301
383
@SuppressWarnings("serial")
302
384
protected class SkipContactException extends Exception { }
304
public void parseLine( String line )
305
throws ParseException, SkipContactException,
308
// ignore empty lines
309
if( line.trim() == "" ) return;
311
// split line into name and value parts (this may turn out to be
312
// unwanted if the line is a subsequent line in a multi-line
313
// value, but we have to do this now to check for and handle VCF
315
String[] props = line.split( ":", 2 );
316
for( int i = 0; i < props.length; i++ )
317
props[ i ] = props[ i ].trim();
319
// if we haven't yet got a version, we won't be paring anything!
386
private String extractCollonPartFromLine( ByteBuffer buffer,
387
String line, boolean former )
391
// get a US-ASCII version of the line for processing, unless we were
395
line = new String( buffer.array(), buffer.position(),
396
buffer.limit() - buffer.position(), "US-ASCII" );
398
catch( UnsupportedEncodingException e ) {
399
// we know US-ASCII is supported, so appease the compiler...
404
// split line into name and value parts and check to make sure we
405
// only got 2 parts and that the first part is not zero in length
406
String[] parts = line.split( ":", 2 );
407
if( parts.length == 2 && parts[ 0 ].length() > 0 )
408
ret = parts[ former? 0 : 1 ];
413
private String extractNameAndParamsFromLine( ByteBuffer buffer,
416
return extractCollonPartFromLine( buffer, line, true );
419
private String extractValueFromLine( ByteBuffer buffer, String line )
421
return extractCollonPartFromLine( buffer, line, false );
424
public void parseLine( ByteBuffer buffer, String line,
425
boolean next_line_looks_folded )
426
throws ParseException, SkipContactException,
429
// do we have a version yet?
320
430
if( _version == null )
322
// is this a version?
323
if( props.length == 2 && props[ 0 ].equals( "VERSION" ) )
432
// tentatively get name and params from line
433
String name_and_params =
434
extractNameAndParamsFromLine( buffer, line );
436
// is it a version line?
437
if( name_and_params != null &&
438
name_and_params.equals( "VERSION" ) )
325
// yes, check/store it
326
if( !props[ 1 ].equals( "2.1" ) &&
327
!props[ 1 ].equals( "3.0" ) )
441
String value = extractValueFromLine( buffer, line );
442
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
328
443
throw new ParseException( R.string.error_vcf_version );
329
_version = props[ 1 ];
331
// parse any other lines we've accumulated so far
333
for( int i = 0; i < _lines.size(); i++ )
334
parseLine( _lines.get( i ) );
446
// parse any buffers we've been accumulating while we waited
448
if( _buffers != null )
449
for( int i = 0; i < _buffers.size(); i++ )
450
parseLine( _buffers.get( i ), null,
451
i + 1 < _buffers.size() &&
452
_buffers.get( i + 1 ).hasRemaining() &&
453
_buffers.get( i + 1 ).get(
454
_buffers.get( i + 1 ).position() ) == ' ' );
339
// no, so stash this line till we have a version
341
_lines = new Vector< String >();
459
// no, so stash this line till we get a version
460
if( _buffers == null )
461
_buffers = new Vector< ByteBuffer >();
462
_buffers.add( buffer );
347
if( _parser_in_multiline )
467
// name and params and the position in the buffer where the
468
// "value" part of the line start
469
String name_and_params;
472
if( _parser_in_encoded_multiline ||
473
_parser_in_folded_multiline )
349
475
// if we're currently in a multi-line value, use the stored
350
476
// property name and parameters
351
props = new String[ 2 ];
352
props[ 0 ] = _parser_current_name_and_params;
353
props[ 1 ] = line.trim();
477
name_and_params = _parser_current_name_and_params;
479
pos = buffer.position();
