(root)/android/import-contacts
» Revision
45
(compared to revision 39)
: src/org/waxworlds/edam/importcontacts/VcardImporter.java
To get this branch, use:
bzr branch
http://bzr.ed.am/android/import-contacts
« back to all changes in this revision
Viewing changes to src/org/waxworlds/edam/importcontacts/VcardImporter.java
- browse files at revision 45
- compare with another revision
- download diff
- download tarball
- view history from revision 45
- Committer: edam
- Date: 2011-06-04 13:16:20 UTC
- Revision ID: edam@waxworlds.org-20110604131620-hean9j66r3f5i72h
- fix UTF-8 unencoding by default on v3.0 vCards
- proper unescaping of strings in v3.0 vCards (also works for v2.1, although the 2.1 specs don't mention it)
- proper unescaping of strings in v3.0 vCards (also works for v2.1, although the 2.1 specs don't mention it)
- files renamed:
- src/org/waxworlds/edam/importcontacts/VCFImporter.java => src/org/waxworlds/edam/importcontacts/VcardImporter.java
- files modified:
- NEWS
added
removed
src/org/waxworlds/edam/importcontacts/VcardImporter.java
1
1
/*
2
2
* VCFImporter.java
3
3
*
4
* Copyright (C) 2009 Tim Marston <edam@waxworlds.org>
4
* Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
5
5
*
6
6
* This file is part of the Import Contacts program (hereafter referred
7
7
* to as "this program"). For more information, see
34
34
import java.nio.ByteBuffer;
35
35
import java.util.ArrayList;
36
36
import java.util.Arrays;
37
import java.util.HashMap;
37
38
import java.util.HashSet;
38
39
import java.util.Iterator;
39
40
import java.util.List;
47
48
import android.provider.Contacts;
48
49
import android.provider.Contacts.PhonesColumns;
49
50
50
public class VCFImporter extends Importer
51
public class VcardImporter extends Importer
51
52
{
52
private int _vCardCount = 0;
53
private int _vcard_count = 0;
53
54
private int _progress = 0;
54
55
55
public VCFImporter( Doit doit )
56
public VcardImporter( Doit doit )
56
57
{
57
58
super( doit );
58
59
}
109
110
countVCardFile( files[ i ] );
110
111
setTmpProgress( i );
111
112
}
112
setProgressMax( _vCardCount ); // will also update tmp progress
113
setProgressMax( _vcard_count ); // will also update tmp progress
113
114
114
115
// import them
115
116
setProgress( 0 );
116
117
for( int i = 0; i < files.length; i++ )
117
118
importVCardFile( files[ i ] );
119
setProgress( _vcard_count );
118
120
}
119
121
120
122
private void countVCardFile( File file ) throws AbortImportException
127
129
128
130
// read
129
131
String line;
130
boolean inVCard = false;
132
boolean in_vcard = false;
131
133
while( ( line = reader.readLine() ) != null )
132
134
{
133
if( !inVCard ) {
135
if( !in_vcard ) {
134
136
// look for vcard beginning
135
137
if( line.matches( "^BEGIN:VCARD" ) ) {
136
inVCard = true;
137
_vCardCount++;
138
in_vcard = true;
139
_vcard_count++;
138
140
}
139
141
}
140
142
else if( line.matches( "^END:VCARD" ) )
141
inVCard = false;
143
in_vcard = false;
142
144
}
143
145
144
146
}
167
169
FileInputStream istream = new FileInputStream( file );
168
170
byte[] content = new byte[ (int)file.length() ];
169
171
istream.read( content );
172
istream = null;
170
173
171
174
// import
