301
387
@SuppressWarnings("serial")
302
388
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!
390
private String extractCollonPartFromLine( ByteBuffer buffer,
391
String line, boolean former )
395
// get a US-ASCII version of the line for processing, unless we were
399
line = new String( buffer.array(), buffer.position(),
400
buffer.limit() - buffer.position(), "US-ASCII" );
402
catch( UnsupportedEncodingException e ) {
403
// we know US-ASCII is supported, so appease the compiler...
408
// split line into name and value parts and check to make sure we
409
// only got 2 parts and that the first part is not zero in length
410
String[] parts = line.split( ":", 2 );
411
if( parts.length == 2 && parts[ 0 ].length() > 0 )
412
ret = parts[ former? 0 : 1 ];
417
private String extractNameAndParamsFromLine( ByteBuffer buffer,
420
return extractCollonPartFromLine( buffer, line, true );
423
private String extractValueFromLine( ByteBuffer buffer, String line )
425
return extractCollonPartFromLine( buffer, line, false );
428
public void parseLine( ByteBuffer buffer, String line,
429
boolean next_line_looks_folded )
430
throws ParseException, SkipContactException,
433
// do we have a version yet?
320
434
if( _version == null )
322
// is this a version?
323
if( props.length == 2 && props[ 0 ].equals( "VERSION" ) )
436
// tentatively get name and params from line
437
String name_and_params =
438
extractNameAndParamsFromLine( buffer, line );
440
// is it a version line?
441
if( name_and_params != null &&
442
name_and_params.equals( "VERSION" ) )
325
// yes, check/store it
326
if( !props[ 1 ].equals( "2.1" ) &&
327
!props[ 1 ].equals( "3.0" ) )
445
String value = extractValueFromLine( buffer, line );
446
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
328
447
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 ) );
450
// parse any buffers we've been accumulating while we waited
452
if( _buffers != null )
453
for( int i = 0; i < _buffers.size(); i++ )
454
parseLine( _buffers.get( i ), null,
455
i + 1 < _buffers.size() &&
456
_buffers.get( i + 1 ).hasRemaining() &&
457
_buffers.get( i + 1 ).get(
458
_buffers.get( i + 1 ).position() ) == ' ' );
339
// no, so stash this line till we have a version
341
_lines = new Vector< String >();
463
// no, so stash this line till we get a version
464
if( _buffers == null )
465
_buffers = new Vector< ByteBuffer >();
466
_buffers.add( buffer );
347
if( _parser_in_multiline )
471
// name and params and the position in the buffer where the
472
// "value" part of the line start
473
String name_and_params;
476
if( _parser_multiline_state != MULTILINE_NONE )
349
478
// if we're currently in a multi-line value, use the stored
350
479
// property name and parameters
351
props = new String[ 2 ];
352
props[ 0 ] = _parser_current_name_and_params;
353
props[ 1 ] = line.trim();
480
name_and_params = _parser_current_name_and_params;
482
// skip some initial line characters, depending on the type
483
// of multi-line we're handling
484
pos = buffer.position();
485
switch( _parser_multiline_state )
487
case MULTILINE_FOLDED:
490
case MULTILINE_ENCODED:
491
while( pos < buffer.limit() && (
492
buffer.get( pos ) == ' ' ||
493
buffer.get( pos ) == '\t' ) )
502
// take us out of multi-line so that we can re-detect that
503
// this line is a multi-line or not
504
_parser_multiline_state = MULTILINE_NONE;
357
// for normal lines, check the property name/value bits
358
if( props.length < 2 || props[ 0 ].length() == 0 )
508
// get name and params from line, and since we're not
509
// parsing a subsequent line in a multi-line, this should
510
// not fail, or it's an error
512
extractNameAndParamsFromLine( buffer, line );
513
if( name_and_params == null )
359
514
throw new ParseException(
360
515
R.string.error_vcf_malformed );
362
// ignore empty properties
363
if( props[ 1 ].length() < 1 )
517
// calculate how many chars to skip from beginning of line
518
// so we skip the property "name:" part
519
pos = buffer.position() + name_and_params.length() + 1;
366
521
// reset the saved multi-line state
367
_parser_current_name_and_params = props[ 0 ];
522
_parser_current_name_and_params = name_and_params;
368
523
_parser_buffered_value_so_far = "";
526
// get value from buffer, as raw bytes
528
value = ByteBuffer.wrap( buffer.array(), pos,
529
buffer.limit() - pos );
371
531
// get parameter parts
372
String[] params = props[ 0 ].split( ";" );
373
for( int i = 0; i < params.length; i++ )
374
params[ i ] = params[ i ].trim();
532
String[] name_param_parts = name_and_params.split( ";", -1 );
533
for( int i = 0; i < name_param_parts.length; i++ )
534
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" ) )
536
// parse encoding parameter
537
String encoding = checkParam( name_param_parts, "ENCODING" );
538
if( encoding != null ) encoding = encoding.toUpperCase();
539
if( encoding != null && !encoding.equals( "8BIT" ) &&
540
!encoding.equals( "QUOTED-PRINTABLE" ) )
386
541
//&& !encoding.equals( "BASE64" ) )
388
543
throw new ParseException( R.string.error_vcf_encoding );
