262
391
@SuppressWarnings("serial")
263
392
protected class SkipContactException extends Exception { }
265
public void parseLine( String line )
266
throws ParseException, SkipContactException,
269
// get property halves
270
String[] props = line.split( ":" );
271
for( int i = 0; i < props.length; i++ )
272
props[ i ] = props[ i ].trim();
273
if( props.length < 2 ||
274
props[ 0 ].length() < 1 || props[ 1 ].length() < 1 )
275
throw new ParseException( R.string.error_vcf_malformed );
394
private String extractCollonPartFromLine( ByteBuffer buffer,
395
String line, boolean former )
399
// get a US-ASCII version of the line for processing, unless we were
403
line = new String( buffer.array(), buffer.position(),
404
buffer.limit() - buffer.position(), "US-ASCII" );
406
catch( UnsupportedEncodingException e ) {
407
// we know US-ASCII is supported, so appease the compiler...
412
// split line into name and value parts and check to make sure we
413
// only got 2 parts and that the first part is not zero in length
414
String[] parts = line.split( ":", 2 );
415
if( parts.length == 2 && parts[ 0 ].length() > 0 )
416
ret = parts[ former? 0 : 1 ];
421
private String extractNameAndParamsFromLine( ByteBuffer buffer,
424
return extractCollonPartFromLine( buffer, line, true );
427
private String extractValueFromLine( ByteBuffer buffer, String line )
429
return extractCollonPartFromLine( buffer, line, false );
432
public void parseLine( ByteBuffer buffer, String line,
433
boolean next_line_looks_folded )
434
throws ParseException, SkipContactException,
437
// do we have a version yet?
277
438
if( _version == null )
279
if( props[ 0 ].equals( "VERSION" ) )
440
// tentatively get name and params from line
441
String name_and_params =
442
extractNameAndParamsFromLine( buffer, line );
444
// is it a version line?
445
if( name_and_params != null &&
446
name_and_params.equals( "VERSION" ) )
282
if( !props[ 1 ].equals( "2.1" ) &&
283
!props[ 1 ].equals( "3.0" ) )
449
String value = extractValueFromLine( buffer, line );
450
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
284
451
throw new ParseException( R.string.error_vcf_version );
285
_version = props[ 1 ];
287
// parse any other lines we've accumulated so far
289
for( int i = 0; i < _lines.size(); i++ )
290
parseLine( _lines.get( i ) );
454
// parse any buffers we've been accumulating while we waited
456
if( _buffers != null )
457
for( int i = 0; i < _buffers.size(); i++ )
458
parseLine( _buffers.get( i ), null,
459
i + 1 < _buffers.size() &&
460
_buffers.get( i + 1 ).hasRemaining() &&
461
_buffers.get( i + 1 ).get(
462
_buffers.get( i + 1 ).position() ) == ' ' );
295
// stash this line till we have a version
297
_lines = new Vector< String >();
467
// no, so stash this line till we get a version
468
if( _buffers == null )
469
_buffers = new Vector< ByteBuffer >();
470
_buffers.add( buffer );
475
// name and params and the position in the buffer where the
476
// "value" part of the line start
477
String name_and_params;
480
if( _parser_multiline_state != MULTILINE_NONE )
482
// if we're currently in a multi-line value, use the stored
483
// property name and parameters
484
name_and_params = _parser_current_name_and_params;
486
// skip some initial line characters, depending on the type
487
// of multi-line we're handling
488
pos = buffer.position();
489
switch( _parser_multiline_state )
491
case MULTILINE_FOLDED:
494
case MULTILINE_ENCODED:
495
while( pos < buffer.limit() && (
496
buffer.get( pos ) == ' ' ||
497
buffer.get( pos ) == '\t' ) )
506
// take us out of multi-line so that we can re-detect that
507
// this line is a multi-line or not
508
_parser_multiline_state = MULTILINE_NONE;
512
// get name and params from line, and since we're not
513
// parsing a subsequent line in a multi-line, this should
514
// not fail, or it's an error
516
extractNameAndParamsFromLine( buffer, line );
517
if( name_and_params == null )
518
throw new ParseException(
519
R.string.error_vcf_malformed );
521
// calculate how many chars to skip from beginning of line
522
// so we skip the property "name:" part
523
pos = buffer.position() + name_and_params.length() + 1;
525
// reset the saved multi-line state
526
_parser_current_name_and_params = name_and_params;
527
_parser_buffered_value_so_far = "";
530
// get value from buffer, as raw bytes
532
value = ByteBuffer.wrap( buffer.array(), pos,
533
buffer.limit() - pos );
303
535
// get parameter parts
304
String[] params = props[ 0 ].split( ";" );
305
for( int i = 0; i < params.length; i++ )
306
params[ i ] = params[ i ].trim();
536
String[] name_param_parts = name_and_params.split( ";", -1 );
