198
202
buffer.limit() - buffer.position(), "US-ASCII" );
200
204
catch( UnsupportedEncodingException e ) {
201
// we know US-ASCII is supported, so appease the compiler...
205
// we know US-ASCII *is* supported, so appease the compiler...
205
209
if( vcard == null ) {
206
210
// look for vcard beginning
207
if( line.matches( "^BEGIN:VCARD" ) ) {
208
setProgress( ++_progress );
211
if( line.matches( "^BEGIN[ \t]*:[ \t]*VCARD" ) ) {
212
setProgress( _progress++ );
214
vcard_start_line = cli.getLineNumber();
213
218
// look for vcard content or ending
214
if( line.matches( "^END:VCARD" ) )
219
if( line.matches( "^END[ \t]*:[ \t]*VCARD" ) )
216
// store vcard and do away with it
221
// finalise the vcard/contact
218
vcard.finaliseParsing();
223
vcard.finaliseVcard();
225
// pass the finalised contact to the importer
219
226
importContact( vcard );
221
catch( VCard.ParseException e ) {
224
getText( R.string.error_vcf_parse ).toString()
225
+ fileName + "\n" + e.getMessage() ) )
227
finish( ACTION_ABORT );
230
catch( VCard.SkipContactException e ) {
228
catch( Vcard.ParseException e ) {
230
getText( R.string.error_vcf_parse ).toString()
232
getText( R.string.error_vcf_parse_line ).toString()
233
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
235
finish( ACTION_ABORT );
240
catch( ContactData.ContactNotIdentifiableException e ) {
242
getText( R.string.error_vcf_parse ).toString()
244
getText( R.string.error_vcf_parse_line ).toString()
245
+ vcard_start_line + ":\n" + getText(
246
R.string.error_vcf_notenoughinfo ).toString()
249
finish( ACTION_ABORT );
255
// discard this vcard
535
571
for( int i = 0; i < name_param_parts.length; i++ )
536
572
name_param_parts[ i ] = name_param_parts[ i ].trim();
574
// determine whether we care about this entry
575
final HashSet< String > interesting_fields =
576
new HashSet< String >( Arrays.asList( new String[] { "N",
577
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
579
boolean is_interesting_field =
580
interesting_fields.contains(
581
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
538
583
// parse encoding parameter
539
584
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" ) )
585
if( encoding != null )
586
encoding = encoding.toUpperCase( Locale.US );
587
if( is_interesting_field && encoding != null &&
588
!encoding.equalsIgnoreCase( "8BIT" ) &&
589
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
590
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
545
592
throw new ParseException( R.string.error_vcf_encoding );
548
595
// parse charset parameter
549
596
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" ) )
597
if( charset != null )
598
charset = charset.toUpperCase( Locale.US );
599
if( charset != null &&
600
!charset.equalsIgnoreCase( "US-ASCII" ) &&
601
!charset.equalsIgnoreCase( "ASCII" ) &&
602
!charset.equalsIgnoreCase( "UTF-8" ) )
555
604
throw new ParseException( R.string.error_vcf_charset );
558
607
// do unencoding (or default to a fake unencoding result with
559
608
// the raw string)
560
609
UnencodeResult unencoding_result = null;
561
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
610
if( encoding != null &&
611
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
562
613
unencoding_result = unencodeQuotedPrintable( value );
563
// else if( encoding != null && encoding.equals( "BASE64" ) )
615
// else if( encoding != null &&
616
// encoding.equalsIgnoreCase( "BASE64" ) )
564
618
// unencoding_result = unencodeBase64( props[ 1 ], charset );
565
620
if( unencoding_result != null ) {
566
621
value = unencoding_result.getBuffer();
567
622
if( unencoding_result.isAnotherLineRequired() )
568
623
_parser_multiline_state = MULTILINE_ENCODED;
571
// convert 8-bit ASCII charset to US-ASCII
572
if( charset == null || charset.equals( "ASCII" ) ) {
626
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
627
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
628
if( ( charset == null && _version.equals( "2.1" ) ) ||
629
( charset != null && (
