197
187
throws AbortImportException
199
189
// go through lines
201
int vcard_start_line = 0;
202
191
ContentLineIterator cli = new ContentLineIterator( content );
203
192
while( cli.hasNext() )
205
ContentLine content_line = cli.next();
207
// get a US-ASCII version of the string, for processing
208
String line = content_line.getUsAsciiLine();
210
if( vcard == null ) {
194
ByteBuffer buffer = cli.next();
196
// get a US-ASCII version of the line for processing
199
line = new String( buffer.array(), buffer.position(),
200
buffer.limit() - buffer.position(), "US-ASCII" );
202
catch( UnsupportedEncodingException e ) {
203
// we know US-ASCII is supported, so appease the compiler...
207
if( vCard == null ) {
211
208
// look for vcard beginning
212
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
213
setProgress( _progress++ );
215
vcard_start_line = cli.getLineNumber();
209
if( line.matches( "^BEGIN:VCARD" ) ) {
210
setProgress( ++_progress );
219
215
// look for vcard content or ending
220
if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
216
if( line.matches( "^END:VCARD" ) )
222
// finalise the vcard/contact
218
// store vcard and do away with it
224
vcard.finaliseVcard();
226
// pass the finalised contact to the importer
227
importContact( vcard );
229
catch( Vcard.ParseException e ) {
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
220
vCard.finaliseParsing();
221
importContact( vCard );
223
catch( VCard.ParseException e ) {
226
getText( R.string.error_vcf_parse ).toString()
227
+ fileName + "\n" + e.getMessage() ) )
229
finish( ACTION_ABORT );
232
catch( VCard.SkipContactException e ) {
261
240
// try giving the line to the vcard
263
vcard.parseLine( content_line );
242
vCard.parseLine( buffer, line,
243
cli.doesNextLineLookFolded() );
265
catch( Vcard.ParseException e ) {
245
catch( VCard.ParseException e ) {
267
247
if( !showContinue(
268
248
getText( R.string.error_vcf_parse ).toString()
270
getText( R.string.error_vcf_parse_line ).toString()
271
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
249
+ fileName + "\n" + e.getMessage() ) )
273
251
finish( ACTION_ABORT );
276
// Although we're continuing, we still need to abort
277
// this vCard. Further lines will be ignored until we
254
// although we're continuing, we still need to abort
255
// this vCard. Further lines will be ignored until we
278
256
// get to another BEGIN:VCARD line.
281
catch( Vcard.SkipImportException e ) {
259
catch( VCard.SkipContactException e ) {
283
// Abort this vCard. Further lines will be ignored until
261
// abort this vCard. Further lines will be ignored until
284
262
// we get to another BEGIN:VCARD line.
294
private ByteBuffer _buffer;
295
private boolean _folded_next;
296
private String _line;
298
public ContentLine( ByteBuffer buffer, boolean folded_next )
301
_folded_next = folded_next;
305
public ByteBuffer getBuffer()
310
public boolean doesNextLineLookFolded()
315
public String getUsAsciiLine()
317
// generated line and cache it
318
if( _line == null ) {
320
_line = new String( _buffer.array(), _buffer.position(),
321
_buffer.limit() - _buffer.position(), "US-ASCII" );
323
catch( UnsupportedEncodingException e ) {
324
// we know US-ASCII *is* supported, so appease the
329
// return cached line
334
class ContentLineIterator implements Iterator< ContentLine >
270
class ContentLineIterator implements Iterator< ByteBuffer >
336
272
protected byte[] _content = null;
337
273
protected int _pos = 0;
338
protected int _line = 0;
340
275
public ContentLineIterator( byte[] content )
394
323
* onto the end of this one?
