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;
182
176
importVCardFileContent( content, file.getName() );
184
catch( OutOfMemoryError e ) {
185
showError( R.string.error_outofmemory );
187
178
catch( FileNotFoundException e ) {
188
179
showError( getText( R.string.error_filenotfound ) +
189
180
file.getName() );
202
193
ContentLineIterator cli = new ContentLineIterator( content );
203
194
while( cli.hasNext() )
205
ContentLine content_line = cli.next();
196
ByteBuffer buffer = cli.next();
207
// get a US-ASCII version of the string, for processing
208
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...
210
209
if( vcard == null ) {
211
210
// look for vcard beginning
212
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
211
if( line.matches( "^BEGIN:VCARD" ) ) {
213
212
setProgress( _progress++ );
214
213
vcard = new Vcard();
215
214
vcard_start_line = cli.getLineNumber();
261
260
// try giving the line to the vcard
263
vcard.parseLine( content_line );
262
vcard.parseLine( buffer, line,
263
cli.doesNextLineLookFolded() );
265
265
catch( Vcard.ParseException e ) {
273
273
finish( ACTION_ABORT );
276
// Although we're continuing, we still need to abort
277
// 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
278
278
// get to another BEGIN:VCARD line.
281
281
catch( Vcard.SkipImportException e ) {
283
// Abort this vCard. Further lines will be ignored until
283
// abort this vCard. Further lines will be ignored until
284
284
// 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 >
292
class ContentLineIterator implements Iterator< ByteBuffer >
336
294
protected byte[] _content = null;
337
295
protected int _pos = 0;
362
320
_pos > initial_pos )? _pos - 1 : _pos;
365
return new ContentLine(
366
ByteBuffer.wrap( _content, initial_pos,
368
doesNextLineLookFolded() );
323
return ByteBuffer.wrap( _content, initial_pos,
371
327
// we didn't find one, but were there bytes left?
376
return new ContentLine(
377
ByteBuffer.wrap( _content, initial_pos,
379
doesNextLineLookFolded() );
332
return ByteBuffer.wrap( _content, initial_pos,
394
348
* onto the end of this one?
397
private boolean doesNextLineLookFolded()
351
public boolean doesNextLineLookFolded()
399
353
return _pos > 0 && _pos < _content.length &&
400
_content[ _pos - 1 ] == '\n' &&
401
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
354
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
404
357
public int getLineNumber()
416
369
private final static int MULTILINE_NONE = 0;
417
370
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
418
371
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
419
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
372
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
421
374
private String _version = null;
422
private Vector< ContentLine > _content_lines = null;
375
private Vector< ByteBuffer > _buffers = null;
423
376
private int _name_level = NAMELEVEL_NONE;
424
377
private int _parser_multiline_state = MULTILINE_NONE;
425
378
private String _parser_current_name_and_params = null;
468
421
@SuppressWarnings("serial")
469
422
protected class SkipImportException extends Exception { }
471
private String extractCollonPartFromLine( ContentLine content_line,
424
private String extractCollonPartFromLine( ByteBuffer buffer,
425
String line, boolean former )
474
427
String ret = null;
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...
476
442
// split line into name and value parts and check to make sure we
477
443
// only got 2 parts and that the first part is not zero in length
478
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
444
String[] parts = line.split( ":", 2 );
479
445
if( parts.length == 2 && parts[ 0 ].length() > 0 )
480
446
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 )
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 )
496
464
throws ParseException, SkipImportException,
497
465
AbortImportException
502
470
// tentatively get name and params from line
503
471
String name_and_params =
504
extractNameAndParamsFromLine( content_line );
472
extractNameAndParamsFromLine( buffer, line );
506
474
// is it a version line?
