205
187
throws AbortImportException
207
189
// go through lines
209
int vcard_start_line = 0;
210
191
ContentLineIterator cli = new ContentLineIterator( content );
211
192
while( cli.hasNext() )
213
ContentLine content_line = cli.next();
215
// get a US-ASCII version of the string, for processing
216
String line = content_line.getUsAsciiLine();
218
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 ) {
219
208
// look for vcard beginning
220
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
221
setProgress( _progress++ );
223
vcard_start_line = cli.getLineNumber();
209
if( line.matches( "^BEGIN:VCARD" ) ) {
210
setProgress( ++_progress );
227
215
// look for vcard content or ending
228
if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
216
if( line.matches( "^END:VCARD" ) )
230
// finalise the vcard/contact
218
// store vcard and do away with it
232
vcard.finaliseVcard();
234
// pass the finalised contact to the importer
235
importContact( vcard );
237
catch( Vcard.ParseException e ) {
239
getText( R.string.error_vcf_parse ).toString()
241
getText( R.string.error_vcf_parse_line ).toString()
242
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
244
finish( ACTION_ABORT );
249
catch( ContactData.ContactNotIdentifiableException e ) {
251
getText( R.string.error_vcf_parse ).toString()
253
getText( R.string.error_vcf_parse_line ).toString()
254
+ vcard_start_line + ":\n" + getText(
255
R.string.error_vcf_notenoughinfo ).toString()
258
finish( ACTION_ABORT );
264
// 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 ) {
269
240
// try giving the line to the vcard
271
vcard.parseLine( content_line );
242
vCard.parseLine( buffer, line,
243
cli.doesNextLineLookFolded() );
273
catch( Vcard.ParseException e ) {
245
catch( VCard.ParseException e ) {
275
247
if( !showContinue(
276
248
getText( R.string.error_vcf_parse ).toString()
278
getText( R.string.error_vcf_parse_line ).toString()
279
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
249
+ fileName + "\n" + e.getMessage() ) )
281
251
finish( ACTION_ABORT );
284
// Although we're continuing, we still need to abort
285
// 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
286
256
// get to another BEGIN:VCARD line.
289
catch( Vcard.SkipImportException e ) {
259
catch( VCard.SkipContactException e ) {
291
// Abort this vCard. Further lines will be ignored until
261
// abort this vCard. Further lines will be ignored until
292
262
// we get to another BEGIN:VCARD line.
302
private ByteBuffer _buffer;
303
private boolean _folded_next;
304
private String _line;
306
public ContentLine( ByteBuffer buffer, boolean folded_next )
309
_folded_next = folded_next;
313
public ByteBuffer getBuffer()
318
public boolean doesNextLineLookFolded()
323
public String getUsAsciiLine()
325
// generated line and cache it
326
if( _line == null ) {
328
_line = new String( _buffer.array(), _buffer.position(),
329
_buffer.limit() - _buffer.position(), "US-ASCII" );
331
catch( UnsupportedEncodingException e ) {
332
// we know US-ASCII *is* supported, so appease the
337
// return cached line
342
class ContentLineIterator implements Iterator< ContentLine >
270
class ContentLineIterator implements Iterator< ByteBuffer >
344
272
protected byte[] _content = null;
345
273
protected int _pos = 0;
346
protected int _line = 0;
348
275
public ContentLineIterator( byte[] content )
402
323
* onto the end of this one?
405
private boolean doesNextLineLookFolded()
326
public boolean doesNextLineLookFolded()
407
328
return _pos > 0 && _pos < _content.length &&
408
_content[ _pos - 1 ] == '\n' &&
409
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
412
public int getLineNumber()
329
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
418
private class Vcard extends ContactData
333
private class VCard extends ContactData
420
335
private final static int NAMELEVEL_NONE = 0;
421
private final static int NAMELEVEL_N = 1;
422
private final static int NAMELEVEL_FN = 2;
336
private final static int NAMELEVEL_FN = 1;
337
private final static int NAMELEVEL_N = 2;
424
339
private final static int MULTILINE_NONE = 0;
425
340
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
426
341
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
427
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
342
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
429
344
private String _version = null;
430
private Vector< ContentLine > _content_lines = null;
345
private Vector< ByteBuffer > _buffers = null;
431
346
private int _name_level = NAMELEVEL_NONE;
432
347
private int _parser_multiline_state = MULTILINE_NONE;
433
348
private String _parser_current_name_and_params = null;
470
385
public ParseException( int res )
472
super( VcardImporter.this.getText( res ).toString() );
387
super( VCFImporter.this.getText( res ).toString() );
476
391
@SuppressWarnings("serial")
477
protected class SkipImportException extends Exception { }
392
protected class SkipContactException extends Exception { }
479
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...
