200
187
// go through lines
201
188
Vcard vcard = null;
202
int vcard_start_line = 0;
203
189
ContentLineIterator cli = new ContentLineIterator( content );
204
190
while( cli.hasNext() )
206
ContentLine content_line = cli.next();
192
ByteBuffer buffer = cli.next();
208
// get a US-ASCII version of the string, for processing
209
String line = content_line.getUsAsciiLine();
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...
211
205
if( vcard == null ) {
212
206
// look for vcard beginning
213
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
214
setProgress( _progress++ );
207
if( line.matches( "^BEGIN:VCARD" ) ) {
208
setProgress( ++_progress );
215
209
vcard = new Vcard();
216
vcard_start_line = cli.getLineNumber();
220
213
// look for vcard content or ending
221
if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
214
if( line.matches( "^END:VCARD" ) )
223
// finalise the vcard/contact
216
// store vcard and do away with it
225
vcard.finaliseVcard();
227
// pass the finalised contact to the importer
218
vcard.finaliseParsing();
228
219
importContact( vcard );
230
221
catch( Vcard.ParseException e ) {
232
getText( R.string.error_vcf_parse ).toString()
234
getText( R.string.error_vcf_parse_line ).toString()
235
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
237
finish( ACTION_ABORT );
242
catch( ContactData.ContactNotIdentifiableException e ) {
244
getText( R.string.error_vcf_parse ).toString()
246
getText( R.string.error_vcf_parse_line ).toString()
247
+ vcard_start_line + ":\n" + getText(
248
R.string.error_vcf_notenoughinfo ).toString()
251
finish( ACTION_ABORT );
257
// discard this vcard
224
getText( R.string.error_vcf_parse ).toString()
225
+ fileName + "\n" + e.getMessage() ) )
227
finish( ACTION_ABORT );
230
catch( Vcard.SkipContactException e ) {
262
238
// try giving the line to the vcard
264
vcard.parseLine( content_line );
240
vcard.parseLine( buffer, line,
241
cli.doesNextLineLookFolded() );
266
243
catch( Vcard.ParseException e ) {
268
245
if( !showContinue(
269
246
getText( R.string.error_vcf_parse ).toString()
271
getText( R.string.error_vcf_parse_line ).toString()
272
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
247
+ fileName + "\n" + e.getMessage() ) )
274
249
finish( ACTION_ABORT );
277
// Although we're continuing, we still need to abort
278
// this vCard. Further lines will be ignored until we
252
// although we're continuing, we still need to abort
253
// this vCard. Further lines will be ignored until we
279
254
// get to another BEGIN:VCARD line.
282
catch( Vcard.SkipImportException e ) {
257
catch( Vcard.SkipContactException e ) {
284
// Abort this vCard. Further lines will be ignored until
259
// abort this vCard. Further lines will be ignored until
285
260
// we get to another BEGIN:VCARD line.
395
321
* onto the end of this one?
398
private boolean doesNextLineLookFolded()
324
public boolean doesNextLineLookFolded()
400
326
return _pos > 0 && _pos < _content.length &&
401
_content[ _pos - 1 ] == '\n' &&
402
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
405
public int getLineNumber()
327
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
411
331
private class Vcard extends ContactData
413
333
private final static int NAMELEVEL_NONE = 0;
414
private final static int NAMELEVEL_N = 1;
415
private final static int NAMELEVEL_FN = 2;
334
private final static int NAMELEVEL_FN = 1;
335
private final static int NAMELEVEL_N = 2;
417
337
private final static int MULTILINE_NONE = 0;
418
338
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
419
339
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
420
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
340
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
422
342
private String _version = null;
423
private Vector< ContentLine > _content_lines = null;
343
private Vector< ByteBuffer > _buffers = null;
424
344
private int _name_level = NAMELEVEL_NONE;
425
345
private int _parser_multiline_state = MULTILINE_NONE;
426
346
private String _parser_current_name_and_params = null;
469
389
@SuppressWarnings("serial")
470
protected class SkipImportException extends Exception { }
390
protected class SkipContactException extends Exception { }
472
private String extractCollonPartFromLine( ContentLine content_line,
392
private String extractCollonPartFromLine( ByteBuffer buffer,
393
String line, boolean former )
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...
475
410
// split line into name and value parts and check to make sure we
476
411
// only got 2 parts and that the first part is not zero in length
477
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
412
String[] parts = line.split( ":", 2 );
478
413
if( parts.length == 2 && parts[ 0 ].length() > 0 )
479
return parts[ former? 0 : 1 ].trim();
484
private String extractNameAndParamsFromLine( ContentLine content_line )
486
return extractCollonPartFromLine( content_line, true );
489
private String extractValueFromLine( ContentLine content_line )
491
return extractCollonPartFromLine( content_line, false );
494
public void parseLine( ContentLine content_line )
495
throws ParseException, SkipImportException,
414
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 )
432
throws ParseException, SkipContactException,
496
433
AbortImportException
498
435
// do we have a version yet?
