187
193
// go through lines
188
194
Vcard vcard = null;
195
int vcard_start_line = 0;
189
196
ContentLineIterator cli = new ContentLineIterator( content );
190
197
while( cli.hasNext() )
192
ByteBuffer buffer = cli.next();
199
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...
201
// get a US-ASCII version of the string, for processing
202
String line = content_line.getUsAsciiLine();
205
204
if( vcard == null ) {
206
205
// look for vcard beginning
207
if( line.matches( "^BEGIN:VCARD" ) ) {
208
setProgress( ++_progress );
206
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD" ) ) {
207
setProgress( _progress++ );
209
208
vcard = new Vcard();
209
vcard_start_line = cli.getLineNumber();
213
213
// look for vcard content or ending
214
if( line.matches( "^END:VCARD" ) )
214
if( line.matches( "^END[ \t]*:[ \t]*VCARD" ) )
216
216
// finalise the vcard/contact
218
218
vcard.finaliseVcard();
220
// pass the finalised contact to the importer
221
importContact( vcard );
220
223
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 );
225
getText( R.string.error_vcf_parse ).toString()
227
getText( R.string.error_vcf_parse_line ).toString()
228
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
230
finish( ACTION_ABORT );
235
catch( ContactData.ContactNotIdentifiableException e ) {
237
getText( R.string.error_vcf_parse ).toString()
239
getText( R.string.error_vcf_parse_line ).toString()
240
+ vcard_start_line + ":\n" + getText(
241
R.string.error_vcf_notenoughinfo ).toString()
244
finish( ACTION_ABORT );
250
// discard this vcard
238
255
// try giving the line to the vcard
240
vcard.parseLine( buffer, line,
241
cli.doesNextLineLookFolded() );
257
vcard.parseLine( content_line );
243
259
catch( Vcard.ParseException e ) {
245
261
if( !showContinue(
246
262
getText( R.string.error_vcf_parse ).toString()
247
+ fileName + "\n" + e.getMessage() ) )
264
getText( R.string.error_vcf_parse_line ).toString()
265
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
249
267
finish( ACTION_ABORT );
321
388
* onto the end of this one?
324
public boolean doesNextLineLookFolded()
391
private boolean doesNextLineLookFolded()
326
393
return _pos > 0 && _pos < _content.length &&
327
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
394
_content[ _pos - 1 ] == '\n' &&
395
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
398
public int getLineNumber()
331
404
private class Vcard extends ContactData
333
406
private final static int NAMELEVEL_NONE = 0;
334
private final static int NAMELEVEL_FN = 1;
335
private final static int NAMELEVEL_N = 2;
407
private final static int NAMELEVEL_N = 1;
408
private final static int NAMELEVEL_FN = 2;
337
410
private final static int MULTILINE_NONE = 0;
338
411
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
339
412
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
340
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
413
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
342
415
private String _version = null;
343
private Vector< ByteBuffer > _buffers = null;
416
private Vector< ContentLine > _content_lines = null;
344
417
private int _name_level = NAMELEVEL_NONE;
345
418
private int _parser_multiline_state = MULTILINE_NONE;
346
419
private String _parser_current_name_and_params = null;
389
462
@SuppressWarnings("serial")
390
463
protected class SkipImportException extends Exception { }
392
private String extractCollonPartFromLine( ByteBuffer buffer,
393
String line, boolean former )
465
private String extractCollonPartFromLine( ContentLine content_line,
395
468
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
470
// split line into name and value parts and check to make sure we
411
471
// only got 2 parts and that the first part is not zero in length
412
String[] parts = line.split( ":", 2 );
472
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
413
473
if( parts.length == 2 && parts[ 0 ].length() > 0 )
414
474
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 )
479
private String extractNameAndParamsFromLine( ContentLine content_line )
481
return extractCollonPartFromLine( content_line, true );
484
private String extractValueFromLine( ContentLine content_line )
486
return extractCollonPartFromLine( content_line, false );
489
public void parseLine( ContentLine content_line )
432
490
throws ParseException, SkipImportException,
433
491
AbortImportException
438
496
// tentatively get name and params from line
439
497
String name_and_params =
440
extractNameAndParamsFromLine( buffer, line );
498
extractNameAndParamsFromLine( content_line );
442
500
// is it a version line?