537
for( int i = 0; i < name_param_parts.length; i++ )
538
name_param_parts[ i ] = name_param_parts[ i ].trim();
540
// parse encoding parameter
541
String encoding = checkParam( name_param_parts, "ENCODING" );
542
if( encoding != null ) encoding = encoding.toUpperCase();
543
if( encoding != null && !encoding.equals( "8BIT" ) &&
544
!encoding.equals( "QUOTED-PRINTABLE" ) )
545
//&& !encoding.equals( "BASE64" ) )
547
throw new ParseException( R.string.error_vcf_encoding );
550
// parse charset parameter
551
String charset = checkParam( name_param_parts, "CHARSET" );
552
if( charset != null ) charset = charset.toUpperCase();
553
if( charset != null && !charset.equals( "US-ASCII" ) &&
554
!charset.equals( "ASCII" ) &&
555
!charset.equals( "UTF-8" ) )
557
throw new ParseException( R.string.error_vcf_charset );
560
// do unencoding (or default to a fake unencoding result with
562
UnencodeResult unencoding_result = null;
563
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
564
unencoding_result = unencodeQuotedPrintable( value );
565
// else if( encoding != null && encoding.equals( "BASE64" ) )
566
// unencoding_result = unencodeBase64( props[ 1 ], charset );
567
if( unencoding_result != null ) {
568
value = unencoding_result.getBuffer();
569
if( unencoding_result.isAnotherLineRequired() )
570
_parser_multiline_state = MULTILINE_ENCODED;
573
// convert 8-bit ASCII charset to US-ASCII
574
if( charset == null || charset.equals( "ASCII" ) ) {
575
value = transcodeAsciiToUtf8( value );
582
string_value = new String( value.array(), value.position(),
583
value.limit() - value.position(), charset );
584
} catch( UnsupportedEncodingException e ) {
585
throw new ParseException( R.string.error_vcf_charset );
588
// for some entries that have semicolon-separated value parts,
589
// check to see if the value ends in an escape character, which
590
// indicates that we have a multi-line value
591
if( ( name_param_parts[ 0 ].equals( "N" ) ||
592
name_param_parts[ 0 ].equals( "ORG" ) ||
593
name_param_parts[ 0 ].equals( "ADR" ) ) &&
594
doesStringEndInAnEscapeChar( string_value ) )
596
_parser_multiline_state = MULTILINE_ESCAPED;
597
string_value = string_value.substring( 0,
598
string_value.length() - 1 );
601
// now we know whether we're in an encoding multi-line,
602
// determine if we're in a v3 folded multi-line or not
603
if( _parser_multiline_state == MULTILINE_NONE &&
604
_version.equals( "3.0" ) && next_line_looks_folded )
606
_parser_multiline_state = MULTILINE_FOLDED;
609
// handle multi-lines by buffering them and parsing them when we
610
// are processing the last line in a multi-line sequence
611
if( _parser_multiline_state != MULTILINE_NONE ) {
612
_parser_buffered_value_so_far += string_value;
615
String complete_value =
616
( _parser_buffered_value_so_far + string_value ).trim();
618
// ignore empty values
619
if( complete_value.length() < 1 ) return;
308
621
// parse some properties
309
if( params[ 0 ].equals( "N" ) )
310
parseN( params, props[ 1 ] );
311
else if( params[ 0 ].equals( "FN" ) )
312
parseFN( params, props[ 1 ] );
313
else if( params[ 0 ].equals( "ORG" ) )
314
parseORG( params, props[ 1 ] );
315
else if( params[ 0 ].equals( "TEL" ) )
316
parseTEL( params, props[ 1 ] );
317
else if( params[ 0 ].equals( "EMAIL" ) )
318
parseEMAIL( params, props[ 1 ] );
622
if( name_param_parts[ 0 ].equals( "N" ) )
623
parseN( name_param_parts, complete_value );
624
else if( name_param_parts[ 0 ].equals( "FN" ) )
625
parseFN( name_param_parts, complete_value );
626
else if( name_param_parts[ 0 ].equals( "ORG" ) )
627
parseORG( name_param_parts, complete_value );
628
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
629
parseTITLE( name_param_parts, complete_value );
630
else if( name_param_parts[ 0 ].equals( "TEL" ) )
631
parseTEL( name_param_parts, complete_value );
632
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
633
parseEMAIL( name_param_parts, complete_value );
634
else if( name_param_parts[ 0 ].equals( "ADR" ) )
635
parseADR( name_param_parts, complete_value );
639
private boolean doesStringEndInAnEscapeChar( String string )
641
// count the number of backslashes at the end of the string
643
for( int a = string.length() - 1; a >= 0; a-- )
644
if( string.charAt( a ) == '\\' )
649
// if there are an even number of backslashes then the final one
651
return ( count & 1 ) == 1;
654
private String[] splitValueBySemicolon( String value )
656
// split string in to parts by semicolon
657
ArrayList< String > parts = new ArrayList< String >(
658
Arrays.asList( value.split( ";" ) ) );
661
for( int a = 0; a < parts.size(); a++ )
