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
133
133
boolean in_vcard = false;
134
134
while( ( line = reader.readLine() ) != null )
137
138
// look for vcard beginning
138
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 )
143
else if( line.matches( "^END:VCARD" ) )
149
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
144
150
in_vcard = false;
148
155
catch( FileNotFoundException e ) {
170
177
FileInputStream istream = new FileInputStream( file );
171
178
byte[] content = new byte[ (int)file.length() ];
172
179
istream.read( content );
176
183
importVCardFileContent( content, file.getName() );
185
catch( OutOfMemoryError e ) {
186
showError( R.string.error_outofmemory );
178
188
catch( FileNotFoundException e ) {
179
189
showError( getText( R.string.error_filenotfound ) +
180
190
file.getName() );
193
203
ContentLineIterator cli = new ContentLineIterator( content );
194
204
while( cli.hasNext() )
196
ByteBuffer buffer = cli.next();
206
ContentLine content_line = cli.next();
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...
208
// get a US-ASCII version of the string, for processing
209
String line = content_line.getUsAsciiLine();
209
211
if( vcard == null ) {
210
212
// look for vcard beginning
211
if( line.matches( "^BEGIN:VCARD" ) ) {
213
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
212
214
setProgress( _progress++ );
213
215
vcard = new Vcard();
214
216
vcard_start_line = cli.getLineNumber();
260
262
// try giving the line to the vcard
262
vcard.parseLine( buffer, line,
263
cli.doesNextLineLookFolded() );
264
vcard.parseLine( content_line );
265
266
catch( Vcard.ParseException e ) {
273
274
finish( ACTION_ABORT );
276
// although we're continuing, we still need to abort
277
// this vCard. Further lines will be ignored until we
277
// Although we're continuing, we still need to abort
278
// this vCard. Further lines will be ignored until we
278
279
// get to another BEGIN:VCARD line.
281
282
catch( Vcard.SkipImportException e ) {
283
// abort this vCard. Further lines will be ignored until
284
// Abort this vCard. Further lines will be ignored until
284
285
// we get to another BEGIN:VCARD line.
292
class ContentLineIterator implements Iterator< ByteBuffer >
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 >
294
337
protected byte[] _content = null;
295
338
protected int _pos = 0;
320
363
_pos > initial_pos )? _pos - 1 : _pos;
323
return ByteBuffer.wrap( _content, initial_pos,
366
return new ContentLine(
367
ByteBuffer.wrap( _content, initial_pos,
369
doesNextLineLookFolded() );
327
372
// we didn't find one, but were there bytes left?
332
return ByteBuffer.wrap( _content, initial_pos,
377
return new ContentLine(
378
ByteBuffer.wrap( _content, initial_pos,
380
doesNextLineLookFolded() );
348
395
* onto the end of this one?
351
public boolean doesNextLineLookFolded()
398
private boolean doesNextLineLookFolded()
353
400
return _pos > 0 && _pos < _content.length &&
354
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
401
_content[ _pos - 1 ] == '\n' &&
402
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
357
405
public int getLineNumber()
369
417
private final static int MULTILINE_NONE = 0;
370
418
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
371
419
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
372
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
420
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
374
422
private String _version = null;
375
private Vector< ByteBuffer > _buffers = null;
423
private Vector< ContentLine > _content_lines = null;
376
424
private int _name_level = NAMELEVEL_NONE;
377
425
private int _parser_multiline_state = MULTILINE_NONE;
378
426
private String _parser_current_name_and_params = null;
421
469
@SuppressWarnings("serial")
422
470
protected class SkipImportException extends Exception { }
424
private String extractCollonPartFromLine( ByteBuffer buffer,
425
String line, boolean former )
472
private String extractCollonPartFromLine( ContentLine content_line,
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...
442
475
// split line into name and value parts and check to make sure we
443
476
// only got 2 parts and that the first part is not zero in length
444
String[] parts = line.split( ":", 2 );
477
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
445
478
if( parts.length == 2 && parts[ 0 ].length() > 0 )
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 )
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 )
464
495
throws ParseException, SkipImportException,
465
496
AbortImportException
470
501
// tentatively get name and params from line
471
502
String name_and_params =
472
extractNameAndParamsFromLine( buffer, line );
503
extractNameAndParamsFromLine( content_line );
474
505
// is it a version line?