172
175
importVCardFileContent( content, file.getName() );
184
187
throws AbortImportException
185
188
{
186
189
// go through lines
187
VCard vCard = null;
190
Vcard vcard = null;
191
int vcard_start_line = 0;
188
192
ContentLineIterator cli = new ContentLineIterator( content );
189
193
while( cli.hasNext() )
190
194
{
201
205
line = "";
202
206
}
203
207
204
if( vCard == null ) {
208
if( vcard == null ) {
205
209
// look for vcard beginning
206
210
if( line.matches( "^BEGIN:VCARD" ) ) {
207
setProgress( ++_progress );
208
vCard = new VCard();
211
setProgress( _progress++ );
212
vcard = new Vcard();
213
vcard_start_line = cli.getLineNumber();
209
214
}
210
215
}
211
216
else {
212
217
// look for vcard content or ending
213
218
if( line.matches( "^END:VCARD" ) )
214
219
{
215
// store vcard and do away with it
220
// finalise the vcard/contact
216
221
try {
217
vCard.finaliseParsing();
218
importContact( vCard );
219
}
220
catch( VCard.ParseException e ) {
221
skipContact();
222
if( !showContinue(
223
getText( R.string.error_vcf_parse ).toString()
224
+ fileName + "\n" + e.getMessage() ) )
225
{
226
finish( ACTION_ABORT );
227
}
228
}
229
catch( VCard.SkipContactException e ) {
230
skipContact();
231
// do nothing
232
}
233
vCard = null;
222
vcard.finaliseVcard();
223
224
// pass the finalised contact to the importer
225
importContact( vcard );
226
}
227
catch( Vcard.ParseException e ) {
228
if( !showContinue(
229
getText( R.string.error_vcf_parse ).toString()
230
+ fileName +
231
getText( R.string.error_vcf_parse_line ).toString()
232
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
233
{
234
finish( ACTION_ABORT );
235
}
236
else
237
skipContact();
238
}
239
catch( ContactData.ContactNotIdentifiableException e ) {
240
if( !showContinue(
241
getText( R.string.error_vcf_parse ).toString()
242
+ fileName +
243
getText( R.string.error_vcf_parse_line ).toString()
244
+ vcard_start_line + ":\n" + getText(
245
R.string.error_vcf_notenoughinfo ).toString()
246
) )
247
{
248
finish( ACTION_ABORT );
249
}
250
else
251
skipContact();
252
}
253
254
// discard this vcard
255
vcard = null;
234
256
}
235
257
else
236
258
{
237
259
// try giving the line to the vcard
238
260
try {
239
vCard.parseLine( buffer, line,
261
vcard.parseLine( buffer, line,
240
262
cli.doesNextLineLookFolded() );
241
263
}
242
catch( VCard.ParseException e ) {
264
catch( Vcard.ParseException e ) {
243
265
skipContact();
244
266
if( !showContinue(
245
267
getText( R.string.error_vcf_parse ).toString()
246
+ fileName + "\n" + e.getMessage() ) )
268
+ fileName +
269
getText( R.string.error_vcf_parse_line ).toString()
270
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
247
271
{
248
272
finish( ACTION_ABORT );
249
273
}
251
275
// although we're continuing, we still need to abort
252
276
// this vCard. Further lines will be ignored until we
253
277
// get to another BEGIN:VCARD line.
254
vCard = null;
278
vcard = null;
255
279
}
256
catch( VCard.SkipContactException e ) {
280
catch( Vcard.SkipImportException e ) {
257
281
skipContact();
258
282
// abort this vCard. Further lines will be ignored until
259
283
// we get to another BEGIN:VCARD line.