397
private boolean doesNextLineLookFolded()
326
public boolean doesNextLineLookFolded()
399
328
return _pos > 0 && _pos < _content.length &&
400
_content[ _pos - 1 ] == '\n' &&
401
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
404
public int getLineNumber()
329
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
410
private class Vcard extends ContactData
333
private class VCard extends ContactData
412
335
private final static int NAMELEVEL_NONE = 0;
413
private final static int NAMELEVEL_N = 1;
414
private final static int NAMELEVEL_FN = 2;
336
private final static int NAMELEVEL_FN = 1;
337
private final static int NAMELEVEL_N = 2;
416
339
private final static int MULTILINE_NONE = 0;
417
340
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
418
341
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
419
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
342
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
421
344
private String _version = null;
422
private Vector< ContentLine > _content_lines = null;
345
private Vector< ByteBuffer > _buffers = null;
423
346
private int _name_level = NAMELEVEL_NONE;
424
347
private int _parser_multiline_state = MULTILINE_NONE;
425
348
private String _parser_current_name_and_params = null;
462
385
public ParseException( int res )
464
super( VcardImporter.this.getText( res ).toString() );
387
super( VCFImporter.this.getText( res ).toString() );
468
391
@SuppressWarnings("serial")
469
protected class SkipImportException extends Exception { }
392
protected class SkipContactException extends Exception { }
471
private String extractCollonPartFromLine( ContentLine content_line,
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...
474
412
// split line into name and value parts and check to make sure we
475
413
// only got 2 parts and that the first part is not zero in length
476
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
414
String[] parts = line.split( ":", 2 );
477
415
if( parts.length == 2 && parts[ 0 ].length() > 0 )
478
return parts[ former? 0 : 1 ].trim();
483
private String extractNameAndParamsFromLine( ContentLine content_line )
485
return extractCollonPartFromLine( content_line, true );
488
private String extractValueFromLine( ContentLine content_line )
490
return extractCollonPartFromLine( content_line, false );
493
public void parseLine( ContentLine content_line )
494
throws ParseException, SkipImportException,
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,
495
435
AbortImportException
497
437
// do we have a version yet?
500
440
// tentatively get name and params from line
501
441
String name_and_params =
502
extractNameAndParamsFromLine( content_line );
442
extractNameAndParamsFromLine( buffer, line );
504
444
// is it a version line?
505
445
if( name_and_params != null &&
506
name_and_params.equalsIgnoreCase( "VERSION" ) )
446
name_and_params.equals( "VERSION" ) )
509
String value = extractValueFromLine( content_line );
510
if( value == null || (
511
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
449
String value = extractValueFromLine( buffer, line );
450
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
451
throw new ParseException( R.string.error_vcf_version );
515
452
_version = value;
517
454
// parse any buffers we've been accumulating while we waited
519
if( _content_lines != null )
520
for( int i = 0; i < _content_lines.size(); i++ )
521
parseLine( _content_lines.get( i ) );
522
_content_lines = null;
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() ) == ' ' );
526
467
// no, so stash this line till we get a version
527
if( _content_lines == null )
528
_content_lines = new Vector< ContentLine >();
529
_content_lines.add( content_line );
468
if( _buffers == null )
469
_buffers = new Vector< ByteBuffer >();
470
_buffers.add( buffer );
534
475
// name and params and the position in the buffer where the
535
// "value" part of the line starts
476
// "value" part of the line start
536
477
String name_and_params;
594
530
// get value from buffer, as raw bytes
595
531
ByteBuffer value;
596
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
597
content_line.getBuffer().limit() - pos );
532
value = ByteBuffer.wrap( buffer.array(), pos,
533
buffer.limit() - pos );