546
// parse charset parameter
547
String charset = checkParam( name_param_parts, "CHARSET" );
548
if( charset != null ) charset = charset.toUpperCase();
549
if( charset != null && !charset.equals( "US-ASCII" ) &&
550
!charset.equals( "ASCII" ) &&
551
!charset.equals( "UTF-8" ) )
553
throw new ParseException( R.string.error_vcf_charset );
391
556
// do unencoding (or default to a fake unencoding result with
392
557
// the raw string)
393
UnencodeResult result;
558
UnencodeResult unencoding_result = null;
394
559
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
395
result = unencodeQuotedPrintable( props[ 1 ], charset );
560
unencoding_result = unencodeQuotedPrintable( value );
396
561
// 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 );
562
// unencoding_result = unencodeBase64( props[ 1 ], charset );
563
if( unencoding_result != null ) {
564
value = unencoding_result.getBuffer();
565
if( unencoding_result.isAnotherLineRequired() )
566
_parser_multiline_state = MULTILINE_ENCODED;
569
// convert 8-bit ASCII charset to US-ASCII
570
if( charset == null || charset.equals( "ASCII" ) ) {
571
value = transcodeAsciiToUtf8( value );
402
575
// process charset
404
props[ 1 ] = new String( result.getBytes(), 0,
405
result.getNumBytes(),
406
charset == null? "UTF-8" : charset );
578
string_value = new String( value.array(), value.position(),
579
value.limit() - value.position(), charset );
407
580
} catch( UnsupportedEncodingException e ) {
408
581
throw new ParseException( R.string.error_vcf_charset );
411
// handle multi-line requests
412
_parser_in_multiline = result.isAnotherLineRequired();
413
if( _parser_in_multiline ) {
414
_parser_buffered_value_so_far += props[ 1 ];
584
// for some entries that have semicolon-separated value parts,
585
// check to see if the value ends in an escape character, which
586
// indicates that we have a multi-line value
587
if( ( name_param_parts[ 0 ].equals( "N" ) ||
588
name_param_parts[ 0 ].equals( "ORG" ) ||
589
name_param_parts[ 0 ].equals( "ADR" ) ) &&
590
doesStringEndInAnEscapeChar( string_value ) )
592
_parser_multiline_state = MULTILINE_ESCAPED;
593
string_value = string_value.substring( 0,
594
string_value.length() - 1 );
597
// now we know whether we're in an encoding multi-line,
598
// determine if we're in a v3 folded multi-line or not
599
if( _parser_multiline_state == MULTILINE_NONE &&
600
_version.equals( "3.0" ) && next_line_looks_folded )
602
_parser_multiline_state = MULTILINE_FOLDED;
605
// handle multi-lines by buffering them and parsing them when we
606
// are processing the last line in a multi-line sequence
607
if( _parser_multiline_state != MULTILINE_NONE ) {
608
_parser_buffered_value_so_far += string_value;
611
String complete_value =
612
( _parser_buffered_value_so_far + string_value ).trim();
418
// add on buffered multi-line content
419
String value = _parser_buffered_value_so_far + props[ 1 ];
614
// ignore empty values
615
if( complete_value.length() < 1 ) return;
421
617
// 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 );
618
if( name_param_parts[ 0 ].equals( "N" ) )
619
parseN( name_param_parts, complete_value );
620
else if( name_param_parts[ 0 ].equals( "FN" ) )
621
parseFN( name_param_parts, complete_value );
622
else if( name_param_parts[ 0 ].equals( "ORG" ) )
623
parseORG( name_param_parts, complete_value );
624
else if( name_param_parts[ 0 ].equals( "TEL" ) )
625
parseTEL( name_param_parts, complete_value );
626
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
627
parseEMAIL( name_param_parts, complete_value );
628
else if( name_param_parts[ 0 ].equals( "ADR" ) )
629
parseADR( name_param_parts, complete_value );
633
private boolean doesStringEndInAnEscapeChar( String string )
635
// count the number of backslashes at the end of the string
637
for( int a = string.length() - 1; a >= 0; a-- )
638
if( string.charAt( a ) == '\\' )
643
// if there are an even number of backslashes then the final one
645
return ( count & 1 ) == 1;
648
private String[] splitValueBySemicolon( String value )
650
// split string in to parts by semicolon
651
ArrayList< String > parts = new ArrayList< String >(
652
Arrays.asList( value.split( ";" ) ) );
655
for( int a = 0; a < parts.size(); a++ )
657
String str = parts.get( a );
659
// look for parts that end in an escape character, but ignore
660
// the final part. We've already detected escape chars at the
661
// end of the final part in parseLine() and handled multi-lines
663
if( a < parts.size() - 1 &&
664
doesStringEndInAnEscapeChar( str ) )
666
// join the next part to this part and remove the next part
667
parts.set( a, str.substring( 0, str.length() - 1 ) +
668
';' + parts.get( a + 1 ) );
669
parts.remove( a + 1 );
671
// re-visit this part
676
// trim and replace string
681
String[] ret = new String[ parts.size() ];
682
return parts.toArray( ret );
435
685
private void parseN( String[] params, String value )
436
throws ParseException, SkipContactException,
686
throws ParseException, SkipContactException,
439
689
// already got a better name?