481
// for folded multi-lines, skip the single space at the
482
// start of the next line
483
if( _parser_in_folded_multiline )
486
// else, this must be an encoded multi-line, so skip any
487
// whitespace we find at the start of the next line
489
while( pos < buffer.limit() && (
490
buffer.get( pos ) == ' ' ||
491
buffer.get( pos ) == '\t' ) )
357
// for normal lines, check the property name/value bits
358
if( props.length < 2 || props[ 0 ].length() == 0 )
498
// get name and params from line, and since we're not
499
// parsing a subsequent line in a multi-line, this should
500
// not fail, or it's an error
502
extractNameAndParamsFromLine( buffer, line );
503
if( name_and_params == null )
359
504
throw new ParseException(
360
505
R.string.error_vcf_malformed );
362
// ignore empty properties
363
if( props[ 1 ].length() < 1 )
507
// calculate how many chars to skip from beginning of line
508
// so we skip the property "name:" part
509
pos = buffer.position() + name_and_params.length() + 1;
366
511
// reset the saved multi-line state
367
_parser_current_name_and_params = props[ 0 ];
512
_parser_current_name_and_params = name_and_params;
368
513
_parser_buffered_value_so_far = "";
516
// get value from buffer, as raw bytes
518
value = ByteBuffer.wrap( buffer.array(), pos,
519
buffer.limit() - pos );
371
521
// get parameter parts
372
String[] params = props[ 0 ].split( ";" );
373
for( int i = 0; i < params.length; i++ )
374
params[ i ] = params[ i ].trim();
522
String[] name_param_parts = name_and_params.split( ";", -1 );
523
for( int i = 0; i < name_param_parts.length; i++ )
524
name_param_parts[ i ] = name_param_parts[ i ].trim();
376
// parse charset and encoding parameters
377
String charset, encoding;
378
if( ( charset = checkParam( params, "CHARSET" ) ) != null &&
379
!charset.equals( "UTF-8" ) && !charset.equals( "UTF-16" ) )
381
throw new ParseException( R.string.error_vcf_charset );
383
if( ( encoding = checkParam( params, "ENCODING" ) ) != null &&
384
!encoding.equals( "QUOTED-PRINTABLE" ) &&
385
!encoding.equals( "8BIT" ) )
526
// parse encoding parameter
527
String encoding = checkParam( name_param_parts, "ENCODING" );
528
if( encoding != null ) encoding = encoding.toUpperCase();
529
if( encoding != null && !encoding.equals( "8BIT" ) &&
530
!encoding.equals( "QUOTED-PRINTABLE" ) )
386
531
//&& !encoding.equals( "BASE64" ) )
388
533
throw new ParseException( R.string.error_vcf_encoding );
536
// parse charset parameter
537
String charset = checkParam( name_param_parts, "CHARSET" );
538
if( charset != null ) charset = charset.toUpperCase();
539
if( charset != null && !charset.equals( "US-ASCII" ) &&
540
!charset.equals( "ASCII" ) &&
541
!charset.equals( "UTF-8" ) )
543
throw new ParseException( R.string.error_vcf_charset );
391
546
// do unencoding (or default to a fake unencoding result with
392
547
// the raw string)
393
UnencodeResult result;
548
UnencodeResult unencoding_result = null;
394
549
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
395
result = unencodeQuotedPrintable( props[ 1 ], charset );
550
unencoding_result = unencodeQuotedPrintable( value );
396
551
// else if( encoding != null && encoding.equals( "BASE64" ) )
397
// result = unencodeBase64( props[ 1 ], charset );
399
result = new UnencodeResult( false, props[ 1 ].getBytes(),
400
props[ 1 ].getBytes().length );
552
// unencoding_result = unencodeBase64( props[ 1 ], charset );
553
if( unencoding_result != null ) {
554
value = unencoding_result.getBuffer();
555
_parser_in_encoded_multiline =
556
unencoding_result.isAnotherLineRequired();
559
// convert 8-bit ASCII charset to US-ASCII