599
535
// get parameter parts
600
536
String[] name_param_parts = name_and_params.split( ";", -1 );
601
537
for( int i = 0; i < name_param_parts.length; i++ )
602
538
name_param_parts[ i ] = name_param_parts[ i ].trim();
604
// determine whether we care about this entry
605
final HashSet< String > interesting_fields =
606
new HashSet< String >( Arrays.asList( new String[] { "N",
607
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
609
boolean is_interesting_field =
610
interesting_fields.contains(
611
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
613
540
// parse encoding parameter
614
541
String encoding = checkParam( name_param_parts, "ENCODING" );
615
if( encoding != null )
616
encoding = encoding.toUpperCase( Locale.US );
617
if( is_interesting_field && encoding != null &&
618
!encoding.equalsIgnoreCase( "8BIT" ) &&
619
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
620
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
542
if( encoding != null ) encoding = encoding.toUpperCase();
543
if( encoding != null && !encoding.equals( "8BIT" ) &&
544
!encoding.equals( "QUOTED-PRINTABLE" ) )
545
//&& !encoding.equals( "BASE64" ) )
622
547
throw new ParseException( R.string.error_vcf_encoding );
625
550
// parse charset parameter
626
551
String charset = checkParam( name_param_parts, "CHARSET" );
627
if( charset != null )
628
charset = charset.toUpperCase( Locale.US );
629
if( charset != null &&
630
!charset.equalsIgnoreCase( "US-ASCII" ) &&
631
!charset.equalsIgnoreCase( "ASCII" ) &&
632
!charset.equalsIgnoreCase( "UTF-8" ) )
552
if( charset != null ) charset = charset.toUpperCase();
553
if( charset != null && !charset.equals( "US-ASCII" ) &&
554
!charset.equals( "ASCII" ) &&
555
!charset.equals( "UTF-8" ) )
634
557
throw new ParseException( R.string.error_vcf_charset );
637
560
// do unencoding (or default to a fake unencoding result with
638
561
// the raw string)
639
562
UnencodeResult unencoding_result = null;
640
if( encoding != null &&
641
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
563
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
643
564
unencoding_result = unencodeQuotedPrintable( value );
645
// else if( encoding != null &&
646
// encoding.equalsIgnoreCase( "BASE64" ) )
565
// else if( encoding != null && encoding.equals( "BASE64" ) )
648
566
// unencoding_result = unencodeBase64( props[ 1 ], charset );
650
567
if( unencoding_result != null ) {
651
568
value = unencoding_result.getBuffer();
652
569
if( unencoding_result.isAnotherLineRequired() )
653
570
_parser_multiline_state = MULTILINE_ENCODED;
656
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
657
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
658
if( ( charset == null && _version.equals( "2.1" ) ) ||
659
( charset != null && (
660
charset.equalsIgnoreCase( "ASCII" ) ||
661
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
573
// convert 8-bit ASCII charset to US-ASCII
574
if( charset == null || charset.equals( "ASCII" ) ) {
663
575
value = transcodeAsciiToUtf8( value );
666
// process charset (value is now in UTF-8)
667
580
String string_value;
669
582
string_value = new String( value.array(), value.position(),
670
value.limit() - value.position(), "UTF-8" );
583
value.limit() - value.position(), charset );
671
584
} catch( UnsupportedEncodingException e ) {
672
585
throw new ParseException( R.string.error_vcf_charset );
706
619
if( complete_value.length() < 1 ) return;
708
621
// parse some properties
709
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
622
if( name_param_parts[ 0 ].equals( "N" ) )
710
623
parseN( name_param_parts, complete_value );
711
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
624
else if( name_param_parts[ 0 ].equals( "FN" ) )
712
625
parseFN( name_param_parts, complete_value );
713
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
626
else if( name_param_parts[ 0 ].equals( "ORG" ) )
714
627
parseORG( name_param_parts, complete_value );
715
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
628
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
716
629
parseTITLE( name_param_parts, complete_value );
717
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
630
else if( name_param_parts[ 0 ].equals( "TEL" ) )