501
438
// tentatively get name and params from line
502
439
String name_and_params =
503
extractNameAndParamsFromLine( content_line );
440
extractNameAndParamsFromLine( buffer, line );
505
442
// is it a version line?
506
443
if( name_and_params != null &&
507
name_and_params.equalsIgnoreCase( "VERSION" ) )
444
name_and_params.equals( "VERSION" ) )
510
String value = extractValueFromLine( content_line );
511
if( value == null || (
512
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
447
String value = extractValueFromLine( buffer, line );
448
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
514
449
throw new ParseException( R.string.error_vcf_version );
516
450
_version = value;
518
452
// parse any buffers we've been accumulating while we waited
520
if( _content_lines != null )
521
for( int i = 0; i < _content_lines.size(); i++ )
522
parseLine( _content_lines.get( i ) );
523
_content_lines = null;
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() ) == ' ' );
527
465
// no, so stash this line till we get a version
528
if( _content_lines == null )
529
_content_lines = new Vector< ContentLine >();
530
_content_lines.add( content_line );
466
if( _buffers == null )
467
_buffers = new Vector< ByteBuffer >();
468
_buffers.add( buffer );
535
473
// name and params and the position in the buffer where the
536
// "value" part of the line starts
474
// "value" part of the line start
537
475
String name_and_params;
595
528
// get value from buffer, as raw bytes
596
529
ByteBuffer value;
597
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
598
content_line.getBuffer().limit() - pos );
530
value = ByteBuffer.wrap( buffer.array(), pos,
531
buffer.limit() - pos );
600
533
// get parameter parts
601
534
String[] name_param_parts = name_and_params.split( ";", -1 );
602
535
for( int i = 0; i < name_param_parts.length; i++ )
603
536
name_param_parts[ i ] = name_param_parts[ i ].trim();
605
// determine whether we care about this entry
606
final HashSet< String > interesting_fields =
607
new HashSet< String >( Arrays.asList( new String[] { "N",
608
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
610
boolean is_interesting_field =
611
interesting_fields.contains(
612
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
614
538
// parse encoding parameter
615
539
String encoding = checkParam( name_param_parts, "ENCODING" );
616
if( encoding != null )
617
encoding = encoding.toUpperCase( Locale.US );
618
if( is_interesting_field && encoding != null &&
619
!encoding.equalsIgnoreCase( "8BIT" ) &&
620
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
621
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
540
if( encoding != null ) encoding = encoding.toUpperCase();
541
if( encoding != null && !encoding.equals( "8BIT" ) &&
542
!encoding.equals( "QUOTED-PRINTABLE" ) )
543
//&& !encoding.equals( "BASE64" ) )
623
545
throw new ParseException( R.string.error_vcf_encoding );
626
548
// parse charset parameter
627
549
String charset = checkParam( name_param_parts, "CHARSET" );
628
if( charset != null )
629
charset = charset.toUpperCase( Locale.US );
630
if( charset != null &&
631
!charset.equalsIgnoreCase( "US-ASCII" ) &&
632
!charset.equalsIgnoreCase( "ASCII" ) &&
633
!charset.equalsIgnoreCase( "UTF-8" ) )
550
if( charset != null ) charset = charset.toUpperCase();
551
if( charset != null && !charset.equals( "US-ASCII" ) &&
552
!charset.equals( "ASCII" ) &&
553
!charset.equals( "UTF-8" ) )
635
555
throw new ParseException( R.string.error_vcf_charset );
638
558
// do unencoding (or default to a fake unencoding result with
639
559
// the raw string)
640
560
UnencodeResult unencoding_result = null;
641
if( encoding != null &&
642
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
561
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
644
562
unencoding_result = unencodeQuotedPrintable( value );
646
// else if( encoding != null &&
647
// encoding.equalsIgnoreCase( "BASE64" ) )
563
// else if( encoding != null && encoding.equals( "BASE64" ) )
649
564
// unencoding_result = unencodeBase64( props[ 1 ], charset );
651
565
if( unencoding_result != null ) {
652
566
value = unencoding_result.getBuffer();
653
567
if( unencoding_result.isAnotherLineRequired() )
654
568
_parser_multiline_state = MULTILINE_ENCODED;
657
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
658
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
659
if( ( charset == null && _version.equals( "2.1" ) ) ||
660
( charset != null && (
661
charset.equalsIgnoreCase( "ASCII" ) ||
662
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
571
// convert 8-bit ASCII charset to US-ASCII
572
if( charset == null || charset.equals( "ASCII" ) ) {
664
573
value = transcodeAsciiToUtf8( value );
667
// process charset (value is now in UTF-8)
