187
199
// go through lines
188
200
Vcard vcard = null;
201
int vcard_start_line = 0;
189
202
ContentLineIterator cli = new ContentLineIterator( content );
190
203
while( cli.hasNext() )
192
ByteBuffer buffer = cli.next();
205
ContentLine content_line = cli.next();
194
// get a US-ASCII version of the line for processing
197
line = new String( buffer.array(), buffer.position(),
198
buffer.limit() - buffer.position(), "US-ASCII" );
200
catch( UnsupportedEncodingException e ) {
201
// we know US-ASCII is supported, so appease the compiler...
207
// get a US-ASCII version of the string, for processing
208
String line = content_line.getUsAsciiLine();
205
210
if( vcard == null ) {
206
211
// look for vcard beginning
207
if( line.matches( "^BEGIN:VCARD" ) ) {
208
setProgress( ++_progress );
212
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD.*$" ) ) {
213
setProgress( _progress++ );
209
214
vcard = new Vcard();
215
vcard_start_line = cli.getLineNumber();
213
219
// look for vcard content or ending
214
if( line.matches( "^END:VCARD" ) )
220
if( line.matches( "^END[ \t]*:[ \t]*VCARD.*$" ) )
216
222
// finalise the vcard/contact
218
224
vcard.finaliseVcard();
226
// pass the finalised contact to the importer
227
importContact( vcard );
220
229
catch( Vcard.ParseException e ) {
223
getText( R.string.error_vcf_parse ).toString()
224
+ fileName + "\n" + e.getMessage() ) )
226
finish( ACTION_ABORT );
230
// pass the finalised contact to the importer
231
importContact( vcard );
231
getText( R.string.error_vcf_parse ).toString()
233
getText( R.string.error_vcf_parse_line ).toString()
234
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
236
finish( ACTION_ABORT );
241
catch( ContactData.ContactNotIdentifiableException e ) {
243
getText( R.string.error_vcf_parse ).toString()
245
getText( R.string.error_vcf_parse_line ).toString()
246
+ vcard_start_line + ":\n" + getText(
247
R.string.error_vcf_notenoughinfo ).toString()
250
finish( ACTION_ABORT );
256
// discard this vcard
238
261
// try giving the line to the vcard
240
vcard.parseLine( buffer, line,
241
cli.doesNextLineLookFolded() );
263
vcard.parseLine( content_line );
243
265
catch( Vcard.ParseException e ) {
245
267
if( !showContinue(
246
268
getText( R.string.error_vcf_parse ).toString()
247
+ fileName + "\n" + e.getMessage() ) )
270
getText( R.string.error_vcf_parse_line ).toString()
271
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
249
273
finish( ACTION_ABORT );
321
394
* onto the end of this one?
324
public boolean doesNextLineLookFolded()
397
private boolean doesNextLineLookFolded()
326
399
return _pos > 0 && _pos < _content.length &&
327
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
400
_content[ _pos - 1 ] == '\n' &&
401
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
404
public int getLineNumber()
331
410
private class Vcard extends ContactData
333
412
private final static int NAMELEVEL_NONE = 0;
334
private final static int NAMELEVEL_FN = 1;
335
private final static int NAMELEVEL_N = 2;
413
private final static int NAMELEVEL_N = 1;
414
private final static int NAMELEVEL_FN = 2;
337
416
private final static int MULTILINE_NONE = 0;
338
417
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
339
418
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
340
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
419
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
342
421
private String _version = null;
343
private Vector< ByteBuffer > _buffers = null;
422
private Vector< ContentLine > _content_lines = null;
344
423
private int _name_level = NAMELEVEL_NONE;
345
424
private int _parser_multiline_state = MULTILINE_NONE;
346
425
private String _parser_current_name_and_params = null;
389
468
@SuppressWarnings("serial")
390
469
protected class SkipImportException extends Exception { }
392
private String extractCollonPartFromLine( ByteBuffer buffer,
393
String line, boolean former )
471
private String extractCollonPartFromLine( ContentLine content_line,
395
474
String ret = null;
397
// get a US-ASCII version of the line for processing, unless we were
401
line = new String( buffer.array(), buffer.position(),
402
buffer.limit() - buffer.position(), "US-ASCII" );
404
catch( UnsupportedEncodingException e ) {
405
// we know US-ASCII is supported, so appease the compiler...