560
if( charset == null || charset.equals( "ASCII" ) ) {
561
value = transcodeAsciiToUtf8( value );
402
565
// process charset
404
props[ 1 ] = new String( result.getBytes(), 0,
405
result.getNumBytes(),
406
charset == null? "UTF-8" : charset );
568
string_value = new String( value.array(), value.position(),
569
value.limit() - value.position(), charset );
407
570
} catch( UnsupportedEncodingException e ) {
408
571
throw new ParseException( R.string.error_vcf_charset );
574
// now we know whether we're in an encoding multi-line,
575
// determine if we're in a v3 folded multi-line or not
576
_parser_in_folded_multiline = !_parser_in_encoded_multiline &&
577
_version.equals( "3.0" ) && next_line_looks_folded;
411
579
// handle multi-line requests
412
_parser_in_multiline = result.isAnotherLineRequired();
413
if( _parser_in_multiline ) {
414
_parser_buffered_value_so_far += props[ 1 ];
580
if( _parser_in_encoded_multiline ||
581
_parser_in_folded_multiline )
583
_parser_buffered_value_so_far += string_value;
418
587
// add on buffered multi-line content
419
String value = _parser_buffered_value_so_far + props[ 1 ];
588
String complete_value =
589
_parser_buffered_value_so_far + string_value;
591
// ignore empty values
592
if( complete_value.length() < 1 ) return;
421
594
// parse some properties
422
if( params[ 0 ].equals( "N" ) )
423
parseN( params, value );
424
else if( params[ 0 ].equals( "FN" ) )
425
parseFN( params, value );
426
else if( params[ 0 ].equals( "ORG" ) )
427
parseORG( params, value );
428
else if( params[ 0 ].equals( "TEL" ) )
429
parseTEL( params, value );
430
else if( params[ 0 ].equals( "EMAIL" ) )
431
parseEMAIL( params, value );
595
if( name_param_parts[ 0 ].equals( "N" ) )
596
parseN( name_param_parts, complete_value );
597
else if( name_param_parts[ 0 ].equals( "FN" ) )
598
parseFN( name_param_parts, complete_value );
599
else if( name_param_parts[ 0 ].equals( "ORG" ) )
600
parseORG( name_param_parts, complete_value );
601
else if( name_param_parts[ 0 ].equals( "TEL" ) )
602
parseTEL( name_param_parts, complete_value );
603
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
604
parseEMAIL( name_param_parts, complete_value );
435
608
private void parseN( String[] params, String value )
436
throws ParseException, SkipContactException,
609
throws ParseException, SkipContactException,
439
612
// already got a better name?
440
613
if( _name_level >= NAMELEVEL_N ) return;
442
615
// get name parts
443
String[] nameparts = value.split( ";" );
444
for( int i = 0; i < nameparts.length; i++ )
445
nameparts[ i ] = nameparts[ i ].trim();
616
String[] name_parts = value.split( ";" );
617
for( int i = 0; i < name_parts.length; i++ )
618
name_parts[ i ] = name_parts[ i ].trim();
449
if( nameparts.length > 1 && nameparts[ 1 ].length() > 0 )
450
value += nameparts[ 1 ];
451
if( nameparts[ 0 ].length() > 0 )
452
value += ( value.length() == 0? "" : " " ) + nameparts[ 0 ];
622
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
623
value += name_parts[ 1 ];
624
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
625
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
455
628
setName( value );
497
670
private void parseTEL( String[] params, String value )
498
throws ParseException
671
throws ParseException
500
673
if( value.length() == 0 ) return;
502
675
Set< String > types = extractTypes( params, Arrays.asList(
503
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
504
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
676
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
677
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
506
679
// here's the logic...