663
String str = parts.get( a );
665
// look for parts that end in an escape character, but ignore
666
// the final part. We've already detected escape chars at the
667
// end of the final part in parseLine() and handled multi-lines
669
if( a < parts.size() - 1 &&
670
doesStringEndInAnEscapeChar( str ) )
672
// join the next part to this part and remove the next part
673
parts.set( a, str.substring( 0, str.length() - 1 ) +
674
';' + parts.get( a + 1 ) );
675
parts.remove( a + 1 );
677
// re-visit this part
682
// trim and replace string
687
String[] ret = new String[ parts.size() ];
688
return parts.toArray( ret );
322
691
private void parseN( String[] params, String value )
323
throws ParseException, SkipContactException,
326
693
// already got a better name?
327
if( _nameLevel >= NAMELEVEL_N ) return;
694
if( _name_level >= NAMELEVEL_N ) return;
329
696
// get name parts
330
String[] nameparts = value.split( ";" );
331
for( int i = 0; i < nameparts.length; i++ )
332
nameparts[ i ] = nameparts[ i ].trim();
697
String[] name_parts = splitValueBySemicolon( value );
336
if( nameparts.length > 1 && nameparts[ 1 ].length() > 0 )
337
value += nameparts[ 1 ];
338
if( nameparts[ 0 ].length() > 0 )
339
value += ( value.length() == 0? "" : " " ) + nameparts[ 0 ];
701
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
702
value += name_parts[ 1 ];
703
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
704
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
342
setName( undoCharsetAndEncoding( params, value ) );
343
_nameLevel = NAMELEVEL_N;
345
// check now to see if we need to import this contact (to avoid
346
// parsing the rest of the vCard unnecessarily)
347
if( !isImportRequired( getName() ) )
348
throw new SkipContactException();
708
_name_level = NAMELEVEL_N;
351
711
private void parseFN( String[] params, String value )
352
throws ParseException, SkipContactException
354
713
// already got a better name?
355
if( _nameLevel >= NAMELEVEL_FN ) return;
714
if( _name_level >= NAMELEVEL_FN ) return;
358
setName( undoCharsetAndEncoding( params, value ) );
359
_nameLevel = NAMELEVEL_FN;
718
_name_level = NAMELEVEL_FN;
362
721
private void parseORG( String[] params, String value )
363
throws ParseException, SkipContactException
365
// already got a better name?
366
if( _nameLevel >= NAMELEVEL_ORG ) return;
369
String[] orgparts = value.split( ";" );
370
for( int i = 0; i < orgparts.length; i++ )
371
orgparts[ i ] = orgparts[ i ].trim();
374
if( orgparts[ 0 ].length() == 0 && orgparts.length > 1 )
375
value = orgparts[ 1 ];
377
value = orgparts[ 0 ];
380
setName( undoCharsetAndEncoding( params, value ) );
381
_nameLevel = NAMELEVEL_ORG;
724
String[] org_parts = splitValueBySemicolon( value );
725
if( org_parts == null || org_parts.length < 1 ) return;
727
// build organisation name
728
StringBuilder builder = new StringBuilder(
729
String.valueOf( org_parts[ 0 ] ) );
730
for( int a = 1; a < org_parts.length; a++ )
731
builder.append( ", " ).append( org_parts[ a ] );
732
String organisation = builder.toString();
734
// set organisation name (using a title we've previously found)
735
addOrganisation( organisation, _cached_title, true );
737
// if we've not previously found a title, store this organisation
738
// name (we'll need it when we find a title to update the
739
// organisation, by name), else if we *have* previously found a
740
// title, clear it (since we just used it)
741
if( _cached_title == null )
742
_cached_organisation = organisation;
744
_cached_title = null;
747
private void parseTITLE( String[] params, String value )
749
// if we previously had an organisation, look it up and append this
751
if( _cached_organisation != null && hasOrganisations() ) {
752
HashMap< String, ExtraDetail > datas = getOrganisations();
753
ExtraDetail detail = datas.get( _cached_organisation );
755
detail.setExtra( value );
758
// same as when handling organisation, if we've not previously found
759
// an organisation we store this title, else we clear it (since we
760
// just appended this title to it)
761
if( _cached_organisation == null )
762
_cached_title = value;
764
_cached_organisation = null;
384
767
private void parseTEL( String[] params, String value )
385
throws ParseException
387
769
if( value.length() == 0 ) return;
389
771
Set< String > types = extractTypes( params, Arrays.asList(
390
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
391
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
772
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
773
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
393
775
// here's the logic...