482
412
// split line into name and value parts and check to make sure we
483
413
// only got 2 parts and that the first part is not zero in length
484
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
414
String[] parts = line.split( ":", 2 );
485
415
if( parts.length == 2 && parts[ 0 ].length() > 0 )
486
return parts[ former? 0 : 1 ].trim();
491
private String extractNameAndParamsFromLine( ContentLine content_line )
493
return extractCollonPartFromLine( content_line, true );
496
private String extractValueFromLine( ContentLine content_line )
498
return extractCollonPartFromLine( content_line, false );
501
public void parseLine( ContentLine content_line )
502
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,
503
435
AbortImportException
505
437
// do we have a version yet?
508
440
// tentatively get name and params from line
509
441
String name_and_params =
510
extractNameAndParamsFromLine( content_line );
442
extractNameAndParamsFromLine( buffer, line );
512
444
// is it a version line?
513
445
if( name_and_params != null &&
514
name_and_params.equalsIgnoreCase( "VERSION" ) )
446
name_and_params.equals( "VERSION" ) )
517
String value = extractValueFromLine( content_line );
518
if( value == null || (
519
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
449
String value = extractValueFromLine( buffer, line );
450
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
521
451
throw new ParseException( R.string.error_vcf_version );
523
452
_version = value;
525
454
// parse any buffers we've been accumulating while we waited
527
if( _content_lines != null )
528
for( int i = 0; i < _content_lines.size(); i++ )
529
parseLine( _content_lines.get( i ) );
530
_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() ) == ' ' );
534
467
// no, so stash this line till we get a version
535
if( _content_lines == null )
536
_content_lines = new Vector< ContentLine >();
537
_content_lines.add( content_line );
468
if( _buffers == null )
469
_buffers = new Vector< ByteBuffer >();
470
_buffers.add( buffer );
542
475
// name and params and the position in the buffer where the
543
// "value" part of the line starts
476
// "value" part of the line start
544
477
String name_and_params;
602
530
// get value from buffer, as raw bytes
603
531
ByteBuffer value;
604
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
605
content_line.getBuffer().limit() - pos );
532
value = ByteBuffer.wrap( buffer.array(), pos,
533
buffer.limit() - pos );
607
535
// get parameter parts
608
536
String[] name_param_parts = name_and_params.split( ";", -1 );
609
537
for( int i = 0; i < name_param_parts.length; i++ )
610
538
name_param_parts[ i ] = name_param_parts[ i ].trim();
612
// determine whether we care about this entry
613
final HashSet< String > interesting_fields =
614
new HashSet< String >( Arrays.asList( new String[] { "N",
615
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
617
boolean is_interesting_field =
618
interesting_fields.contains(
619
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
621
540
// parse encoding parameter
622
541
String encoding = checkParam( name_param_parts, "ENCODING" );
623
if( encoding != null )
624
encoding = encoding.toUpperCase( Locale.US );
625
if( is_interesting_field && encoding != null &&
626
!encoding.equalsIgnoreCase( "8BIT" ) &&
627
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
628
//&& !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" ) )
630
547
throw new ParseException( R.string.error_vcf_encoding );
633
550
// parse charset parameter
634
551
String charset = checkParam( name_param_parts, "CHARSET" );
635
if( charset != null )
636
charset = charset.toUpperCase( Locale.US );
637
if( charset != null &&
638
!charset.equalsIgnoreCase( "US-ASCII" ) &&
639
!charset.equalsIgnoreCase( "ASCII" ) &&
640
!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" ) )
642
557
throw new ParseException( R.string.error_vcf_charset );
645
560
// do unencoding (or default to a fake unencoding result with
646
561
// the raw string)
647
562
UnencodeResult unencoding_result = null;
648
if( encoding != null &&
649
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
563
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
651
564
unencoding_result = unencodeQuotedPrintable( value );
653
// else if( encoding != null &&
654
// encoding.equalsIgnoreCase( "BASE64" ) )
565
// else if( encoding != null && encoding.equals( "BASE64" ) )
656
566
// unencoding_result = unencodeBase64( props[ 1 ], charset );
658