410
476
// split line into name and value parts and check to make sure we
411
477
// only got 2 parts and that the first part is not zero in length
412
String[] parts = line.split( ":", 2 );
478
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
413
479
if( parts.length == 2 && parts[ 0 ].length() > 0 )
414
480
ret = parts[ former? 0 : 1 ];
419
private String extractNameAndParamsFromLine( ByteBuffer buffer,
422
return extractCollonPartFromLine( buffer, line, true );
425
private String extractValueFromLine( ByteBuffer buffer, String line )
427
return extractCollonPartFromLine( buffer, line, false );
430
public void parseLine( ByteBuffer buffer, String line,
431
boolean next_line_looks_folded )
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 )
432
496
throws ParseException, SkipImportException,
433
497
AbortImportException
438
502
// tentatively get name and params from line
439
503
String name_and_params =
440
extractNameAndParamsFromLine( buffer, line );
504
extractNameAndParamsFromLine( content_line );
442
506
// is it a version line?
443
507
if( name_and_params != null &&
444
name_and_params.equals( "VERSION" ) )
508
name_and_params.equalsIgnoreCase( "VERSION" ) )
447
String value = extractValueFromLine( buffer, line );
511
String value = extractValueFromLine( content_line ).trim();
448
512
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
449
513
throw new ParseException( R.string.error_vcf_version );
450
514
_version = value;
452
516
// parse any buffers we've been accumulating while we waited
454
if( _buffers != null )
455
for( int i = 0; i < _buffers.size(); i++ )
456
parseLine( _buffers.get( i ), null,
457
i + 1 < _buffers.size() &&
458
_buffers.get( i + 1 ).hasRemaining() &&
459
_buffers.get( i + 1 ).get(
460
_buffers.get( i + 1 ).position() ) == ' ' );
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;
465
525
// no, so stash this line till we get a version
466
if( _buffers == null )
467
_buffers = new Vector< ByteBuffer >();
468
_buffers.add( buffer );
526
if( _content_lines == null )
527
_content_lines = new Vector< ContentLine >();
528
_content_lines.add( content_line );
473
533
// name and params and the position in the buffer where the
474
// "value" part of the line start
534
// "value" part of the line starts
475
535
String name_and_params;
528
593
// get value from buffer, as raw bytes
529
594
ByteBuffer value;
530
value = ByteBuffer.wrap( buffer.array(), pos,
531
buffer.limit() - pos );
595
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
596
content_line.getBuffer().limit() - pos );
533
598
// get parameter parts
534
599
String[] name_param_parts = name_and_params.split( ";", -1 );
535
600
for( int i = 0; i < name_param_parts.length; i++ )
536
601
name_param_parts[ i ] = name_param_parts[ i ].trim();
603
// determine whether we care about this entry
604
final HashSet< String > interesting_fields =
605
new HashSet< String >( Arrays.asList( new String[] { "N",
606
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
608
boolean is_interesting_field =
609
interesting_fields.contains(
610
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
538
612
// parse encoding parameter
539
613
String encoding = checkParam( name_param_parts, "ENCODING" );
540
if( encoding != null ) encoding = encoding.toUpperCase();
541
if( encoding != null && !encoding.equals( "8BIT" ) &&
542
!encoding.equals( "QUOTED-PRINTABLE" ) )
543
//&& !encoding.equals( "BASE64" ) )
614
if( encoding != null )
615
encoding = encoding.toUpperCase( Locale.US );
616
if( is_interesting_field && encoding != null &&
617
!encoding.equalsIgnoreCase( "8BIT" ) &&
618
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
619
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
545
621
throw new ParseException( R.string.error_vcf_encoding );
548
624
// parse charset parameter
549
625
String charset = checkParam( name_param_parts, "CHARSET" );
550
if( charset != null ) charset = charset.toUpperCase();
551
if( charset != null && !charset.equals( "US-ASCII" ) &&
552
!charset.equals( "ASCII" ) &&
553
!charset.equals( "UTF-8" ) )
626
if( charset != null )
627
charset = charset.toUpperCase( Locale.US );
628
if( charset != null &&
629
!charset.equalsIgnoreCase( "US-ASCII" ) &&
630
!charset.equalsIgnoreCase( "ASCII" ) &&
631
!charset.equalsIgnoreCase( "UTF-8" ) )
555
633
throw new ParseException( R.string.error_vcf_charset );
558
636
// do unencoding (or default to a fake unencoding result with
559
637
// the raw string)
560
638
UnencodeResult unencoding_result = null;
561
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
639
if( encoding != null &&
640
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
562
642