260
vCard = null;
284
vcard = null;
261
285
}
262
286
}
263
287
}
268
292
{
269
293
protected byte[] _content = null;
270
294
protected int _pos = 0;
295
protected int _line = 0;
271
296
272
297
public ContentLineIterator( byte[] content )
273
298
{
293
318
int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
294
319
_pos > initial_pos )? _pos - 1 : _pos;
295
320
_pos++;
321
_line++;
296
322
return ByteBuffer.wrap( _content, initial_pos,
297
323
to - initial_pos );
298
324
}
301
327
if( _pos != initial_pos ) {
302
328
int to = _pos;
303
329
_pos++;
330
_line++;
304
331
return ByteBuffer.wrap( _content, initial_pos,
305
332
to - initial_pos );
306
333
}
325
352
return _pos > 0 && _pos < _content.length &&
326
353
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
327
354
}
355
356
public int getLineNumber()
357
{
358
return _line;
359
}
328
360
}
329
361
330
private class VCard extends ContactData
362
private class Vcard extends ContactData
331
363
{
332
364
private final static int NAMELEVEL_NONE = 0;
333
private final static int NAMELEVEL_ORG = 1;
365
private final static int NAMELEVEL_N = 1;
334
366
private final static int NAMELEVEL_FN = 2;
335
private final static int NAMELEVEL_N = 3;
336
367
337
368
private final static int MULTILINE_NONE = 0;
338
369
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
345
376
private int _parser_multiline_state = MULTILINE_NONE;
346
377
private String _parser_current_name_and_params = null;
347
378
private String _parser_buffered_value_so_far = "";
379
private String _cached_organisation = null;
380
private String _cached_title = null;
348
381
349
382
protected class UnencodeResult
350
383
{
380
413
381
414
public ParseException( int res )
382
415
{
383
super( VCFImporter.this.getText( res ).toString() );
416
super( VcardImporter.this.getText( res ).toString() );
384
417
}
385
418
}
386
419
387
420
@SuppressWarnings("serial")
388
protected class SkipContactException extends Exception { }
421
protected class SkipImportException extends Exception { }
389
422
390
423
private String extractCollonPartFromLine( ByteBuffer buffer,
391
424
String line, boolean former )
427
460
428
461
public void parseLine( ByteBuffer buffer, String line,
429
462
boolean next_line_looks_folded )
430
throws ParseException, SkipContactException,
463
throws ParseException, SkipImportException,
431
464
AbortImportException
432
465
{
433
466
// do we have a version yet?
533
566
for( int i = 0; i < name_param_parts.length; i++ )
534
567
name_param_parts[ i ] = name_param_parts[ i ].trim();
535
568
569
// determine whether we care about this entry
570
final HashSet< String > interesting_fields =
571
new HashSet< String >( Arrays.asList( new String[]
572
{ "N", "FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR" }
573
) );
574
boolean is_interesting_field =
575
interesting_fields.contains( name_param_parts[ 0 ] );
576
536
577
// parse encoding parameter
537
578
String encoding = checkParam( name_param_parts, "ENCODING" );
538
579
if( encoding != null ) encoding = encoding.toUpperCase();
539
if( encoding != null && !encoding.equals( "8BIT" ) &&
580
if( is_interesting_field && encoding != null &&
581
!encoding.equals( "8BIT" ) &&
540
582
!encoding.equals( "QUOTED-PRINTABLE" ) )
541
583
//&& !encoding.equals( "BASE64" ) )
542
584
{
546
588
// parse charset parameter
547
589
String charset = checkParam( name_param_parts, "CHARSET" );
548
590
if( charset != null ) charset = charset.toUpperCase();
549
if( charset != null && !charset.equals( "US-ASCII" ) &&
591
if( charset != null &&
592
!charset.equals( "US-ASCII" ) &&
550
593
!charset.equals( "ASCII" ) &&
551
594
!charset.equals( "UTF-8" ) )
552
595
{
566
609
_parser_multiline_state = MULTILINE_ENCODED;
567
610
}
568
611
569
// convert 8-bit ASCII charset to US-ASCII
570
if( charset == null || charset.equals( "ASCII" ) ) {
612
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
613
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
614
if( ( charset == null && _version.equals( "2.1" ) ) ||
615
( charset != null && (
616
charset.equals( "ASCII" ) ||
617
charset.equals( "US-ASCII" ) ) ) )
618
{
571
619
value = transcodeAsciiToUtf8( value );
572
charset = "UTF-8";
573
620
}
574
621
575
// process charset
622
// process charset (value is now in UTF-8)
576
623
String string_value;
577
624
try {
578
625