475
506
if( name_and_params != null &&
476
name_and_params.equals( "VERSION" ) )
507
name_and_params.equalsIgnoreCase( "VERSION" ) )
479
String value = extractValueFromLine( buffer, line );
480
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
510
String value = extractValueFromLine( content_line );
511
if( value == null || (
512
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
481
514
throw new ParseException( R.string.error_vcf_version );
482
516
_version = value;
484
518
// parse any buffers we've been accumulating while we waited
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() ) == ' ' );
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;
497
527
// no, so stash this line till we get a version
498
if( _buffers == null )
499
_buffers = new Vector< ByteBuffer >();
500
_buffers.add( buffer );
528
if( _content_lines == null )
529
_content_lines = new Vector< ContentLine >();
530
_content_lines.add( content_line );
505
535
// name and params and the position in the buffer where the
506
// "value" part of the line start
536
// "value" part of the line starts
507
537
String name_and_params;
516
546
// skip some initial line characters, depending on the type
517
547
// of multi-line we're handling
518
pos = buffer.position();
548
pos = content_line.getBuffer().position();
519
549
switch( _parser_multiline_state )
521
551
case MULTILINE_FOLDED:
524
554
case MULTILINE_ENCODED:
525
while( pos < buffer.limit() && (
526
buffer.get( pos ) == ' ' ||
527
buffer.get( pos ) == '\t' ) )
555
while( pos < content_line.getBuffer().limit() && (
556
content_line.getBuffer().get( pos ) == ' ' ||
557
content_line.getBuffer().get( pos ) == '\t' ) )
573
if( content_line.getUsAsciiLine().trim().length() == 0 )
542
576
// get name and params from line, and since we're not
543
577
// parsing a subsequent line in a multi-line, this should
544
578
// not fail, or it's an error
545
579
name_and_params =
546
extractNameAndParamsFromLine( buffer, line );
580
extractNameAndParamsFromLine( content_line );
547
581
if( name_and_params == null )
548
582
throw new ParseException(
549
583
R.string.error_vcf_malformed );
551
585
// calculate how many chars to skip from beginning of line
552
586
// so we skip the property "name:" part
553
pos = buffer.position() + name_and_params.length() + 1;
587
pos = content_line.getBuffer().position() +
588
name_and_params.length() + 1;
555
590
// reset the saved multi-line state
556
591
_parser_current_name_and_params = name_and_params;
560
595
// get value from buffer, as raw bytes
561
596
ByteBuffer value;
562
value = ByteBuffer.wrap( buffer.array(), pos,
563
buffer.limit() - pos );
597
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
598
content_line.getBuffer().limit() - pos );
565
600
// get parameter parts
566
601
String[] name_param_parts = name_and_params.split( ";", -1 );
573
608
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
575
610
boolean is_interesting_field =
576
interesting_fields.contains( name_param_parts[ 0 ] );
611
interesting_fields.contains(
612
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
578
614
// parse encoding parameter
579
615
String encoding = checkParam( name_param_parts, "ENCODING" );
580
616
if( encoding != null )
581
617
encoding = encoding.toUpperCase( Locale.US );
582
618
if( is_interesting_field && encoding != null &&
583
!encoding.equals( "8BIT" ) &&
584
!encoding.equals( "QUOTED-PRINTABLE" ) )
585
//&& !encoding.equals( "BASE64" ) )
619
!encoding.equalsIgnoreCase( "8BIT" ) &&
620
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
621
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
587
623
throw new ParseException( R.string.error_vcf_encoding );
592
628
if( charset != null )
593
629
charset = charset.toUpperCase( Locale.US );
594
630
if( charset != null &&
595
!charset.equals( "US-ASCII" ) &&
596
!charset.equals( "ASCII" ) &&
597
!charset.equals( "UTF-8" ) )
631
!charset.equalsIgnoreCase( "US-ASCII" ) &&
632
!charset.equalsIgnoreCase( "ASCII" ) &&
633
!charset.equalsIgnoreCase( "UTF-8" ) )
599
635
throw new ParseException( R.string.error_vcf_charset );
602
638
// do unencoding (or default to a fake unencoding result with
603
639
// the raw string)
604
640
UnencodeResult unencoding_result = null;
605
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
641
if( encoding != null &&
642
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
606
644
unencoding_result = unencodeQuotedPrintable( value );
607
// else if( encoding != null && encoding.equals( "BASE64" ) )
646
// else if( encoding != null &&
647
// encoding.equalsIgnoreCase( "BASE64" ) )
608
649
// unencoding_result = unencodeBase64( props[ 1 ], charset );
609
651
if( unencoding_result != null ) {
610
652
value = unencoding_result.getBuffer();
611
653
if( unencoding_result.isAnotherLineRequired() )
616
658
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
617
659
if( ( charset == null && _version.equals( "2.1" ) ) ||
618
660
( charset != null && (
619
charset.equals( "ASCII" ) ||
620
charset.equals( "US-ASCII" ) ) ) )
661
charset.equalsIgnoreCase( "ASCII" ) ||
662
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
622
664