718
631
parseTEL( name_param_parts, complete_value );
719
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
632
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
720
633
parseEMAIL( name_param_parts, complete_value );
721
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
634
else if( name_param_parts[ 0 ].equals( "ADR" ) )
722
635
parseADR( name_param_parts, complete_value );
723
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
724
parseLABEL( name_param_parts, complete_value );
725
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
726
parseNOTE( name_param_parts, complete_value );
727
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
728
parseBDAY( name_param_parts, complete_value );
744
651
return ( count & 1 ) == 1;
747
private String[] splitValueByCharacter( String value, char character )
654
private String[] splitValueBySemicolon( String value )
749
// split string in to parts by specified character
656
// split string in to parts by semicolon
750
657
ArrayList< String > parts = new ArrayList< String >(
751
Arrays.asList( value.split( "" + character ) ) );
658
Arrays.asList( value.split( ";" ) ) );
753
660
// go through parts
754
661
for( int a = 0; a < parts.size(); a++ )
756
663
String str = parts.get( a );
758
// Look for parts that end in an escape character, but ignore
759
// the final part. We've already detected escape chars at the
665
// look for parts that end in an escape character, but ignore
666
// the final part. We've already detected escape chars at the
760
667
// end of the final part in parseLine() and handled multi-lines
762
669
if( a < parts.size() - 1 &&
763
670
doesStringEndInAnEscapeChar( str ) )
765
// append the escaped character, join the next part to this
766
// part and remove the next part
672
// join the next part to this part and remove the next part
767
673
parts.set( a, str.substring( 0, str.length() - 1 ) +
768
character + parts.get( a + 1 ) );
674
';' + parts.get( a + 1 ) );
769
675
parts.remove( a + 1 );
771
677
// re-visit this part
782
688
return parts.toArray( ret );
785
private String unescapeValue( String value )
787
StringBuilder ret = new StringBuilder( value.length() );
788
boolean in_escape = false;
789
for( int a = 0; a < value.length(); a++ )
791
int c = value.codePointAt( a );
793
// process a normal character
798
ret.append( Character.toChars( c ) );
802
// process an escape sequence
808
// add tab (invalid/non-standard, but accepted)
819
// add escaped character
820
ret.append( Character.toChars( c ) );
823
// unknown escape sequence, so add it unescaped
824
// (invalid/non-standard, but accepted)
826
ret.append( Character.toChars( c ) );
831
return ret.toString();
834
691
private void parseN( String[] params, String value )
836
693
// already got a better name?
837
694
if( _name_level >= NAMELEVEL_N ) return;
839
696
// get name parts
840
String[] name_parts = splitValueByCharacter( value, ';' );
697
String[] name_parts = splitValueBySemicolon( value );
844
final int[] part_order = { 3, 1, 2, 0, 4 };
845
for( int a = 0; a < part_order.length; a++ )
846
if( name_parts.length > part_order[ a ] &&
847
name_parts[ part_order[ a ] ].length() > 0 )
849
// split this part in to it's comma-separated bits
850
String[] name_part_parts = splitValueByCharacter(
851
name_parts[ part_order[ a ] ], ',' );
852
for( int b = 0; b < name_part_parts.length; b++ )
853
if( name_part_parts[ b ].length() > 0 )
855
if( value.length() > 0 ) value += " ";
856
value += name_part_parts[ b ];
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 ];
861
setName( unescapeValue( value ) );
862
708
_name_level = NAMELEVEL_N;
957
801
"PREF", "WORK", "HOME", "INTERNET" ) );
959
803
// add email address
960
boolean is_preferred = types.contains( "PREF" );
804
boolean preferred = types.contains( "PREF" );
962
806
if( types.contains( "WORK" ) )
807
type = Contacts.ContactMethods.TYPE_WORK;
809
type = Contacts.ContactMethods.TYPE_HOME;
967
addEmail( unescapeValue( value ), type, is_preferred );
811
addEmail( value, type, preferred );
970
814
private void parseADR( String[] params, String value )
972
816
// get address parts
973
String[] adr_parts = splitValueByCharacter( value, ';' );
817