string_value = new String( value.array(), value.position(),
579
value.limit() - value.position(), charset );
626
value.limit() - value.position(), "UTF-8" );
580
627
} catch( UnsupportedEncodingException e ) {
581
628
throw new ParseException( R.string.error_vcf_charset );
582
629
}
621
668
parseFN( name_param_parts, complete_value );
622
669
else if( name_param_parts[ 0 ].equals( "ORG" ) )
623
670
parseORG( name_param_parts, complete_value );
671
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
672
parseTITLE( name_param_parts, complete_value );
624
673
else if( name_param_parts[ 0 ].equals( "TEL" ) )
625
674
parseTEL( name_param_parts, complete_value );
626
675
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
645
694
return ( count & 1 ) == 1;
646
695
}
647
696
648
private String[] splitValueBySemicolon( String value )
697
private String[] splitValueByCharacter( String value, char character )
649
698
{
650
// split string in to parts by semicolon
699
// split string in to parts by specified character
651
700
ArrayList< String > parts = new ArrayList< String >(
652
Arrays.asList( value.split( ";" ) ) );
701
Arrays.asList( value.split( "" + character ) ) );
653
702
654
703
// go through parts
655
704
for( int a = 0; a < parts.size(); a++ )
663
712
if( a < parts.size() - 1 &&
664
713
doesStringEndInAnEscapeChar( str ) )
665
714
{
666
// join the next part to this part and remove the next part
715
// append the escaped character, join the next part to this
716
// part and remove the next part
667
717
parts.set( a, str.substring( 0, str.length() - 1 ) +
668
';' + parts.get( a + 1 ) );
718
character + parts.get( a + 1 ) );
669
719
parts.remove( a + 1 );
670
720
671
721
// re-visit this part
682
732
return parts.toArray( ret );
683
733
}
684
734
735
private String unescapeValue( String value )
736
{
737
StringBuilder ret = new StringBuilder( value.length() );
738
boolean in_escape = false;
739
for( int a = 0; a < value.length(); a++ )
740
{
741
int c = value.codePointAt( a );
742
743
// process a normal character
744
if( !in_escape ) {
745
if( c == '\\' )
746
in_escape = true;
747
else
748
ret.append( Character.toChars( c ) );
749
continue;
750
}
751
752
// process an escape sequence
753
in_escape = false;
754
switch( c )
755
{
756
case 'N':
757
case 'n':
758
// add newline
759
ret.append( '\n' );
760
break;
761
case '\\':
762
case ',':
763
case ';':
764
// add escaped character
765
ret.append( Character.toChars( c ) );
766
break;
767
default:
768
// unknown escape sequence, so add it unescaped
769
ret.append( "\\" );
770
ret.append( Character.toChars( c ) );
771
break;
772
}
773
}
774
775
return ret.toString();
776
}
777
685
778
private void parseN( String[] params, String value )
686
throws ParseException, SkipContactException,
687
AbortImportException
688
779
{
689
780
// already got a better name?
690
781
if( _name_level >= NAMELEVEL_N ) return;
691
782
692
783
// get name parts
693
String[] name_parts = splitValueBySemicolon( value );
784
String[] name_parts = splitValueByCharacter( value, ';' );
694
785
695
786
// build name
696
787
value = "";
697
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
698
value += name_parts[ 1 ];
699
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
700
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
788
final int[] part_order = { 3, 1, 2, 0, 4 };
789
for( int a = 0; a < part_order.length; a++ )
790
if( name_parts.length > part_order[ a ] &&
791
name_parts[ part_order[ a ] ].length() > 0 )
792
{
793
// split this part in to it's comma-separated bits
794
String[] name_part_parts = splitValueByCharacter(
795
name_parts[ part_order[ a ] ], ',' );
796
for( int b = 0; b < name_part_parts.length; b++ )
797
if( name_part_parts[ b ].length() > 0 )
798
{
799
if( value.length() == 0 ) value += " ";
800
value += name_part_parts[ b ];
801
}
802
}
701
803
702
804
// set name
703
setName( value );
805
setName( unescapeValue( value ) );
704
806
_name_level = NAMELEVEL_N;
705
706
// check now to see if we need to import this contact (to avoid
707
// parsing the rest of the vCard unnecessarily)
708
if( !isImportRequired( getName() ) )
709
throw new SkipContactException();
710
807
}
711
808
712
809
private void parseFN( String[] params, String value )
713
throws ParseException, SkipContactException
714
810
{
715
811
// already got a better name?