394
776
boolean preferred = types.contains( "PREF" );
395
if( types.contains( "VOICE" ) )
396
if( types.contains( "WORK" ) )
397
addPhone( value, PhonesColumns.TYPE_WORK, preferred );
399
addPhone( value, PhonesColumns.TYPE_HOME, preferred );
400
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
401
addPhone( value, PhonesColumns.TYPE_MOBILE, preferred );
402
778
if( types.contains( "FAX" ) )
403
779
if( types.contains( "HOME" ) )
404
addPhone( value, PhonesColumns.TYPE_FAX_HOME, preferred );
780
type = PhonesColumns.TYPE_FAX_HOME;
406
addPhone( value, PhonesColumns.TYPE_FAX_WORK, preferred );
407
if( types.contains( "PAGER" ) )
408
addPhone( value, PhonesColumns.TYPE_PAGER, preferred );
782
type = PhonesColumns.TYPE_FAX_WORK;
783
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
784
type = PhonesColumns.TYPE_MOBILE;
785
else if( types.contains( "PAGER" ) )
786
type = PhonesColumns.TYPE_PAGER;
787
else if( types.contains( "WORK" ) )
788
type = PhonesColumns.TYPE_WORK;
790
type = PhonesColumns.TYPE_HOME;
793
addNumber( value, type, preferred );
411
796
public void parseEMAIL( String[] params, String value )
413
798
if( value.length() == 0 ) return;
415
800
Set< String > types = extractTypes( params, Arrays.asList(
416
"PREF", "WORK", "HOME", "INTERNET" ) );
801
"PREF", "WORK", "HOME", "INTERNET" ) );
418
// here's the logic...
419
804
boolean preferred = types.contains( "PREF" );
420
if( types.contains( "WORK" ) )
421
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
423
addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
806
if( types.contains( "WORK" ) )
807
type = Contacts.ContactMethods.TYPE_WORK;
809
type = Contacts.ContactMethods.TYPE_HOME;
811
addEmail( value, type, preferred );
814
private void parseADR( String[] params, String value )
817
String[] adr_parts = splitValueBySemicolon( value );
821
for( int a = 0; a < adr_parts.length; a++ ) {
822
if( value.length() > 0 ) value += "\n";
823
value += adr_parts[ a ].trim();
826
Set< String > types = extractTypes( params, Arrays.asList(
827
"PREF", "WORK", "HOME", "INTERNET" ) );
831
if( types.contains( "WORK" ) )
832
type = Contacts.ContactMethods.TYPE_WORK;
834
type = Contacts.ContactMethods.TYPE_HOME;
836
addAddress( value, type );
426
839
public void finaliseParsing()
427
throws ParseException, SkipContactException,
840
throws ParseException, SkipContactException,
430
843
// missing version (and data is present)
431
if( _version == null && _lines != null )
844
if( _version == null && _buffers != null )
432
845
throw new ParseException( R.string.error_vcf_malformed );
434
// missing name properties?
435
if( _nameLevel == NAMELEVEL_NONE )
436
throw new ParseException( R.string.error_vcf_noname );
438
// check if we should import this one? If we've already got an 'N'-
439
// type name, this will already have been done by parseN() so we
440
// mustn't do this here (or it could prompt twice!)