567
if( unencoding_result != null ) {
659
568
value = unencoding_result.getBuffer();
660
569
if( unencoding_result.isAnotherLineRequired() )
661
570
_parser_multiline_state = MULTILINE_ENCODED;
664
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
665
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
666
if( ( charset == null && _version.equals( "2.1" ) ) ||
667
( charset != null && (
668
charset.equalsIgnoreCase( "ASCII" ) ||
669
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
573
// convert 8-bit ASCII charset to US-ASCII
574
if( charset == null || charset.equals( "ASCII" ) ) {
671
575
value = transcodeAsciiToUtf8( value );
674
// process charset (value is now in UTF-8)
675
580
String string_value;
677
582
string_value = new String( value.array(), value.position(),
678
value.limit() - value.position(), "UTF-8" );
583
value.limit() - value.position(), charset );
679
584
} catch( UnsupportedEncodingException e ) {
680
585
throw new ParseException( R.string.error_vcf_charset );
714
619
if( complete_value.length() < 1 ) return;
716
621
// parse some properties
717
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
622
if( name_param_parts[ 0 ].equals( "N" ) )
718
623
parseN( name_param_parts, complete_value );
719
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
624
else if( name_param_parts[ 0 ].equals( "FN" ) )
720
625
parseFN( name_param_parts, complete_value );
721
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
626
else if( name_param_parts[ 0 ].equals( "ORG" ) )
722
627
parseORG( name_param_parts, complete_value );
723
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
628
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
724
629
parseTITLE( name_param_parts, complete_value );
725
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
630
else if( name_param_parts[ 0 ].equals( "TEL" ) )
726
631
parseTEL( name_param_parts, complete_value );
727
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
632
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
728
633
parseEMAIL( name_param_parts, complete_value );
729
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
634
else if( name_param_parts[ 0 ].equals( "ADR" ) )
730
635
parseADR( name_param_parts, complete_value );
731
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
732
parseLABEL( name_param_parts, complete_value );
733
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
734
parseNOTE( name_param_parts, complete_value );
735
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
736
parseBDAY( name_param_parts, complete_value );
752
651
return ( count & 1 ) == 1;
755
private String[] splitValueByCharacter( String value, char character )
654
private String[] splitValueBySemicolon( String value )
757
// split string in to parts by specified character
656
// split string in to parts by semicolon
758
657
ArrayList< String > parts = new ArrayList< String >(
759
Arrays.asList( value.split( "" + character ) ) );
658
Arrays.asList( value.split( ";" ) ) );
761
660
// go through parts
762
661
for( int a = 0; a < parts.size(); a++ )
764
663
String str = parts.get( a );
766
// Look for parts that end in an escape character, but ignore
767
// 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
768
667
// end of the final part in parseLine() and handled multi-lines
770
669
if( a < parts.size() - 1 &&
771
670
doesStringEndInAnEscapeChar( str ) )
773
// append the escaped character, join the next part to this
774
// part and remove the next part
672
// join the next part to this part and remove the next part
775
673
parts.set( a, str.substring( 0, str.length() - 1 ) +
776
character + parts.get( a + 1 ) );
674
';' + parts.get( a + 1 ) );
777
675
parts.remove( a + 1 );
779
677
// re-visit this part
790
688
return parts.toArray( ret );
793
private String unescapeValue( String value )
795
StringBuilder ret = new StringBuilder( value.length() );
796
boolean in_escape = false;
797
for( int a = 0; a < value.length(); a++ )
799
int c = value.codePointAt( a );
801
// process a normal character
806
ret.append( Character.toChars( c ) );
810
// process an escape sequence
816
// add tab (invalid/non-standard, but accepted)
827
// add escaped character
828
ret.append( Character.toChars( c ) );
831
// unknown escape sequence, so add it unescaped
832
// (invalid/non-standard, but accepted)
834
ret.append( Character.toChars( c ) );
839
return ret.toString();
842
691
private void parseN( String[] params, String value )
844
693
// already got a better name?