716
812
if( _name_level >= NAMELEVEL_FN ) return;
717
813
718
814
// set name
719
setName( value );
815
setName( unescapeValue( value ) );
720
816
_name_level = NAMELEVEL_FN;
721
817
}
722
818
723
819
private void parseORG( String[] params, String value )
724
throws ParseException, SkipContactException
725
820
{
726
// already got a better name?
727
if( _name_level >= NAMELEVEL_ORG ) return;
728
729
821
// get org parts
730
String[] org_parts = splitValueBySemicolon( value );
731
732
// build name
733
if( org_parts.length > 1 && org_parts[ 0 ].length() == 0 )
734
value = org_parts[ 1 ];
735
else if( org_parts.length > 1 && org_parts[ 1 ].length() > 0 )
736
value = org_parts[ 0 ] + ", " + org_parts[ 1 ];
737
else
738
value = org_parts[ 0 ];
739
740
// set name
741
setName( value );
742
_name_level = NAMELEVEL_ORG;
822
String[] org_parts = splitValueByCharacter( value, ';' );
823
if( org_parts == null || org_parts.length < 1 ) return;
824
825
// build organisation name
826
StringBuilder builder = new StringBuilder(
827
String.valueOf( org_parts[ 0 ] ) );
828
for( int a = 1; a < org_parts.length; a++ )
829
builder.append( ", " ).append( org_parts[ a ] );
830
String organisation = unescapeValue( builder.toString() );
831
832
// set organisation name (using a title we've previously found)
833
addOrganisation( organisation, _cached_title, true );
834
835
// if we've not previously found a title, store this organisation
836
// name (we'll need it when we find a title to update the
837
// organisation, by name), else if we *have* previously found a
838
// title, clear it (since we just used it)
839
if( _cached_title == null )
840
_cached_organisation = organisation;
841
else
842
_cached_title = null;
843
}
844
845
private void parseTITLE( String[] params, String value )
846
{
847
value = unescapeValue( value );
848
849
// if we previously had an organisation, look it up and append this
850
// title to it
851
if( _cached_organisation != null && hasOrganisations() ) {
852
HashMap< String, ExtraDetail > datas = getOrganisations();
853
ExtraDetail detail = datas.get( _cached_organisation );
854
if( detail != null )
855
detail.setExtra( value );
856
}
857
858
// same as when handling organisation, if we've not previously found
859
// an organisation we store this title, else we clear it (since we
860
// just appended this title to it)
861
if( _cached_organisation == null )
862
_cached_title = value;
863
else
864
_cached_organisation = null;
743
865
}
744
866
745
867
private void parseTEL( String[] params, String value )
746
throws ParseException
747
868
{
748
869
if( value.length() == 0 ) return;
749
870
752
873
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
753
874
754
875
// here's the logic...