441
if( _nameLevel < NAMELEVEL_N && !isImportRequired( getName() ) )
442
throw new SkipContactException();
445
private String undoCharsetAndEncoding( String[] params, String value )
446
throws ParseException
448
// check encoding/charset
449
String charset, encoding;
450
if( ( charset = checkParam( params, "CHARSET" ) ) != null &&
451
!charset.equals( "UTF-8" ) && !charset.equals( "UTF-16" ) )
452
throw new ParseException( R.string.error_vcf_charset );
453
if( ( encoding = checkParam( params, "ENCODING" ) ) != null &&
454
!encoding.equals( "QUOTED-PRINTABLE" ) )
455
throw new ParseException( R.string.error_vcf_encoding );
458
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
459
return unencodeQuotedPrintable( value, charset );
847
// check if we should import this contact
849
if( !isImportRequired( this ) )
850
throw new SkipContactException();
852
catch( ContactNeedsMoreInfoException e ) {
853
throw new ParseException( R.string.error_vcf_notenoughinfo );
465
857
private String checkParam( String[] params, String name )
467
Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
859
Pattern p = Pattern.compile(
860
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
468
861
for( int i = 0; i < params.length; i++ ) {
469
862
Matcher m = p.matcher( params[ i ] );
470
863
if( m.matches() )
476
869
private Set< String > extractTypes( String[] params,
477
List< String > validTypes )
870
List< String > valid_types )
479
872
HashSet< String > types = new HashSet< String >();
481
874
// get 3.0-style TYPE= param
483
if( ( typeParam = checkParam( params, "TYPE" ) ) != null ) {
484
String[] bits = typeParam.split( "," );
485
for( int i = 0; i < bits.length; i++ )
486
if( validTypes.contains( bits[ i ] ) )
487
types.add( bits[ i ] );
876
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
877
String[] parts = type_param.split( "," );
878
for( int i = 0; i < parts.length; i++ )
879
if( valid_types.contains( parts[ i ] ) )
880
types.add( parts[ i ] );
490
883
// get 2.1-style type param
491
884
if( _version.equals( "2.1" ) ) {
492
885
for( int i = 1; i < params.length; i++ )
493
if( validTypes.contains( params[ i ] ) )
886
if( valid_types.contains( params[ i ] ) )
494
887
types.add( params[ i ] );
500
private String unencodeQuotedPrintable( String str, String charset )
893
private UnencodeResult unencodeQuotedPrintable( ByteBuffer in )
502
// default encoding scheme
503
if( charset == null ) charset = "UTF-8";
895
boolean another = false;
505
// unencode quoted-pritable encoding, as per RFC1521 section 5.1
506
byte[] bytes = new byte[ str.length() ];
897
// unencode quoted-printable encoding, as per RFC1521 section 5.1
898
byte[] out = new byte[ in.limit() - in.position() ];
508
for( int i = 0; i < str.length(); i++, j++ ) {
509
char ch = str.charAt( i );
510
if( ch == '=' && i < str.length() - 2 ) {
512
Character.digit( str.charAt( i + 1 ), 16 ) * 16 +
513
Character.digit( str.charAt( i + 2 ), 16 ) );
900
for( int i = in.position(); i < in.limit(); i++ )
902
// get next char and process...
903
byte ch = in.array()[ i ];
904
if( ch == '=' && i < in.limit() - 2 )
906
// we found a =XX format byte, add it
908
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
909
Character.digit( in.array()[ i + 2 ], 16 ) );
912
else if( ch == '=' && i == in.limit() - 1 )
914
// we found a '=' at the end of a line signifying a multi-
915
// line string, so we don't add it.
517
bytes[ j ] = (byte)ch;
520
return new String( bytes, 0, j, charset );
521
} catch( UnsupportedEncodingException e ) { }
920
// just a normal char...
925
return new UnencodeResult( another, ByteBuffer.wrap( out, 0, j ) );
928
private ByteBuffer transcodeAsciiToUtf8( ByteBuffer in )
931
byte[] out = new byte[ ( in.limit() - in.position() ) * 2 ];
933
for( int a = in.position(); a < in.limit(); a++ )
935
// if char is < 127, keep it as-is
936
if( in.array()[ a ] >= 0 )
937
out[ j++ ] = in.array()[ a ];
939
// else, convert it to UTF-8
941
int b = 0xff & (int)in.array()[ a ];
942
out[ j++ ] = (byte)( 0xc0 | ( b >> 6 ) );
943
out[ j++ ] = (byte)( 0x80 | ( b & 0x3f ) );
947
return ByteBuffer.wrap( out, 0, j );