630
charset.equalsIgnoreCase( "ASCII" ) ||
631
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
573
633
value = transcodeAsciiToUtf8( value );
636
// process charset (value is now in UTF-8)
578
637
String string_value;
580
639
string_value = new String( value.array(), value.position(),
581
value.limit() - value.position(), charset );
640
value.limit() - value.position(), "UTF-8" );
582
641
} catch( UnsupportedEncodingException e ) {
583
642
throw new ParseException( R.string.error_vcf_charset );
617
676
if( complete_value.length() < 1 ) return;
619
678
// parse some properties
620
if( name_param_parts[ 0 ].equals( "N" ) )
679
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
621
680
parseN( name_param_parts, complete_value );
622
else if( name_param_parts[ 0 ].equals( "FN" ) )
681
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
623
682
parseFN( name_param_parts, complete_value );
624
else if( name_param_parts[ 0 ].equals( "ORG" ) )
683
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
625
684
parseORG( name_param_parts, complete_value );
626
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
685
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
627
686
parseTITLE( name_param_parts, complete_value );
628
else if( name_param_parts[ 0 ].equals( "TEL" ) )
687
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
629
688
parseTEL( name_param_parts, complete_value );
630
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
689
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
631
690
parseEMAIL( name_param_parts, complete_value );
632
else if( name_param_parts[ 0 ].equals( "ADR" ) )
691
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
633
692
parseADR( name_param_parts, complete_value );
693
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
694
parseLABEL( name_param_parts, complete_value );
695
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
696
parseNOTE( name_param_parts, complete_value );
686
750
return parts.toArray( ret );
753
private String unescapeValue( String value )
755
StringBuilder ret = new StringBuilder( value.length() );
756
boolean in_escape = false;
757
for( int a = 0; a < value.length(); a++ )
759
int c = value.codePointAt( a );
761
// process a normal character
766
ret.append( Character.toChars( c ) );
770
// process an escape sequence
776
// add tab (invalid/non-standard, but accepted)
787
// add escaped character
788
ret.append( Character.toChars( c ) );
791
// unknown escape sequence, so add it unescaped
792
// (invalid/non-standard, but accepted)
794
ret.append( Character.toChars( c ) );
799
return ret.toString();
689
802
private void parseN( String[] params, String value )
691
804
// already got a better name?
692
805
if( _name_level >= NAMELEVEL_N ) return;
694
807
// get name parts
695
String[] name_parts = splitValueBySemicolon( value );
808
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 ];
812
final int[] part_order = { 3, 1, 2, 0, 4 };
813
for( int a = 0; a < part_order.length; a++ )
814
if( name_parts.length > part_order[ a ] &&
815
name_parts[ part_order[ a ] ].length() > 0 )
817
// split this part in to it's comma-separated bits
818
String[] name_part_parts = splitValueByCharacter(
819
name_parts[ part_order[ a ] ], ',' );
820
for( int b = 0; b < name_part_parts.length; b++ )
821
if( name_part_parts[ b ].length() > 0 )
823
if( value.length() == 0 ) value += " ";
824
value += name_part_parts[ b ];
829
setName( unescapeValue( value ) );
706
830
_name_level = NAMELEVEL_N;
802
928
boolean is_preferred = types.contains( "PREF" );
804
930
if( types.contains( "WORK" ) )
805
type = Contacts.ContactMethods.TYPE_WORK;
807
type = Contacts.ContactMethods.TYPE_HOME;
809
addEmail( value, type, is_preferred );
935
addEmail( unescapeValue( value ), type, is_preferred );
812
938
private void parseADR( String[] params, String value )
814
940
// get address parts
815
String[] adr_parts = splitValueBySemicolon( value );
941
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 );
