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 )
474
397
String ret = null;
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...
476
412
// split line into name and value parts and check to make sure we
477
413
// only got 2 parts and that the first part is not zero in length
478
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
414
String[] parts = line.split( ":", 2 );
479
415
if( parts.length == 2 && parts[ 0 ].length() > 0 )
480
416
ret = parts[ former? 0 : 1 ];
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 )
496
throws ParseException, SkipImportException,
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,
497
435
AbortImportException
499
437
// do we have a version yet?
502
440
// tentatively get name and params from line
503
441
String name_and_params =
504
extractNameAndParamsFromLine( content_line );
442
extractNameAndParamsFromLine( buffer, line );
506
444
// is it a version line?
507
445
if( name_and_params != null &&
508
name_and_params.equalsIgnoreCase( "VERSION" ) )
446
name_and_params.equals( "VERSION" ) )
511
String value = extractValueFromLine( content_line ).trim();
449
String value = extractValueFromLine( buffer, line );
512
450
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
513
451
throw new ParseException( R.string.error_vcf_version );
514
452
_version = value;
516
454
// parse any buffers we've been accumulating while we waited
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;
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() ) == ' ' );
525
467
// no, so stash this line till we get a version
526
if( _content_lines == null )
527
_content_lines = new Vector< ContentLine >();
528
_content_lines.add( content_line );
468
if( _buffers == null )
469
_buffers = new Vector< ByteBuffer >();
470
_buffers.add( buffer );
533
475
// name and params and the position in the buffer where the
534
// "value" part of the line starts
476
// "value" part of the line start
535
477
String name_and_params;
593
530
// get value from buffer, as raw bytes
594
531
ByteBuffer value;
595
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
596
content_line.getBuffer().limit() - pos );
532
value = ByteBuffer.wrap( buffer.array(), pos,
533
buffer.limit() - pos );
598
535
// get parameter parts
599
536
String[] name_param_parts = name_and_params.split( ";", -1 );
600
537
for( int i = 0; i < name_param_parts.length; i++ )
601
538
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 ) );
612
540
// parse encoding parameter
613
541
String encoding = checkParam( name_param_parts, "ENCODING" );
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" ) )
542
if( encoding != null ) encoding = encoding.toUpperCase();
543
if( encoding != null && !encoding.equals( "8BIT" ) &&
544
!encoding.equals( "QUOTED-PRINTABLE" ) )
545
//&& !encoding.equals( "BASE64" ) )
621
547
throw new ParseException( R.string.error_vcf_encoding );
624
550
// parse charset parameter
625
551
String charset = checkParam( name_param_parts, "CHARSET" );
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" ) )
552
if( charset != null ) charset = charset.toUpperCase();
553
if( charset != null && !charset.equals( "US-ASCII" ) &&
554
!charset.equals( "ASCII" ) &&
555
!charset.equals( "UTF-8" ) )
633
557
throw new ParseException( R.string.error_vcf_charset );
636
560
// do unencoding (or default to a fake unencoding result with
637
561
// the raw string)
638
562
UnencodeResult unencoding_result = null;
639
if( encoding != null &&
640
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
563
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
642
564
unencoding_result = unencodeQuotedPrintable( value );
644
// else if( encoding != null &&
645
// encoding.equalsIgnoreCase( "BASE64" ) )
565
// else if( encoding != null && encoding.equals( "BASE64" ) )
647
566
// unencoding_result = unencodeBase64( props[ 1 ], charset );
649
567
if( unencoding_result != null ) {
650
568
value = unencoding_result.getBuffer();
651
569
if( unencoding_result.isAnotherLineRequired() )
652
570
_parser_multiline_state = MULTILINE_ENCODED;
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
// convert 8-bit ASCII charset to US-ASCII
574
if( charset == null || charset.equals( "ASCII" ) ) {
662
575
value = transcodeAsciiToUtf8( value );
665
// process charset (value is now in UTF-8)
666
580
String string_value;
668
582
string_value = new String( value.array(), value.position(),
669
value.limit() - value.position(), "UTF-8" );
583
value.limit() - value.position(), charset );
670
584
} catch( UnsupportedEncodingException e ) {
671
585
throw new ParseException( R.string.error_vcf_charset );
705
619
if( complete_value.length() < 1 ) return;
707
621
// parse some properties
708
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
622
if( name_param_parts[ 0 ].equals( "N" ) )
709
623
parseN( name_param_parts, complete_value );
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
624
else if( name_param_parts[ 0 ].equals( "FN" ) )
711
625
parseFN( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
626
else if( name_param_parts[ 0 ].equals( "ORG" ) )
713
627
parseORG( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
628
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
715
629
parseTITLE( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
630
else if( name_param_parts[ 0 ].equals( "TEL" ) )
717
631
parseTEL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
632
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
719
633
parseEMAIL( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
634
else if( name_param_parts[ 0 ].equals( "ADR" ) )
721
635
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 );
741
651
return ( count & 1 ) == 1;
744
private String[] splitValueByCharacter( String value, char character )
654
private String[] splitValueBySemicolon( String value )
746
// split string in to parts by specified character
656
// split string in to parts by semicolon
747
657
ArrayList< String > parts = new ArrayList< String >(
748
Arrays.asList( value.split( "" + character ) ) );
658
Arrays.asList( value.split( ";" ) ) );
750
660
// go through parts
751
661
for( int a = 0; a < parts.size(); a++ )
753
663
String str = parts.get( a );
755
// Look for parts that end in an escape character, but ignore
756
// 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
757
667
// end of the final part in parseLine() and handled multi-lines
759
669
if( a < parts.size() - 1 &&
760
670
doesStringEndInAnEscapeChar( str ) )
762
// append the escaped character, join the next part to this
763
// part and remove the next part
672
// join the next part to this part and remove the next part
764
673
parts.set( a, str.substring( 0, str.length() - 1 ) +
765
character + parts.get( a + 1 ) );
674
';' + parts.get( a + 1 ) );
766
675
parts.remove( a + 1 );
768
677
// re-visit this part
779
688
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();
831
691
private void parseN( String[] params, String value )
833
693
// already got a better name?