443
501
if( name_and_params != null &&
444
name_and_params.equals( "VERSION" ) )
502
name_and_params.equalsIgnoreCase( "VERSION" ) )
447
String value = extractValueFromLine( buffer, line );
505
String value = extractValueFromLine( content_line );
448
506
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
449
507
throw new ParseException( R.string.error_vcf_version );
450
508
_version = value;
452
510
// 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() ) == ' ' );
512
if( _content_lines != null )
513
for( int i = 0; i < _content_lines.size(); i++ )
514
parseLine( _content_lines.get( i ) );
515
_content_lines = null;
465
519
// no, so stash this line till we get a version
466
if( _buffers == null )
467
_buffers = new Vector< ByteBuffer >();
468
_buffers.add( buffer );
520
if( _content_lines == null )
521
_content_lines = new Vector< ContentLine >();
522
_content_lines.add( content_line );
473
527
// name and params and the position in the buffer where the
474
// "value" part of the line start
528
// "value" part of the line starts
475
529
String name_and_params;
528
587
// get value from buffer, as raw bytes
529
588
ByteBuffer value;
530
value = ByteBuffer.wrap( buffer.array(), pos,
531
buffer.limit() - pos );
589
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
590
content_line.getBuffer().limit() - pos );
533
592
// get parameter parts
534
593
String[] name_param_parts = name_and_params.split( ";", -1 );
535
594
for( int i = 0; i < name_param_parts.length; i++ )
536
595
name_param_parts[ i ] = name_param_parts[ i ].trim();
597
// determine whether we care about this entry
598
final HashSet< String > interesting_fields =
599
new HashSet< String >( Arrays.asList( new String[] { "N",
600
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
602
boolean is_interesting_field =
603
interesting_fields.contains(
604
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
538
606
// parse encoding parameter
539
607
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" ) )
608
if( encoding != null )
609
encoding = encoding.toUpperCase( Locale.US );
610
if( is_interesting_field && encoding != null &&
611
!encoding.equalsIgnoreCase( "8BIT" ) &&
612
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
613
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
545
615
throw new ParseException( R.string.error_vcf_encoding );
548
618
// parse charset parameter
549
619
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" ) )
620
if( charset != null )
621
charset = charset.toUpperCase( Locale.US );
622
if( charset != null &&
623
!charset.equalsIgnoreCase( "US-ASCII" ) &&
624
!charset.equalsIgnoreCase( "ASCII" ) &&
625
!charset.equalsIgnoreCase( "UTF-8" ) )
555
627
throw new ParseException( R.string.error_vcf_charset );
558
630
// do unencoding (or default to a fake unencoding result with
559
631
// the raw string)
560
632
UnencodeResult unencoding_result = null;
561
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
633
if( encoding != null &&
634
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
562
636
unencoding_result = unencodeQuotedPrintable( value );
563
// else if( encoding != null && encoding.equals( "BASE64" ) )
638
// else if( encoding != null &&
639
// encoding.equalsIgnoreCase( "BASE64" ) )
564
641
// unencoding_result = unencodeBase64( props[ 1 ], charset );
565
643
if( unencoding_result != null ) {
566
644
value = unencoding_result.getBuffer();
567
645
if( unencoding_result.isAnotherLineRequired() )
568
646
_parser_multiline_state = MULTILINE_ENCODED;
571
// convert 8-bit ASCII charset to US-ASCII
572
if( charset == null || charset.equals( "ASCII" ) ) {
649
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
650
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
651
if( ( charset == null && _version.equals( "2.1" ) ) ||
652
( charset != null && (
653
charset.equalsIgnoreCase( "ASCII" ) ||
654
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
573
656
value = transcodeAsciiToUtf8( value );
659
// process charset (value is now in UTF-8)
578
660
String string_value;
580
662
string_value = new String( value.array(), value.position(),
581
value.limit() - value.position(), charset );
663
value.limit() - value.position(), "UTF-8" );
582
664
} catch( UnsupportedEncodingException e ) {
583
665
throw new ParseException( R.string.error_vcf_charset );
617
699
if( complete_value.length() < 1 ) return;
619
701
// parse some properties
620
if( name_param_parts[ 0 ].equals( "N" ) )
702
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
621
703
parseN( name_param_parts, complete_value );
622
else if( name_param_parts[ 0 ].equals( "FN" ) )
704
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
623
705
parseFN( name_param_parts, complete_value );
624
else if( name_param_parts[ 0 ].equals( "ORG" ) )
706
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
625
707
parseORG( name_param_parts, complete_value );
626
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
708
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
627
709
parseTITLE( name_param_parts, complete_value );
628
else if( name_param_parts[ 0 ].equals( "TEL" ) )
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
629
711
parseTEL( name_param_parts, complete_value );
630
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
631
713
parseEMAIL( name_param_parts, complete_value );
632
else if( name_param_parts[ 0 ].equals( "ADR" ) )
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
633
715
parseADR( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
717
parseLABEL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
719
parseNOTE( name_param_parts, complete_value );
686
773
return parts.toArray( ret );
776
private String unescapeValue( String value )
778
StringBuilder ret = new StringBuilder( value.length() );
779
boolean in_escape = false;
780
for( int a = 0; a < value.length(); a++ )
782
int c = value.codePointAt( a );
784
// process a normal character
789
ret.append( Character.toChars( c ) );
793
// process an escape sequence
799
// add tab (invalid/non-standard, but accepted)
810
// add escaped character
811
ret.append( Character.toChars( c ) );
814
// unknown escape sequence, so add it unescaped
815
// (invalid/non-standard, but accepted)
817
ret.append( Character.toChars( c ) );
822
return ret.toString();
689
825
private void parseN( String[] params, String value )
691
827
// already got a better name?