837
public void finaliseParsing()
838
throws ParseException, SkipContactException,
945
for( int a = 0; a < adr_parts.length; a++ )
946
if( adr_parts[ a ].length() > 0 )
948
// version 3.0 vCards allow further splitting by comma
949
if( _version.equals( "3.0" ) )
951
// split this part in to it's comma-separated bits and
952
// add them on individual lines
953
String[] adr_part_parts =
954
splitValueByCharacter( adr_parts[ a ], ',' );
955
for( int b = 0; b < adr_part_parts.length; b++ )
956
if( adr_part_parts[ b ].length() > 0 )
958
if( value.length() > 0 ) value += "\n";
959
value += adr_part_parts[ b ];
964
// add this part on an individual line
965
if( value.length() > 0 ) value += "\n";
966
value += adr_parts[ a ];
970
Set< String > types = extractTypes( params, Arrays.asList(
971
"PREF", "WORK", "HOME" ) );
975
if( types.contains( "WORK" ) )
980
addAddress( unescapeValue( value ), type );
983
private void parseLABEL( String[] params, String value )
985
Set< String > types = extractTypes( params, Arrays.asList(
986
"PREF", "WORK", "HOME" ) );
990
if( types.contains( "WORK" ) )
995
addAddress( unescapeValue( value ), type );
998
private void parseNOTE( String[] params, String value )
1000
addNote( unescapeValue( value ) );
1003
public void finaliseVcard()
1004
throws ParseException, ContactNotIdentifiableException
841
1006
// missing version (and data is present)
842
1007
if( _version == null && _buffers != null )
843
1008
throw new ParseException( R.string.error_vcf_malformed );
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 );
1010
// finalise the parent class
1015
* Amongst the params, find the value of the first, only, of any with
1016
* the specified name
1019
* @return a value, or null
855
1021
private String checkParam( String[] params, String name )
1023
String[] res = checkParams( params, name );
1024
return res.length > 0? res[ 0 ] : null;
1028
* Amongst the params, find the values of any with the specified name
1031
* @return an array of values, or null
1033
private String[] checkParams( String[] params, String name )
1035
HashSet< String > ret = new HashSet< String >();
857
1037
Pattern p = Pattern.compile(
858
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1038
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1039
Pattern.CASE_INSENSITIVE );
859
1040
for( int i = 0; i < params.length; i++ ) {
860
1041
Matcher m = p.matcher( params[ i ] );
861
1042
if( m.matches() )
1043
ret.add( m.group( 2 ) );
1046
return (String[]) ret.toArray( new String[ ret.size() ] );
1050
* Amongst the params, return any type values present. For v2.1 vCards,
1051
* those types are just parameters. For v3.0, they are prefixed with
1052
* "TYPE=". There may also be multiple type parameters.
1053
* @param params an array of params to look for types in
1054
* @param valid_types an list of upper-case type values to look for
1055
* @return a set of present type values
867
1057
private Set< String > extractTypes( String[] params,
868
1058
List< String > valid_types )
870
1060
HashSet< String > types = new HashSet< String >();
872
1062
// get 3.0-style TYPE= param
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 ] );
1063
String type_params[] = checkParams( params, "TYPE" );
1064
for( int a = 0; a < type_params.length; a++ )
1066
// check for a comma-separated list of types (why? I don't think
1067
// this is in the specs!)
1068
String[] parts = type_params[ a ].split( "," );
1069
for( int i = 0; i < parts.length; i++ ) {
1070
String ucpart = parts[ i ].toUpperCase( Locale.US );
1071
if( valid_types.contains( ucpart ) )
1072
types.add( ucpart );
881
1076
// get 2.1-style type param
882
1077
if( _version.equals( "2.1" ) ) {
883
for( int i = 1; i < params.length; i++ )
884
if( valid_types.contains( params[ i ] ) )
885
types.add( params[ i ] );
1078
for( int i = 1; i < params.length; i++ ) {
1079
String ucparam = params[ i ].toUpperCase( Locale.US );
1080
if( valid_types.contains( ucparam ) )
1081
types.add( ucparam );