unencoding_result = unencodeQuotedPrintable( value );
563
// else if( encoding != null && encoding.equals( "BASE64" ) )
644
// else if( encoding != null &&
645
// encoding.equalsIgnoreCase( "BASE64" ) )
564
647
// unencoding_result = unencodeBase64( props[ 1 ], charset );
565
649
if( unencoding_result != null ) {
566
650
value = unencoding_result.getBuffer();
567
651
if( unencoding_result.isAnotherLineRequired() )
568
652
_parser_multiline_state = MULTILINE_ENCODED;
571
// convert 8-bit ASCII charset to US-ASCII
572
if( charset == null || charset.equals( "ASCII" ) ) {
655
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
656
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
657
if( ( charset == null && _version.equals( "2.1" ) ) ||
658
( charset != null && (
659
charset.equalsIgnoreCase( "ASCII" ) ||
660
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
573
662
value = transcodeAsciiToUtf8( value );
665
// process charset (value is now in UTF-8)
578
666
String string_value;
580
668
string_value = new String( value.array(), value.position(),
581
value.limit() - value.position(), charset );
669
value.limit() - value.position(), "UTF-8" );
582
670
} catch( UnsupportedEncodingException e ) {
583
671
throw new ParseException( R.string.error_vcf_charset );
617
705
if( complete_value.length() < 1 ) return;
619
707
// parse some properties
620
if( name_param_parts[ 0 ].equals( "N" ) )
708
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
621
709
parseN( name_param_parts, complete_value );
622
else if( name_param_parts[ 0 ].equals( "FN" ) )
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
623
711
parseFN( name_param_parts, complete_value );
624
else if( name_param_parts[ 0 ].equals( "ORG" ) )
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
625
713
parseORG( name_param_parts, complete_value );
626
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
627
715
parseTITLE( name_param_parts, complete_value );
628
else if( name_param_parts[ 0 ].equals( "TEL" ) )
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
629
717
parseTEL( name_param_parts, complete_value );
630
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
631
719
parseEMAIL( name_param_parts, complete_value );
632
else if( name_param_parts[ 0 ].equals( "ADR" ) )
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
633
721
parseADR( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
723
parseLABEL( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
725
parseNOTE( name_param_parts, complete_value );
686
779
return parts.toArray( ret );
782
private String unescapeValue( String value )
784
StringBuilder ret = new StringBuilder( value.length() );
785
boolean in_escape = false;
786
for( int a = 0; a < value.length(); a++ )
788
int c = value.codePointAt( a );
790
// process a normal character
795
ret.append( Character.toChars( c ) );
799
// process an escape sequence
805
// add tab (invalid/non-standard, but accepted)
816
// add escaped character
817
ret.append( Character.toChars( c ) );
820
// unknown escape sequence, so add it unescaped
821
// (invalid/non-standard, but accepted)
823
ret.append( Character.toChars( c ) );
828
return ret.toString();
689
831
private void parseN( String[] params, String value )
691
833
// already got a better name?
692
834
if( _name_level >= NAMELEVEL_N ) return;
694
836
// get name parts
695
String[] name_parts = splitValueBySemicolon( value );
837
String[] name_parts = splitValueByCharacter( value, ';' );
699
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
700
value += name_parts[ 1 ];
701
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
702
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
841
final int[] part_order = { 3, 1, 2, 0, 4 };
842
for( int a = 0; a < part_order.length; a++ )
843
if( name_parts.length > part_order[ a ] &&
844
name_parts[ part_order[ a ] ].length() > 0 )
846
// split this part in to it's comma-separated bits
847
String[] name_part_parts = splitValueByCharacter(
848
name_parts[ part_order[ a ] ], ',' );
849
for( int b = 0; b < name_part_parts.length; b++ )
850
if( name_part_parts[ b ].length() > 0 )
852
if( value.length() > 0 ) value += " ";
853
value += name_part_parts[ b ];
858
setName( unescapeValue( value ) );
706
859
_name_level = NAMELEVEL_N;
802
957
boolean is_preferred = types.contains( "PREF" );
804
959
if( types.contains( "WORK" ) )
805
type = Contacts.ContactMethods.TYPE_WORK;
807
type = Contacts.ContactMethods.TYPE_HOME;
809
addEmail( value, type, is_preferred );
964
addEmail( unescapeValue( value ), type, is_preferred );
812
967
private void parseADR( String[] params, String value )
814
969
// get address parts
815
String[] adr_parts = splitValueBySemicolon( value );