507
475
if( name_and_params != null &&
508
name_and_params.equalsIgnoreCase( "VERSION" ) )
476
name_and_params.equals( "VERSION" ) )
511
String value = extractValueFromLine( content_line ).trim();
479
String value = extractValueFromLine( buffer, line );
512
480
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
481
throw new ParseException( R.string.error_vcf_version );
514
482
_version = value;
516
484
// 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;
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() ) == ' ' );
525
497
// 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 );
498
if( _buffers == null )
499
_buffers = new Vector< ByteBuffer >();
500
_buffers.add( buffer );
533
505
// name and params and the position in the buffer where the
534
// "value" part of the line starts
506
// "value" part of the line start
535
507
String name_and_params;
544
516
// skip some initial line characters, depending on the type
545
517
// of multi-line we're handling
546
pos = content_line.getBuffer().position();
518
pos = buffer.position();
547
519
switch( _parser_multiline_state )
549
521
case MULTILINE_FOLDED:
552
524
case MULTILINE_ENCODED:
553
while( pos < content_line.getBuffer().limit() && (
554
content_line.getBuffer().get( pos ) == ' ' ||
555
content_line.getBuffer().get( pos ) == '\t' ) )
525
while( pos < buffer.limit() && (
526
buffer.get( pos ) == ' ' ||
527
buffer.get( pos ) == '\t' ) )
571
if( content_line.getUsAsciiLine().trim().length() == 0 )
574
542
// get name and params from line, and since we're not
575
543
// parsing a subsequent line in a multi-line, this should
576
544
// not fail, or it's an error
577
545
name_and_params =
578
extractNameAndParamsFromLine( content_line );
546
extractNameAndParamsFromLine( buffer, line );
579
547
if( name_and_params == null )
580
548
throw new ParseException(
581
549
R.string.error_vcf_malformed );
583
551
// calculate how many chars to skip from beginning of line
584
552
// so we skip the property "name:" part
585
pos = content_line.getBuffer().position() +
586
name_and_params.length() + 1;
553
pos = buffer.position() + name_and_params.length() + 1;
588
555
// reset the saved multi-line state
589
556
_parser_current_name_and_params = name_and_params;
593
560
// get value from buffer, as raw bytes
594
561
ByteBuffer value;
595
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
596
content_line.getBuffer().limit() - pos );
562
value = ByteBuffer.wrap( buffer.array(), pos,
563
buffer.limit() - pos );
598
565
// get parameter parts
599
566
String[] name_param_parts = name_and_params.split( ";", -1 );
606
573
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
608
575
boolean is_interesting_field =
609
interesting_fields.contains(
610
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
576
interesting_fields.contains( name_param_parts[ 0 ] );
612
578
// parse encoding parameter
613
579
String encoding = checkParam( name_param_parts, "ENCODING" );
614
580
if( encoding != null )
615
581
encoding = encoding.toUpperCase( Locale.US );
616
582
if( is_interesting_field && encoding != null &&
617
!encoding.equalsIgnoreCase( "8BIT" ) &&
618
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
619
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
583
!encoding.equals( "8BIT" ) &&
584
!encoding.equals( "QUOTED-PRINTABLE" ) )
585
//&& !encoding.equals( "BASE64" ) )
621
587
throw new ParseException( R.string.error_vcf_encoding );
626
592
if( charset != null )
627
593
charset = charset.toUpperCase( Locale.US );
628
594
if( charset != null &&
629
!charset.equalsIgnoreCase( "US-ASCII" ) &&
630
!charset.equalsIgnoreCase( "ASCII" ) &&
631
!charset.equalsIgnoreCase( "UTF-8" ) )
595
!charset.equals( "US-ASCII" ) &&
596
!charset.equals( "ASCII" ) &&
597
!charset.equals( "UTF-8" ) )
633
599
throw new ParseException( R.string.error_vcf_charset );
636
602
// do unencoding (or default to a fake unencoding result with
637
603
// the raw string)
638
604
UnencodeResult unencoding_result = null;
639
if( encoding != null &&
640
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
605
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
642
606
unencoding_result = unencodeQuotedPrintable( value );
644
// else if( encoding != null &&
645
// encoding.equalsIgnoreCase( "BASE64" ) )
607
// else if( encoding != null && encoding.equals( "BASE64" ) )
647
608
// unencoding_result = unencodeBase64( props[ 1 ], charset );
649
609
if( unencoding_result != null ) {
650
610
value = unencoding_result.getBuffer();
651
611
if( unencoding_result.isAnotherLineRequired() )
656
616
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
657
617
if( ( charset == null && _version.equals( "2.1" ) ) ||
658
618
( charset != null && (
659
charset.equalsIgnoreCase( "ASCII" ) ||
660
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
619
charset.equals( "ASCII" ) ||
620
charset.equals( "US-ASCII" ) ) ) )
662
622
value = transcodeAsciiToUtf8( value );
674
634
// for some entries that have semicolon-separated value parts,
675
635
// check to see if the value ends in an escape character, which
676
636
// 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" ) ) &&
637
if( ( name_param_parts[ 0 ].equals( "N" ) ||
638
name_param_parts[ 0 ].equals( "ORG" ) ||
639
name_param_parts[ 0 ].equals( "ADR" ) ) &&
680
640
doesStringEndInAnEscapeChar( string_value ) )
682
642
_parser_multiline_state = MULTILINE_ESCAPED;
684
644
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
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
649
if( _parser_multiline_state == MULTILINE_NONE &&
690
content_line.doesNextLineLookFolded() )
650
_version.equals( "3.0" ) && next_line_looks_folded )
692
652
_parser_multiline_state = MULTILINE_FOLDED;
705
665
if( complete_value.length() < 1 ) return;
707
667
// parse some properties
708
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
668
if( name_param_parts[ 0 ].equals( "N" ) )
709