507
680
boolean preferred = types.contains( "PREF" );
681
int type = PhonesColumns.TYPE_MOBILE;
508
682
if( types.contains( "VOICE" ) )
509
683
if( types.contains( "WORK" ) )
510
addPhone( value, PhonesColumns.TYPE_WORK, preferred );
684
type = PhonesColumns.TYPE_WORK;
512
addPhone( value, PhonesColumns.TYPE_HOME, preferred );
686
type = PhonesColumns.TYPE_HOME;
513
687
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
514
addPhone( value, PhonesColumns.TYPE_MOBILE, preferred );
688
type = PhonesColumns.TYPE_MOBILE;
515
689
if( types.contains( "FAX" ) )
516
690
if( types.contains( "HOME" ) )
517
addPhone( value, PhonesColumns.TYPE_FAX_HOME, preferred );
691
type = PhonesColumns.TYPE_FAX_HOME;
519
addPhone( value, PhonesColumns.TYPE_FAX_WORK, preferred );
693
type = PhonesColumns.TYPE_FAX_WORK;
520
694
if( types.contains( "PAGER" ) )
521
addPhone( value, PhonesColumns.TYPE_PAGER, preferred );
695
type = PhonesColumns.TYPE_PAGER;
698
addPhone( value, type, preferred );
524
701
public void parseEMAIL( String[] params, String value )
525
throws ParseException
702
throws ParseException
527
704
if( value.length() == 0 ) return;
529
706
Set< String > types = extractTypes( params, Arrays.asList(
530
"PREF", "WORK", "HOME", "INTERNET" ) );
707
"PREF", "WORK", "HOME", "INTERNET" ) );
532
709
// here's the logic...
533
710
boolean preferred = types.contains( "PREF" );
559
736
private String checkParam( String[] params, String name )
561
Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
738
Pattern p = Pattern.compile(
739
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
562
740
for( int i = 0; i < params.length; i++ ) {
563
741
Matcher m = p.matcher( params[ i ] );
564
742
if( m.matches() )
570
748
private Set< String > extractTypes( String[] params,
571
List< String > validTypes )
749
List< String > valid_types )
573
751
HashSet< String > types = new HashSet< String >();
575
753
// get 3.0-style TYPE= param
577
if( ( typeParam = checkParam( params, "TYPE" ) ) != null ) {
578
String[] bits = typeParam.split( "," );
579
for( int i = 0; i < bits.length; i++ )
580
if( validTypes.contains( bits[ i ] ) )
581
types.add( bits[ i ] );
755
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
756
String[] parts = type_param.split( "," );
757
for( int i = 0; i < parts.length; i++ )
758
if( valid_types.contains( parts[ i ] ) )
759
types.add( parts[ i ] );
584
762
// get 2.1-style type param
585
763
if( _version.equals( "2.1" ) ) {
586
764
for( int i = 1; i < params.length; i++ )
587
if( validTypes.contains( params[ i ] ) )
765
if( valid_types.contains( params[ i ] ) )
588
766
types.add( params[ i ] );
594
private UnencodeResult unencodeQuotedPrintable( String str, String charset )
772
private UnencodeResult unencodeQuotedPrintable( ByteBuffer in )
596
774
boolean another = false;
598
// default encoding scheme
599
if( charset == null ) charset = "UTF-8";
601
// unencode quoted-pritable encoding, as per RFC1521 section 5.1
602
byte[] bytes = new byte[ str.length() ];
776
// unencode quoted-printable encoding, as per RFC1521 section 5.1
777
byte[] out = new byte[ in.limit() - in.position() ];
604
for( int i = 0; i < str.length(); i++ )
779
for( int i = in.position(); i < in.limit(); i++ )
606
781
// get next char and process...
607
char ch = str.charAt( i );
608
if( ch == '=' && i < str.length() - 2 )
782
byte ch = in.array()[ i ];
783
if( ch == '=' && i < in.limit() - 2 )
610
785
// we found a =XX format byte, add it
612
Character.digit( str.charAt( i + 1 ), 16 ) * 16 +
613
Character.digit( str.charAt( i + 2 ), 16 ) );
787
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
788
Character.digit( in.array()[ i + 2 ], 16 ) );
616
else if( ch == '=' && i == str.length() - 1 )
791
else if( ch == '=' && i == in.limit() - 1 )
618
793
// we found a '=' at the end of a line signifying a multi-
619
794
// line string, so we don't add it.