692
828
if( _name_level >= NAMELEVEL_N ) return;
694
830
// get name parts
695
String[] name_parts = splitValueBySemicolon( value );
831
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 ];
835
final int[] part_order = { 3, 1, 2, 0, 4 };
836
for( int a = 0; a < part_order.length; a++ )
837
if( name_parts.length > part_order[ a ] &&
838
name_parts[ part_order[ a ] ].length() > 0 )
840
// split this part in to it's comma-separated bits
841
String[] name_part_parts = splitValueByCharacter(
842
name_parts[ part_order[ a ] ], ',' );
843
for( int b = 0; b < name_part_parts.length; b++ )
844
if( name_part_parts[ b ].length() > 0 )
846
if( value.length() > 0 ) value += " ";
847
value += name_part_parts[ b ];
852
setName( unescapeValue( value ) );
706
853
_name_level = NAMELEVEL_N;
802
951
boolean is_preferred = types.contains( "PREF" );
804
953
if( types.contains( "WORK" ) )
805
type = Contacts.ContactMethods.TYPE_WORK;
807
type = Contacts.ContactMethods.TYPE_HOME;
809
addEmail( value, type, is_preferred );
958
addEmail( unescapeValue( value ), type, is_preferred );
812
961
private void parseADR( String[] params, String value )
814
963
// get address parts
815
String[] adr_parts = splitValueBySemicolon( value );
964
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 );
968
for( int a = 0; a < adr_parts.length; a++ )
969
if( adr_parts[ a ].length() > 0 )
971
// version 3.0 vCards allow further splitting by comma
972
if( _version.equals( "3.0" ) )
974
// split this part in to it's comma-separated bits and
975
// add them on individual lines
976
String[] adr_part_parts =
977
splitValueByCharacter( adr_parts[ a ], ',' );
978
for( int b = 0; b < adr_part_parts.length; b++ )
979
if( adr_part_parts[ b ].length() > 0 )
981
if( value.length() > 0 ) value += "\n";
982
value += adr_part_parts[ b ];
987
// add this part on an individual line
988
if( value.length() > 0 ) value += "\n";
989
value += adr_parts[ a ];
993
Set< String > types = extractTypes( params, Arrays.asList(
994
"PREF", "WORK", "HOME" ) );
998
if( types.contains( "WORK" ) )
1003
addAddress( unescapeValue( value ), type );
1006
private void parseLABEL( String[] params, String value )
1008
Set< String > types = extractTypes( params, Arrays.asList(
1009
"PREF", "WORK", "HOME" ) );
1013
if( types.contains( "WORK" ) )
1018
addAddress( unescapeValue( value ), type );
1021
private void parseNOTE( String[] params, String value )
1023
addNote( unescapeValue( value ) );
837
1026
public void finaliseVcard()
838
throws ParseException
1027
throws ParseException, ContactNotIdentifiableException
840
1029
// missing version (and data is present)
841
if( _version == null && _buffers != null )
1030
if( _version == null && _content_lines != null )
842
1031
throw new ParseException( R.string.error_vcf_malformed );
844
1033
// finalise the parent class
848
catch( ContactNotIdentifiableException e ) {
849
throw new ParseException( R.string.error_vcf_notenoughinfo );
1038
* Amongst the params, find the value of the first, only, of any with
1039
* the specified name
1042
* @return a value, or null
853
1044
private String checkParam( String[] params, String name )
1046
String[] res = checkParams( params, name );
1047
return res.length > 0? res[ 0 ] : null;
1051
* Amongst the params, find the values of any with the specified name
1054
* @return an array of values, or null
1056
private String[] checkParams( String[] params, String name )
1058
HashSet< String > ret = new HashSet< String >();
855
1060
Pattern p = Pattern.compile(
856
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1061
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1062
Pattern.CASE_INSENSITIVE );
857
1063
for( int i = 0; i < params.length; i++ ) {
858
1064
Matcher m = p.matcher( params[ i ] );
859
1065
if( m.matches() )
1066
ret.add( m.group( 2 ) );
1069
return (String[]) ret.toArray( new String[ ret.size() ] );
1073
* Amongst the params, return any type values present. For v2.1 vCards,
1074
* those types are just parameters. For v3.0, they are prefixed with
1075
* "TYPE=". There may also be multiple type parameters.
1076
* @param params an array of params to look for types in
1077
* @param valid_types an list of upper-case type values to look for
1078
* @return a set of present type values
865
1080
private Set< String > extractTypes( String[] params,
866
1081
List< String > valid_types )
868
1083
HashSet< String > types = new HashSet< String >();
870
1085
// 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 ] );
1086
String type_params[] = checkParams( params, "TYPE" );
1087
for( int a = 0; a < type_params.length; a++ )
1089
// check for a comma-separated list of types (why? I don't think
1090
// this is in the specs!)
1091
String[] parts = type_params[ a ].split( "," );
1092
for( int i = 0; i < parts.length; i++ ) {
1093
String ucpart = parts[ i ].toUpperCase( Locale.US );
1094
if( valid_types.contains( ucpart ) )
1095
types.add( ucpart );
879
1099
// get 2.1-style type param
880
1100
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 ] );
1101
for( int i = 1; i < params.length; i++ ) {
1102
String ucparam = params[ i ].toUpperCase( Locale.US );
1103
if( valid_types.contains( ucparam ) )
1104
types.add( ucparam );