970
String[] adr_parts = splitValueByCharacter( value, ';' );
819
for( int a = 0; a < adr_parts.length; a++ ) {
820
if( value.length() > 0 ) value += "\n";
821
value += adr_parts[ a ].trim();
824
Set< String > types = extractTypes( params, Arrays.asList(
825
"PREF", "WORK", "HOME", "INTERNET" ) );
829
if( types.contains( "WORK" ) )
830
type = Contacts.ContactMethods.TYPE_WORK;
832
type = Contacts.ContactMethods.TYPE_HOME;
834
addAddress( value, type );
974
for( int a = 0; a < adr_parts.length; a++ )
975
if( adr_parts[ a ].length() > 0 )
977
// version 3.0 vCards allow further splitting by comma
978
if( _version.equals( "3.0" ) )
980
// split this part in to it's comma-separated bits and
981
// add them on individual lines
982
String[] adr_part_parts =
983
splitValueByCharacter( adr_parts[ a ], ',' );
984
for( int b = 0; b < adr_part_parts.length; b++ )
985
if( adr_part_parts[ b ].length() > 0 )
987
if( value.length() > 0 ) value += "\n";
988
value += adr_part_parts[ b ];
993
// add this part on an individual line
994
if( value.length() > 0 ) value += "\n";
995
value += adr_parts[ a ];
999
Set< String > types = extractTypes( params, Arrays.asList(
1000
"PREF", "WORK", "HOME" ) );
1004
if( types.contains( "WORK" ) )
1009
addAddress( unescapeValue( value ), type );
1012
private void parseLABEL( String[] params, String value )
1014
Set< String > types = extractTypes( params, Arrays.asList(
1015
"PREF", "WORK", "HOME" ) );
1019
if( types.contains( "WORK" ) )
1024
addAddress( unescapeValue( value ), type );
1027
private void parseNOTE( String[] params, String value )
1029
addNote( unescapeValue( value ) );
837
1032
public void finaliseVcard()
838
throws ParseException
1033
throws ParseException, ContactNotIdentifiableException
840
1035
// missing version (and data is present)
841
if( _version == null && _buffers != null )
1036
if( _version == null && _content_lines != null )
842
1037
throw new ParseException( R.string.error_vcf_malformed );
844
1039
// finalise the parent class
848
catch( ContactNotIdentifiableException e ) {
849
throw new ParseException( R.string.error_vcf_notenoughinfo );
1044
* Amongst the params, find the value of the first, only, of any with
1045
* the specified name
1048
* @return a value, or null
853
1050
private String checkParam( String[] params, String name )
1052
String[] res = checkParams( params, name );
1053
return res.length > 0? res[ 0 ] : null;
1057
* Amongst the params, find the values of any with the specified name
1060
* @return an array of values, or null
1062
private String[] checkParams( String[] params, String name )
1064
HashSet< String > ret = new HashSet< String >();
855
1066
Pattern p = Pattern.compile(
856
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1067
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1068
Pattern.CASE_INSENSITIVE );
857
1069
for( int i = 0; i < params.length; i++ ) {
858
1070
Matcher m = p.matcher( params[ i ] );
859
1071
if( m.matches() )
1072
ret.add( m.group( 2 ) );
1075
return (String[]) ret.toArray( new String[ ret.size() ] );
1079
* Amongst the params, return any type values present. For v2.1 vCards,
1080
* those types are just parameters. For v3.0, they are prefixed with
1081
* "TYPE=". There may also be multiple type parameters.
1082
* @param params an array of params to look for types in
1083
* @param valid_types an list of upper-case type values to look for
1084
* @return a set of present type values
865
1086
private Set< String > extractTypes( String[] params,
866
1087
List< String > valid_types )
868
1089
HashSet< String > types = new HashSet< String >();
870
1091
// get 3.0-style TYPE= param
872
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
873
String[] parts = type_param.split( "," );
874
for( int i = 0; i < parts.length; i++ )
875
if( valid_types.contains( parts[ i ] ) )
876
types.add( parts[ i ] );
1092
String type_params[] = checkParams( params, "TYPE" );
1093
for( int a = 0; a < type_params.length; a++ )
1095
// check for a comma-separated list of types (why? I don't think
1096
// this is in the specs!)
1097
String[] parts = type_params[ a ].split( "," );
1098
for( int i = 0; i < parts.length; i++ ) {
1099
String ucpart = parts[ i ].toUpperCase( Locale.US );
1100
if( valid_types.contains( ucpart ) )
1101
types.add( ucpart );
879
1105
// get 2.1-style type param
880
1106
if( _version.equals( "2.1" ) ) {
881
for( int i = 1; i < params.length; i++ )
882
if( valid_types.contains( params[ i ] ) )
883
types.add( params[ i ] );
1107
for( int i = 1; i < params.length; i++ ) {
1108
String ucparam = params[ i ].toUpperCase( Locale.US );
1109
if( valid_types.contains( ucparam ) )
1110
types.add( ucparam );