678
636
// indicates that we have a multi-line value
679
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
680
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
681
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
637
if( ( name_param_parts[ 0 ].equals( "N" ) ||
638
name_param_parts[ 0 ].equals( "ORG" ) ||
639
name_param_parts[ 0 ].equals( "ADR" ) ) &&
682
640
doesStringEndInAnEscapeChar( string_value ) )
684
642
_parser_multiline_state = MULTILINE_ESCAPED;
686
644
string_value.length() - 1 );
689
// if we know we're not in an encoding-based multi-line, check
690
// to see if we're in a folded multi-line
647
// now we know whether we're in an encoding multi-line,
648
// determine if we're in a v3 folded multi-line or not
691
649
if( _parser_multiline_state == MULTILINE_NONE &&
692
content_line.doesNextLineLookFolded() )
650
_version.equals( "3.0" ) && next_line_looks_folded )
694
652
_parser_multiline_state = MULTILINE_FOLDED;
707
665
if( complete_value.length() < 1 ) return;
709
667
// parse some properties
710
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
668
if( name_param_parts[ 0 ].equals( "N" ) )
711
669
parseN( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
670
else if( name_param_parts[ 0 ].equals( "FN" ) )
713
671
parseFN( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
672
else if( name_param_parts[ 0 ].equals( "ORG" ) )
715
673
parseORG( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
674
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
717
675
parseTITLE( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
676
else if( name_param_parts[ 0 ].equals( "TEL" ) )
719
677
parseTEL( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
678
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
721
679
parseEMAIL( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
680
else if( name_param_parts[ 0 ].equals( "ADR" ) )
723
681
parseADR( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
682
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
725
683
parseLABEL( name_param_parts, complete_value );
726
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
727
parseNOTE( name_param_parts, complete_value );
728
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
729
parseBDAY( name_param_parts, complete_value );
757
711
String str = parts.get( a );
759
// Look for parts that end in an escape character, but ignore
760
// the final part. We've already detected escape chars at the
713
// look for parts that end in an escape character, but ignore
714
// the final part. We've already detected escape chars at the
761
715
// end of the final part in parseLine() and handled multi-lines
763
717
if( a < parts.size() - 1 &&
853
801
for( int b = 0; b < name_part_parts.length; b++ )
854
802
if( name_part_parts[ b ].length() > 0 )
856
if( value.length() > 0 ) value += " ";
804
if( value.length() == 0 ) value += " ";
857
805
value += name_part_parts[ b ];
978
926
for( int a = 0; a < adr_parts.length; a++ )
979
927
if( adr_parts[ a ].length() > 0 )
981
// version 3.0 vCards allow further splitting by comma
982
if( _version.equals( "3.0" ) )
984
// split this part in to it's comma-separated bits and
985
// add them on individual lines
986
String[] adr_part_parts =
987
splitValueByCharacter( adr_parts[ a ], ',' );
988
for( int b = 0; b < adr_part_parts.length; b++ )
989
if( adr_part_parts[ b ].length() > 0 )
991
if( value.length() > 0 ) value += "\n";
992
value += adr_part_parts[ b ];
997
// add this part on an individual line
998
if( value.length() > 0 ) value += "\n";
999
value += adr_parts[ a ];
929
// split this part in to it's comma-separated bits
930
String[] adr_part_parts =
931
splitValueByCharacter( adr_parts[ a ], ',' );
932
for( int b = 0; b < adr_part_parts.length; b++ )
933
if( adr_part_parts[ b ].length() > 0 )
935
if( value.length() > 0 ) value += "\n";
936
value += adr_part_parts[ b ];
1003
940
Set< String > types = extractTypes( params, Arrays.asList(
1028
965
addAddress( unescapeValue( value ), type );
1031
private void parseNOTE( String[] params, String value )
1033
addNote( unescapeValue( value ) );
1036
private void parseBDAY( String[] params, String value )
1038
setBirthday( value );
1041
968
public void finaliseVcard()
1042
969
throws ParseException, ContactNotIdentifiableException
1044
971
// missing version (and data is present)
1045
if( _version == null && _content_lines != null )
972
if( _version == null && _buffers != null )
1046
973
throw new ParseException( R.string.error_vcf_malformed );
1048
975
// finalise the parent class
1075
1000
HashSet< String > ret = new HashSet< String >();
1077
1002
Pattern p = Pattern.compile(
1078
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1079
Pattern.CASE_INSENSITIVE );
1003
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1080
1004
for( int i = 0; i < params.length; i++ ) {
1081
1005
Matcher m = p.matcher( params[ i ] );
1082
1006
if( m.matches() )
1090
* Amongst the params, return any type values present. For v2.1 vCards,
1091
* those types are just parameters. For v3.0, they are prefixed with
1092
* "TYPE=". There may also be multiple type parameters.
1094
* @param params an array of params to look for types in
1095
* @param valid_types an list of upper-case type values to look for
1014
* Amongst the params, return any type values present. For v2.1 vCards,
1015
* those types are just parameters. For v3.0, they are prefixed with
1016
* "TYPE=". There may also be multiple type parameters.
1018
* @param a list of type values to look for
1096
1019
* @return a set of present type values
1098
1021
private Set< String > extractTypes( String[] params,
1104
1027
String type_params[] = checkParams( params, "TYPE" );
1105
1028
for( int a = 0; a < type_params.length; a++ )
1107
// check for a comma-separated list of types (why? I don't think
1108
// this is in the specs!)
1030
// check for a comma-separated list of types (why? this isn't in
1109
1032
String[] parts = type_params[ a ].split( "," );
1110
for( int i = 0; i < parts.length; i++ ) {
1111
String ucpart = parts[ i ].toUpperCase( Locale.US );
1112
if( valid_types.contains( ucpart ) )
1113
types.add( ucpart );
1033
for( int i = 0; i < parts.length; i++ )
1034
if( valid_types.contains( parts[ i ] ) )
1035
types.add( parts[ i ] );
1117
1038
// get 2.1-style type param
1118
1039
if( _version.equals( "2.1" ) ) {
1119
for( int i = 1; i < params.length; i++ ) {
1120
String ucparam = params[ i ].toUpperCase( Locale.US );
1121
if( valid_types.contains( ucparam ) )
1122
types.add( ucparam );
1040
for( int i = 1; i < params.length; i++ )
1041
if( valid_types.contains( params[ i ] ) )
1042
types.add( params[ i ] );
1148
1067
else if( ch == '=' && i == in.limit() - 1 )
1150
1069
// we found a '=' at the end of a line signifying a multi-
1151
// line string, so we don't add it
1070
// line string, so we don't add it.
1152
1071
another = true;