668
578
String string_value;
670
580
string_value = new String( value.array(), value.position(),
671
value.limit() - value.position(), "UTF-8" );
581
value.limit() - value.position(), charset );
672
582
} catch( UnsupportedEncodingException e ) {
673
583
throw new ParseException( R.string.error_vcf_charset );
707
617
if( complete_value.length() < 1 ) return;
709
619
// parse some properties
710
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
620
if( name_param_parts[ 0 ].equals( "N" ) )
711
621
parseN( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
622
else if( name_param_parts[ 0 ].equals( "FN" ) )
713
623
parseFN( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
624
else if( name_param_parts[ 0 ].equals( "ORG" ) )
715
625
parseORG( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
626
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
717
627
parseTITLE( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
628
else if( name_param_parts[ 0 ].equals( "TEL" ) )
719
629
parseTEL( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
630
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
721
631
parseEMAIL( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
632
else if( name_param_parts[ 0 ].equals( "ADR" ) )
723
633
parseADR( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
725
parseLABEL( name_param_parts, complete_value );
726
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
727
parseNOTE( name_param_parts, complete_value );
728
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
729
parseBDAY( name_param_parts, complete_value );
745
649
return ( count & 1 ) == 1;
748
private String[] splitValueByCharacter( String value, char character )
652
private String[] splitValueBySemicolon( String value )
750
// split string in to parts by specified character
654
// split string in to parts by semicolon
751
655
ArrayList< String > parts = new ArrayList< String >(
752
Arrays.asList( value.split( "" + character ) ) );
656
Arrays.asList( value.split( ";" ) ) );
754
658
// go through parts
755
659
for( int a = 0; a < parts.size(); a++ )
757
661
String str = parts.get( a );
759
// Look for parts that end in an escape character, but ignore
760
// the final part. We've already detected escape chars at the
663
// look for parts that end in an escape character, but ignore
664
// the final part. We've already detected escape chars at the
761
665
// end of the final part in parseLine() and handled multi-lines
763
667
if( a < parts.size() - 1 &&
764
668
doesStringEndInAnEscapeChar( str ) )
766
// append the escaped character, join the next part to this
767
// part and remove the next part
670
// join the next part to this part and remove the next part
768
671
parts.set( a, str.substring( 0, str.length() - 1 ) +
769
character + parts.get( a + 1 ) );
672
';' + parts.get( a + 1 ) );
770
673
parts.remove( a + 1 );
772
675
// re-visit this part
783
686
return parts.toArray( ret );
786
private String unescapeValue( String value )
788
StringBuilder ret = new StringBuilder( value.length() );
789
boolean in_escape = false;
790
for( int a = 0; a < value.length(); a++ )
792
int c = value.codePointAt( a );
794
// process a normal character
799
ret.append( Character.toChars( c ) );
803
// process an escape sequence
809
// add tab (invalid/non-standard, but accepted)
820
// add escaped character
821
ret.append( Character.toChars( c ) );
824
// unknown escape sequence, so add it unescaped
825
// (invalid/non-standard, but accepted)
827
ret.append( Character.toChars( c ) );
832
return ret.toString();
835
689
private void parseN( String[] params, String value )
837
691
// already got a better name?
838
692
if( _name_level >= NAMELEVEL_N ) return;
840
694
// get name parts
841
String[] name_parts = splitValueByCharacter( value, ';' );
695
String[] name_parts = splitValueBySemicolon( value );
845
final int[] part_order = { 3, 1, 2, 0, 4 };
846
for( int a = 0; a < part_order.length; a++ )
847
if( name_parts.length > part_order[ a ] &&
848
name_parts[ part_order[ a ] ].length() > 0 )
850
// split this part in to it's comma-separated bits
851
String[] name_part_parts = splitValueByCharacter(
852
name_parts[ part_order[ a ] ], ',' );
853
for( int b = 0; b < name_part_parts.length; b++ )
854
if( name_part_parts[ b ].length() > 0 )
856
if( value.length() > 0 ) value += " ";
857
value += name_part_parts[ b ];
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 ];
862
setName( unescapeValue( value ) );
863
706
_name_level = NAMELEVEL_N;
961
802
boolean is_preferred = types.contains( "PREF" );
963
804
if( types.contains( "WORK" ) )
805
type = Contacts.ContactMethods.TYPE_WORK;
807
type = Contacts.ContactMethods.TYPE_HOME;
968
addEmail( unescapeValue( value ), type, is_preferred );
809