value = transcodeAsciiToUtf8( value );
634
676
// for some entries that have semicolon-separated value parts,
635
677
// check to see if the value ends in an escape character, which
636
678
// indicates that we have a multi-line value
637
if( ( name_param_parts[ 0 ].equals( "N" ) ||
638
name_param_parts[ 0 ].equals( "ORG" ) ||
639
name_param_parts[ 0 ].equals( "ADR" ) ) &&
679
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
680
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
681
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
640
682
doesStringEndInAnEscapeChar( string_value ) )
642
684
_parser_multiline_state = MULTILINE_ESCAPED;
644
686
string_value.length() - 1 );
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
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
649
691
if( _parser_multiline_state == MULTILINE_NONE &&
650
_version.equals( "3.0" ) && next_line_looks_folded )
692
content_line.doesNextLineLookFolded() )
652
694
_parser_multiline_state = MULTILINE_FOLDED;
665
707
if( complete_value.length() < 1 ) return;
667
709
// parse some properties
668
if( name_param_parts[ 0 ].equals( "N" ) )
710
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
669
711
parseN( name_param_parts, complete_value );
670
else if( name_param_parts[ 0 ].equals( "FN" ) )
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
671
713
parseFN( name_param_parts, complete_value );
672
else if( name_param_parts[ 0 ].equals( "ORG" ) )
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
673
715
parseORG( name_param_parts, complete_value );
674
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
675
717
parseTITLE( name_param_parts, complete_value );
676
else if( name_param_parts[ 0 ].equals( "TEL" ) )
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
677
719
parseTEL( name_param_parts, complete_value );
678
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
679
721
parseEMAIL( name_param_parts, complete_value );
680
else if( name_param_parts[ 0 ].equals( "ADR" ) )
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
681
723
parseADR( name_param_parts, complete_value );
682
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
683
725
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 );
711
757
String str = parts.get( a );
713
// look for parts that end in an escape character, but ignore
714
// the final part. We've already detected escape chars at the
759
// Look for parts that end in an escape character, but ignore
760
// the final part. We've already detected escape chars at the
715
761
// end of the final part in parseLine() and handled multi-lines
717
763
if( a < parts.size() - 1 &&
801
853
for( int b = 0; b < name_part_parts.length; b++ )
802
854
if( name_part_parts[ b ].length() > 0 )
804
if( value.length() == 0 ) value += " ";
856
if( value.length() > 0 ) value += " ";
805
857
value += name_part_parts[ b ];
926
978
for( int a = 0; a < adr_parts.length; a++ )
927
979
if( adr_parts[ a ].length() > 0 )
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 ];
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 ];
940
1003
Set< String > types = extractTypes( params, Arrays.asList(
965
1028
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 );
968
1041
public void finaliseVcard()
969
1042
throws ParseException, ContactNotIdentifiableException
971
1044
// missing version (and data is present)
972
if( _version == null && _buffers != null )
1045
if( _version == null && _content_lines != null )
973
1046
throw new ParseException( R.string.error_vcf_malformed );
975
1048
// finalise the parent class
1000
1075
HashSet< String > ret = new HashSet< String >();
1002
1077
Pattern p = Pattern.compile(
1003
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1078
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1079
Pattern.CASE_INSENSITIVE );
1004
1080
for( int i = 0; i < params.length; i++ ) {
1005
1081
Matcher m = p.matcher( params[ i ] );
1006
1082
if( m.matches() )
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
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
1019
1096
* @return a set of present type values
1021
1098
private Set< String > extractTypes( String[] params,
1027
1104
String type_params[] = checkParams( params, "TYPE" );
1028
1105
for( int a = 0; a < type_params.length; a++ )
1030
// check for a comma-separated list of types (why? this isn't in
1107
// check for a comma-separated list of types (why? I don't think
1108
// this is in the specs!)
1032
1109
String[] parts = type_params[ a ].split( "," );
1033
for( int i = 0; i < parts.length; i++ )
1034
if( valid_types.contains( parts[ i ] ) )
1035
types.add( parts[ i ] );
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 );
1038
1117
// get 2.1-style type param
1039
1118
if( _version.equals( "2.1" ) ) {
1040
for( int i = 1; i < params.length; i++ )
1041
if( valid_types.contains( params[ i ] ) )
1042
types.add( params[ i ] );
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 );
1067
1148
else if( ch == '=' && i == in.limit() - 1 )
1069
1150
// we found a '=' at the end of a line signifying a multi-
1070
// line string, so we don't add it.
1151
// line string, so we don't add it
1071
1152
another = true;