String[] adr_parts = splitValueBySemicolon( value );
977
for( int a = 0; a < adr_parts.length; a++ )
978
if( adr_parts[ a ].length() > 0 )
980
// version 3.0 vCards allow further splitting by comma
981
if( _version.equals( "3.0" ) )
983
// split this part in to it's comma-separated bits and
984
// add them on individual lines
985
String[] adr_part_parts =
986
splitValueByCharacter( adr_parts[ a ], ',' );
987
for( int b = 0; b < adr_part_parts.length; b++ )
988
if( adr_part_parts[ b ].length() > 0 )
990
if( value.length() > 0 ) value += "\n";
991
value += adr_part_parts[ b ];
996
// add this part on an individual line
997
if( value.length() > 0 ) value += "\n";
998
value += adr_parts[ a ];
1002
Set< String > types = extractTypes( params, Arrays.asList(
1003
"PREF", "WORK", "HOME" ) );
1007
if( types.contains( "WORK" ) )
1012
addAddress( unescapeValue( value ), type );
1015
private void parseLABEL( String[] params, String value )
1017
Set< String > types = extractTypes( params, Arrays.asList(
1018
"PREF", "WORK", "HOME" ) );
1022
if( types.contains( "WORK" ) )
1027
addAddress( unescapeValue( value ), type );
1030
private void parseNOTE( String[] params, String value )
1032
addNote( unescapeValue( value ) );
1035
private void parseBDAY( String[] params, String value )
1037
setBirthday( value );
1040
public void finaliseVcard()
1041
throws ParseException, ContactNotIdentifiableException
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 );
839
public void finaliseParsing()
840
throws ParseException, SkipContactException,
1043
843
// missing version (and data is present)
1044
if( _version == null && _content_lines != null )
844
if( _version == null && _buffers != null )
1045
845
throw new ParseException( R.string.error_vcf_malformed );
1047
// finalise the parent class
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 );
1052
* Amongst the params, find the value of the first, only, of any with
1053
* the specified name.
1057
* @return a value, or null
1059
857
private String checkParam( String[] params, String name )
1061
String[] res = checkParams( params, name );
1062
return res.length > 0? res[ 0 ] : null;
1066
* Amongst the params, find the values of any with the specified name.
1070
* @return an array of values, or null
1072
private String[] checkParams( String[] params, String name )
1074
HashSet< String > ret = new HashSet< String >();
1076
859
Pattern p = Pattern.compile(
1077
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1078
Pattern.CASE_INSENSITIVE );
860
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1079
861
for( int i = 0; i < params.length; i++ ) {
1080
862
Matcher m = p.matcher( params[ i ] );
1081
863
if( m.matches() )
1082
ret.add( m.group( 2 ) );
1085
return (String[]) ret.toArray( new String[ ret.size() ] );
1089
* Amongst the params, return any type values present. For v2.1 vCards,
1090
* those types are just parameters. For v3.0, they are prefixed with
1091
* "TYPE=". There may also be multiple type parameters.
1093
* @param params an array of params to look for types in
1094
* @param valid_types an list of upper-case type values to look for
1095
* @return a set of present type values
1097
869
private Set< String > extractTypes( String[] params,
1098
870
List< String > valid_types )
1100
872
HashSet< String > types = new HashSet< String >();
1102
874
// get 3.0-style TYPE= param
1103
String type_params[] = checkParams( params, "TYPE" );
1104
for( int a = 0; a < type_params.length; a++ )
1106
// check for a comma-separated list of types (why? I don't think
1107
// this is in the specs!)
1108
String[] parts = type_params[ a ].split( "," );
1109
for( int i = 0; i < parts.length; i++ ) {
1110
String ucpart = parts[ i ].toUpperCase( Locale.US );
1111
if( valid_types.contains( ucpart ) )
1112
types.add( ucpart );
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 ] );
1116
883
// get 2.1-style type param
1117
884
if( _version.equals( "2.1" ) ) {
1118
for( int i = 1; i < params.length; i++ ) {
1119
String ucparam = params[ i ].toUpperCase( Locale.US );
1120
if( valid_types.contains( ucparam ) )
1121
types.add( ucparam );
885
for( int i = 1; i < params.length; i++ )
886
if( valid_types.contains( params[ i ] ) )
887
types.add( params[ i ] );