addEmail( value, type, is_preferred );
971
812
private void parseADR( String[] params, String value )
973
814
// get address parts
974
String[] adr_parts = splitValueByCharacter( value, ';' );
815
String[] adr_parts = splitValueBySemicolon( value );
978
for( int a = 0; a < adr_parts.length; a++ )
979
if( adr_parts[ a ].length() > 0 )
981
// version 3.0 vCards allow further splitting by comma
982
if( _version.equals( "3.0" ) )
984
// split this part in to it's comma-separated bits and
985
// add them on individual lines
986
String[] adr_part_parts =
987
splitValueByCharacter( adr_parts[ a ], ',' );
988
for( int b = 0; b < adr_part_parts.length; b++ )
989
if( adr_part_parts[ b ].length() > 0 )
991
if( value.length() > 0 ) value += "\n";
992
value += adr_part_parts[ b ];
997
// add this part on an individual line
998
if( value.length() > 0 ) value += "\n";
999
value += adr_parts[ a ];
1003
Set< String > types = extractTypes( params, Arrays.asList(
1004
"PREF", "WORK", "HOME" ) );
1008
if( types.contains( "WORK" ) )
1013
addAddress( unescapeValue( value ), type );
1016
private void parseLABEL( String[] params, String value )
1018
Set< String > types = extractTypes( params, Arrays.asList(
1019
"PREF", "WORK", "HOME" ) );
1023
if( types.contains( "WORK" ) )
1028
addAddress( unescapeValue( value ), type );
1031
private void parseNOTE( String[] params, String value )
1033
addNote( unescapeValue( value ) );
1036
private void parseBDAY( String[] params, String value )
1038
setBirthday( value );
1041
public void finaliseVcard()
1042
throws ParseException, ContactNotIdentifiableException
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 );
837
public void finaliseParsing()
838
throws ParseException, SkipContactException,
1044
841
// missing version (and data is present)
1045
if( _version == null && _content_lines != null )
842
if( _version == null && _buffers != null )
1046
843
throw new ParseException( R.string.error_vcf_malformed );
1048
// finalise the parent class
845
// check if we should import this contact
847
if( !isImportRequired( this ) )
848
throw new SkipContactException();
850
catch( ContactNeedsMoreInfoException e ) {
851
throw new ParseException( R.string.error_vcf_notenoughinfo );
1053
* Amongst the params, find the value of the first, only, of any with
1054
* the specified name.
1058
* @return a value, or null
1060
855
private String checkParam( String[] params, String name )
1062
String[] res = checkParams( params, name );
1063
return res.length > 0? res[ 0 ] : null;
1067
* Amongst the params, find the values of any with the specified name.
1071
* @return an array of values, or null
1073
private String[] checkParams( String[] params, String name )
1075
HashSet< String > ret = new HashSet< String >();
1077
857
Pattern p = Pattern.compile(
1078
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1079
Pattern.CASE_INSENSITIVE );
858
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1080
859
for( int i = 0; i < params.length; i++ ) {
1081
860
Matcher m = p.matcher( params[ i ] );
1082
861
if( m.matches() )
1083
ret.add( m.group( 2 ) );
1086
return (String[]) ret.toArray( new String[ ret.size() ] );
1090
* Amongst the params, return any type values present. For v2.1 vCards,
1091
* those types are just parameters. For v3.0, they are prefixed with
1092
* "TYPE=". There may also be multiple type parameters.
1094
* @param params an array of params to look for types in
1095
* @param valid_types an list of upper-case type values to look for
1096
* @return a set of present type values
1098
867
private Set< String > extractTypes( String[] params,
1099
868
List< String > valid_types )
1101
870
HashSet< String > types = new HashSet< String >();
1103
872
// get 3.0-style TYPE= param
1104
String type_params[] = checkParams( params, "TYPE" );
1105
for( int a = 0; a < type_params.length; a++ )
1107
// check for a comma-separated list of types (why? I don't think
1108
// this is in the specs!)
1109
String[] parts = type_params[ a ].split( "," );
1110
for( int i = 0; i < parts.length; i++ ) {
1111
String ucpart = parts[ i ].toUpperCase( Locale.US );
1112
if( valid_types.contains( ucpart ) )
1113
types.add( ucpart );
874
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
875
String[] parts = type_param.split( "," );
876
for( int i = 0; i < parts.length; i++ )
877
if( valid_types.contains( parts[ i ] ) )
878
types.add( parts[ i ] );
1117
881
// get 2.1-style type param
1118
882
if( _version.equals( "2.1" ) ) {
1119
for( int i = 1; i < params.length; i++ ) {
1120
String ucparam = params[ i ].toUpperCase( Locale.US );
1121
if( valid_types.contains( ucparam ) )
1122
types.add( ucparam );
883
for( int i = 1; i < params.length; i++ )
884
if( valid_types.contains( params[ i ] ) )
885
types.add( params[ i ] );