834
694
if( _name_level >= NAMELEVEL_N ) return;
836
696
// get name parts
837
String[] name_parts = splitValueByCharacter( value, ';' );
697
String[] name_parts = splitValueBySemicolon( value );
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 ];
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 ];
858
setName( unescapeValue( value ) );
859
708
_name_level = NAMELEVEL_N;
954
801
"PREF", "WORK", "HOME", "INTERNET" ) );
956
803
// add email address
957
boolean is_preferred = types.contains( "PREF" );
804
boolean preferred = types.contains( "PREF" );
959
806
if( types.contains( "WORK" ) )
807
type = Contacts.ContactMethods.TYPE_WORK;
809
type = Contacts.ContactMethods.TYPE_HOME;
964
addEmail( unescapeValue( value ), type, is_preferred );
811
addEmail( value, type, preferred );
967
814
private void parseADR( String[] params, String value )
969
816
// get address parts
970
String[] adr_parts = splitValueByCharacter( value, ';' );
817
String[] adr_parts = splitValueBySemicolon( value );
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 ) );
1032
public void finaliseVcard()
1033
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,
1035
843
// missing version (and data is present)
1036
if( _version == null && _content_lines != null )
844
if( _version == null && _buffers != null )
1037
845
throw new ParseException( R.string.error_vcf_malformed );
1039
// 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 );
1044
* Amongst the params, find the value of the first, only, of any with
1045
* the specified name.
1049
* @return a value, or null
1051
857
private String checkParam( String[] params, String name )
1053
String[] res = checkParams( params, name );
1054
return res.length > 0? res[ 0 ] : null;
1058
* Amongst the params, find the values of any with the specified name.
1062
* @return an array of values, or null
1064
private String[] checkParams( String[] params, String name )
1066
HashSet< String > ret = new HashSet< String >();
1068
859
Pattern p = Pattern.compile(
1069
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1070
Pattern.CASE_INSENSITIVE );
860
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1071
861
for( int i = 0; i < params.length; i++ ) {
1072
862
Matcher m = p.matcher( params[ i ] );
1073
863
if( m.matches() )
1074
ret.add( m.group( 2 ) );
1077
return (String[]) ret.toArray( new String[ ret.size() ] );
1081
* Amongst the params, return any type values present. For v2.1 vCards,
1082
* those types are just parameters. For v3.0, they are prefixed with
1083
* "TYPE=". There may also be multiple type parameters.
1085
* @param params an array of params to look for types in
1086
* @param valid_types an list of upper-case type values to look for
1087
* @return a set of present type values
1089
869
private Set< String > extractTypes( String[] params,
1090
870
List< String > valid_types )
1092
872
HashSet< String > types = new HashSet< String >();
1094
874
// get 3.0-style TYPE= param
1095
String type_params[] = checkParams( params, "TYPE" );
1096
for( int a = 0; a < type_params.length; a++ )
1098
// check for a comma-separated list of types (why? I don't think
1099
// this is in the specs!)
1100
String[] parts = type_params[ a ].split( "," );
1101
for( int i = 0; i < parts.length; i++ ) {
1102
String ucpart = parts[ i ].toUpperCase( Locale.US );
1103
if( valid_types.contains( ucpart ) )
1104
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 ] );
1108
883
// get 2.1-style type param
1109
884
if( _version.equals( "2.1" ) ) {
1110
for( int i = 1; i < params.length; i++ ) {
1111
String ucparam = params[ i ].toUpperCase( Locale.US );
1112
if( valid_types.contains( ucparam ) )
1113
types.add( ucparam );
885
for( int i = 1; i < params.length; i++ )
886
if( valid_types.contains( params[ i ] ) )
887
types.add( params[ i ] );