669
parseN( name_param_parts, complete_value );
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
670
else if( name_param_parts[ 0 ].equals( "FN" ) )
711
671
parseFN( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
672
else if( name_param_parts[ 0 ].equals( "ORG" ) )
713
673
parseORG( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
674
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
715
675
parseTITLE( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
676
else if( name_param_parts[ 0 ].equals( "TEL" ) )
717
677
parseTEL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
678
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
719
679
parseEMAIL( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
680
else if( name_param_parts[ 0 ].equals( "ADR" ) )
721
681
parseADR( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
682
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
723
683
parseLABEL( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
684
else if( name_param_parts[ 0 ].equals( "NOTE" ) )
725
685
parseNOTE( name_param_parts, complete_value );
726
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
727
parseBDAY( name_param_parts, complete_value );
755
713
String str = parts.get( a );
757
// Look for parts that end in an escape character, but ignore
758
// the final part. We've already detected escape chars at the
715
// look for parts that end in an escape character, but ignore
716
// the final part. We've already detected escape chars at the
759
717
// end of the final part in parseLine() and handled multi-lines
761
719
if( a < parts.size() - 1 &&
851
809
for( int b = 0; b < name_part_parts.length; b++ )
852
810
if( name_part_parts[ b ].length() > 0 )
854
if( value.length() > 0 ) value += " ";
812
if( value.length() == 0 ) value += " ";
855
813
value += name_part_parts[ b ];
976
934
for( int a = 0; a < adr_parts.length; a++ )
977
935
if( adr_parts[ a ].length() > 0 )
979
// version 3.0 vCards allow further splitting by comma
980
if( _version.equals( "3.0" ) )
982
// split this part in to it's comma-separated bits and
983
// add them on individual lines
984
String[] adr_part_parts =
985
splitValueByCharacter( adr_parts[ a ], ',' );
986
for( int b = 0; b < adr_part_parts.length; b++ )
987
if( adr_part_parts[ b ].length() > 0 )
989
if( value.length() > 0 ) value += "\n";
990
value += adr_part_parts[ b ];
995
// add this part on an individual line
996
if( value.length() > 0 ) value += "\n";
997
value += adr_parts[ a ];
937
// split this part in to it's comma-separated bits
938
String[] adr_part_parts =
939
splitValueByCharacter( adr_parts[ a ], ',' );
940
for( int b = 0; b < adr_part_parts.length; b++ )
941
if( adr_part_parts[ b ].length() > 0 )
943
if( value.length() > 0 ) value += "\n";
944
value += adr_part_parts[ b ];
1001
948
Set< String > types = extractTypes( params, Arrays.asList(
1031
978
addNote( unescapeValue( value ) );
1034
private void parseBDAY( String[] params, String value )
1036
setBirthday( value );
1039
981
public void finaliseVcard()
1040
982
throws ParseException, ContactNotIdentifiableException
1042
984
// missing version (and data is present)
1043
if( _version == null && _content_lines != null )
985
if( _version == null && _buffers != null )
1044
986
throw new ParseException( R.string.error_vcf_malformed );
1046
988
// finalise the parent class
1065
* Amongst the params, find the values of any with the specified name.
1006
* Amongst the params, find the values of any with the specified name
1067
1007
* @param params
1069
1009
* @return an array of values, or null
1073
1013
HashSet< String > ret = new HashSet< String >();
1075
1015
Pattern p = Pattern.compile(
1076
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1077
Pattern.CASE_INSENSITIVE );
1016
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1078
1017
for( int i = 0; i < params.length; i++ ) {
1079
1018
Matcher m = p.matcher( params[ i ] );
1080
1019
if( m.matches() )
1088
* Amongst the params, return any type values present. For v2.1 vCards,
1089
* those types are just parameters. For v3.0, they are prefixed with
1090
* "TYPE=". There may also be multiple type parameters.
1092
* @param params an array of params to look for types in
1093
* @param valid_types an list of upper-case type values to look for
1027
* Amongst the params, return any type values present. For v2.1 vCards,
1028
* those types are just parameters. For v3.0, they are prefixed with
1029
* "TYPE=". There may also be multiple type parameters.
1031
* @param a list of type values to look for
1094
1032
* @return a set of present type values
1096
1034
private Set< String > extractTypes( String[] params,
1102
1040
String type_params[] = checkParams( params, "TYPE" );
1103
1041
for( int a = 0; a < type_params.length; a++ )
1105
// check for a comma-separated list of types (why? I don't think
1106
// this is in the specs!)
1043
// check for a comma-separated list of types (why? this isn't in
1107
1045
String[] parts = type_params[ a ].split( "," );
1108
for( int i = 0; i < parts.length; i++ ) {
1109
String ucpart = parts[ i ].toUpperCase( Locale.US );
1110
if( valid_types.contains( ucpart ) )
1111
types.add( ucpart );
1046
for( int i = 0; i < parts.length; i++ )
1047
if( valid_types.contains( parts[ i ] ) )
1048
types.add( parts[ i ] );
1115
1051
// get 2.1-style type param
1116
1052
if( _version.equals( "2.1" ) ) {
1117
for( int i = 1; i < params.length; i++ ) {
1118
String ucparam = params[ i ].toUpperCase( Locale.US );
1119
if( valid_types.contains( ucparam ) )
1120
types.add( ucparam );
1053
for( int i = 1; i < params.length; i++ )
1054
if( valid_types.contains( params[ i ] ) )
1055
types.add( params[ i ] );
1146
1080
else if( ch == '=' && i == in.limit() - 1 )
1148
1082
// we found a '=' at the end of a line signifying a multi-
1149
// line string, so we don't add it
1083
// line string, so we don't add it.
1150
1084
another = true;