755
boolean preferred = types.contains( "PREF" );
756
int type = PhonesColumns.TYPE_MOBILE;
757
if( types.contains( "VOICE" ) )
758
if( types.contains( "WORK" ) )
759
type = PhonesColumns.TYPE_WORK;
760
else
761
type = PhonesColumns.TYPE_HOME;
762
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
763
type = PhonesColumns.TYPE_MOBILE;
876
boolean is_preferred = types.contains( "PREF" );
877
int type;
764
878
if( types.contains( "FAX" ) )
765
879
if( types.contains( "HOME" ) )
766
880
type = PhonesColumns.TYPE_FAX_HOME;
767
881
else
768
882
type = PhonesColumns.TYPE_FAX_WORK;
769
if( types.contains( "PAGER" ) )
883
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
884
type = PhonesColumns.TYPE_MOBILE;
885
else if( types.contains( "PAGER" ) )
770
886
type = PhonesColumns.TYPE_PAGER;
887
else if( types.contains( "WORK" ) )
888
type = PhonesColumns.TYPE_WORK;
889
else
890
type = PhonesColumns.TYPE_HOME;
771
891
772
892
// add phone number
773
addPhone( value, type, preferred );
893
addNumber( value, type, is_preferred );
774
894
}
775
895
776
896
public void parseEMAIL( String[] params, String value )
777
throws ParseException
778
897
{
779
898
if( value.length() == 0 ) return;
780
899
782
901
"PREF", "WORK", "HOME", "INTERNET" ) );
783
902
784
903
// add email address
785
boolean preferred = types.contains( "PREF" );
904
boolean is_preferred = types.contains( "PREF" );
905
int type;
786
906
if( types.contains( "WORK" ) )
787
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
907
type = Contacts.ContactMethods.TYPE_WORK;
788
908
else
789
addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
909
type = Contacts.ContactMethods.TYPE_HOME;
910
911
addEmail( unescapeValue( value ), type, is_preferred );
790
912
}
791
913
792
914
private void parseADR( String[] params, String value )
793
throws ParseException, SkipContactException
794
915
{
795
916
// get address parts
796
String[] adr_parts = splitValueBySemicolon( value );
917
String[] adr_parts = splitValueByCharacter( value, ';' );
797
918
798
919
// build address
799
920
value = "";
800
for( int a = 0; a < adr_parts.length; a++ ) {
801
if( value.length() > 0 ) value += "\n";
802
value += adr_parts[ a ].trim();
803
}
921
for( int a = 0; a < adr_parts.length; a++ )
922
if( adr_parts[ a ].length() > 0 )
923
{
924
// split this part in to it's comma-separated bits
925
String[] adr_part_parts =
926
splitValueByCharacter( adr_parts[ a ], ',' );
927
for( int b = 0; b < adr_part_parts.length; b++ )
928
if( adr_part_parts[ b ].length() > 0 )
929
{
930
if( value.length() > 0 ) value += "\n";
931
value += adr_part_parts[ b ];
932
}
933
}
804
934
805
935
Set< String > types = extractTypes( params, Arrays.asList(
806
936
"PREF", "WORK", "HOME", "INTERNET" ) );
807
937
808
938
// add address
939
int type;
809
940
if( types.contains( "WORK" ) )
810
addAddress( value, Contacts.ContactMethods.TYPE_WORK );
941
type = Contacts.ContactMethods.TYPE_WORK;
811
942
else
812
addAddress( value, Contacts.ContactMethods.TYPE_HOME);
943
type = Contacts.ContactMethods.TYPE_HOME;
944
945
addAddress( unescapeValue( value ), type );
813
946
}
814
947
815
public void finaliseParsing()
816
throws ParseException, SkipContactException,
817
AbortImportException
948
public void finaliseVcard()
949
throws ParseException, ContactNotIdentifiableException
818
950
{
819
951
// missing version (and data is present)
820
952
if( _version == null && _buffers != null )
821
953
throw new ParseException( R.string.error_vcf_malformed );
822
954
823
// missing name properties?
824
if( _name_level == NAMELEVEL_NONE )
825
throw new ParseException( R.string.error_vcf_noname );
826
827
// check if we should import this one? If we've already got an 'N'-
828
// type name, this will already have been done by parseN() so we
829
// mustn't do this here (or it could prompt twice!)
830
if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
831
throw new SkipContactException();
955
// finalise the parent class
956
finalise();
832
957
}
833
958
834
959
private String checkParam( String[] params, String name )
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480