845
694
if( _name_level >= NAMELEVEL_N ) return;
847
696
// get name parts
848
String[] name_parts = splitValueByCharacter( value, ';' );
697
String[] name_parts = splitValueBySemicolon( value );
852
final int[] part_order = { 3, 1, 2, 0, 4 };
853
for( int a = 0; a < part_order.length; a++ )
854
if( name_parts.length > part_order[ a ] &&
855
name_parts[ part_order[ a ] ].length() > 0 )
857
// split this part in to it's comma-separated bits
858
String[] name_part_parts = splitValueByCharacter(
859
name_parts[ part_order[ a ] ], ',' );
860
for( int b = 0; b < name_part_parts.length; b++ )
861
if( name_part_parts[ b ].length() > 0 )
863
if( value.length() > 0 ) value += " ";
864
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 ];
869
setName( unescapeValue( value ) );
870
708
_name_level = NAMELEVEL_N;
965
801
"PREF", "WORK", "HOME", "INTERNET" ) );
967
803
// add email address
968
boolean is_preferred = types.contains( "PREF" );
804
boolean preferred = types.contains( "PREF" );
970
806
if( types.contains( "WORK" ) )
807
type = Contacts.ContactMethods.TYPE_WORK;
809
type = Contacts.ContactMethods.TYPE_HOME;
975
addEmail( unescapeValue( value ), type, is_preferred );
811
addEmail( value, type, preferred );
978
814
private void parseADR( String[] params, String value )
980
816
// get address parts
981
String[] adr_parts = splitValueByCharacter( value, ';' );
817
String[] adr_parts = splitValueBySemicolon( value );
985
for( int a = 0; a < adr_parts.length; a++ )
986
if( adr_parts[ a ].length() > 0 )
988
// version 3.0 vCards allow further splitting by comma
989
if( _version.equals( "3.0" ) )
991
// split this part in to it's comma-separated bits and
992
// add them on individual lines
993
String[] adr_part_parts =
994
splitValueByCharacter( adr_parts[ a ], ',' );
995
for( int b = 0; b < adr_part_parts.length; b++ )
996
if( adr_part_parts[ b ].length() > 0 )
998
if( value.length() > 0 ) value += "\n";
999
value += adr_part_parts[ b ];
1004
// add this part on an individual line
1005
if( value.length() > 0 ) value += "\n";
1006
value += adr_parts[ a ];
1010
Set< String > types = extractTypes( params, Arrays.asList(
1011
"PREF", "WORK", "HOME" ) );
1015
if( types.contains( "WORK" ) )
1020
addAddress( unescapeValue( value ), type );
1023
private void parseLABEL( String[] params, String value )
1025
Set< String > types = extractTypes( params, Arrays.asList(
1026
"PREF", "WORK", "HOME" ) );
1030
if( types.contains( "WORK" ) )
1035
addAddress( unescapeValue( value ), type );
1038
private void parseNOTE( String[] params, String value )
1040
addNote( unescapeValue( value ) );
1043
private void parseBDAY( String[] params, String value )
1045
setBirthday( value );
1048
public void finaliseVcard()
1049
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,
1051
843
// missing version (and data is present)
1052
if( _version == null && _content_lines != null )
844
if( _version == null && _buffers != null )
1053
845
throw new ParseException( R.string.error_vcf_malformed );
1055
// 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 );
1060
* Amongst the params, find the value of the first, only, of any with
1061
* the specified name.
1065
* @return a value, or null
1067
857
private String checkParam( String[] params, String name )
1069
String[] res = checkParams( params, name );
1070
return res.length > 0? res[ 0 ] : null;
1074
* Amongst the params, find the values of any with the specified name.
1078
* @return an array of values, or null
1080
private String[] checkParams( String[] params, String name )
1082
HashSet< String > ret = new HashSet< String >();
1084
859
Pattern p = Pattern.compile(
1085
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1086
Pattern.CASE_INSENSITIVE );
860
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1087
861
for( int i = 0; i < params.length; i++ ) {
1088
862
Matcher m = p.matcher( params[ i ] );
1089
863
if( m.matches() )
1090
ret.add( m.group( 2 ) );
1093
return (String[]) ret.toArray( new String[ ret.size() ] );
1097
* Amongst the params, return any type values present. For v2.1 vCards,
1098
* those types are just parameters. For v3.0, they are prefixed with
1099
* "TYPE=". There may also be multiple type parameters.
1101
* @param params an array of params to look for types in
1102
* @param valid_types an list of upper-case type values to look for
1103
* @return a set of present type values
1105
869
private Set< String > extractTypes( String[] params,
1106
870
List< String > valid_types )
1108
872
HashSet< String > types = new HashSet< String >();
1110
874
// get 3.0-style TYPE= param
1111
String type_params[] = checkParams( params, "TYPE" );
1112
for( int a = 0; a < type_params.length; a++ )
1114
// check for a comma-separated list of types (why? I don't think
1115
// this is in the specs!)
1116
String[] parts = type_params[ a ].split( "," );
1117
for( int i = 0; i < parts.length; i++ ) {
1118
String ucpart = parts[ i ].toUpperCase( Locale.US );
1119
if( valid_types.contains( ucpart ) )
1120
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 ] );
1124
883
// get 2.1-style type param
1125
884
if( _version.equals( "2.1" ) ) {
1126
for( int i = 1; i < params.length; i++ ) {
1127
String ucparam = params[ i ].toUpperCase( Locale.US );
1128
if( valid_types.contains( ucparam ) )
1129
types.add( ucparam );
885
for( int i = 1; i < params.length; i++ )
886
if( valid_types.contains( params[ i ] ) )
887
types.add( params[ i ] );