440
690
if( _name_level >= NAMELEVEL_N ) return;
442
692
// get name parts
443
String[] nameparts = value.split( ";" );
444
for( int i = 0; i < nameparts.length; i++ )
445
nameparts[ i ] = nameparts[ i ].trim();
693
String[] name_parts = splitValueBySemicolon( value );
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 ];
697
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
698
value += name_parts[ 1 ];
699
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
700
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
455
703
setName( value );
497
745
private void parseTEL( String[] params, String value )
498
throws ParseException
746
throws ParseException
500
748
if( value.length() == 0 ) return;
502
750
Set< String > types = extractTypes( params, Arrays.asList(
503
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
504
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
751
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
752
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
506
754
// here's the logic...
507
755
boolean preferred = types.contains( "PREF" );
756
int type = PhonesColumns.TYPE_MOBILE;
508
757
if( types.contains( "VOICE" ) )
509
758
if( types.contains( "WORK" ) )
510
addPhone( value, PhonesColumns.TYPE_WORK, preferred );
759
type = PhonesColumns.TYPE_WORK;
512
addPhone( value, PhonesColumns.TYPE_HOME, preferred );
761
type = PhonesColumns.TYPE_HOME;
513
762
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
514
addPhone( value, PhonesColumns.TYPE_MOBILE, preferred );
763
type = PhonesColumns.TYPE_MOBILE;
515
764
if( types.contains( "FAX" ) )
516
765
if( types.contains( "HOME" ) )
517
addPhone( value, PhonesColumns.TYPE_FAX_HOME, preferred );
766
type = PhonesColumns.TYPE_FAX_HOME;
519
addPhone( value, PhonesColumns.TYPE_FAX_WORK, preferred );
768
type = PhonesColumns.TYPE_FAX_WORK;
520
769
if( types.contains( "PAGER" ) )
521
addPhone( value, PhonesColumns.TYPE_PAGER, preferred );
770
type = PhonesColumns.TYPE_PAGER;
773
addPhone( value, type, preferred );
524
776
public void parseEMAIL( String[] params, String value )
525
throws ParseException
777
throws ParseException
527
779
if( value.length() == 0 ) return;
529
781
Set< String > types = extractTypes( params, Arrays.asList(
530
"PREF", "WORK", "HOME", "INTERNET" ) );
782
"PREF", "WORK", "HOME", "INTERNET" ) );
532
// here's the logic...
533
785
boolean preferred = types.contains( "PREF" );
534
786
if( types.contains( "WORK" ) )
535
787
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
559
834
private String checkParam( String[] params, String name )
561
Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
836
Pattern p = Pattern.compile(
837
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
562
838
for( int i = 0; i < params.length; i++ ) {
563
839
Matcher m = p.matcher( params[ i ] );
564
840
if( m.matches() )
570
846
private Set< String > extractTypes( String[] params,
571
List< String > validTypes )
847
List< String > valid_types )
573
849
HashSet< String > types = new HashSet< String >();
575
851
// 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 ] );
853
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
854
String[] parts = type_param.split( "," );
855
for( int i = 0; i < parts.length; i++ )
856
if( valid_types.contains( parts[ i ] ) )
857
types.add( parts[ i ] );
584
860
// get 2.1-style type param
585
861
if( _version.equals( "2.1" ) ) {
586
862
for( int i = 1; i < params.length; i++ )
587
if( validTypes.contains( params[ i ] ) )
863
if( valid_types.contains( params[ i ] ) )
588
864
types.add( params[ i ] );
594
private UnencodeResult unencodeQuotedPrintable( String str, String charset )
870
private UnencodeResult unencodeQuotedPrintable( ByteBuffer in )
596
872
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() ];
874
// unencode quoted-printable encoding, as per RFC1521 section 5.1
875
byte[] out = new byte[ in.limit() - in.position() ];
604
for( int i = 0; i < str.length(); i++ )
877
for( int i = in.position(); i < in.limit(); i++ )
606
879
// get next char and process...
607
char ch = str.charAt( i );
608
if( ch == '=' && i < str.length() - 2 )
880
byte ch = in.array()[ i ];
881
if( ch == '=' && i < in.limit() - 2 )
610
883
// 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 ) );
885
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
886
Character.digit( in.array()[ i + 2 ], 16 ) );
616
else if( ch == '=' && i == str.length() - 1 )
889
else if( ch == '=' && i == in.limit() - 1 )
618
891
// we found a '=' at the end of a line signifying a multi-
619
892
// line string, so we don't add it.