(root)/android/import-contacts
» Revision
36
(compared to revision 41)
: src/org/waxworlds/edam/importcontacts/VCFImporter.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/VCFImporter.java
- browse files at revision 36
- compare with another revision
- download diff
- download tarball
- view history from revision 36
- Committer: edam
- Date: 2011-03-19 20:33:09 UTC
- Revision ID: edam@waxworlds.org-20110319203309-5dzfyqrxwk94jtin
- formatting: removed some double-indents on overrunning lines
- updated TODO and NEWS
- rewrote central logic of parser so it makes more sense, looks nicer and has a small optimisation (getting name and params from line only when necessary)
- optimised unnecessary mutliple converting of lines to US-ASCII
- re-wrote line extraction from vcards so that we can lookahead for v3 folded lines
- added support for v3 folded lines
- updated TODO and NEWS
- rewrote central logic of parser so it makes more sense, looks nicer and has a small optimisation (getting name and params from line only when necessary)
- optimised unnecessary mutliple converting of lines to US-ASCII
- re-wrote line extraction from vcards so that we can lookahead for v3 folded lines
- added support for v3 folded lines
- files removed:
- src/org/waxworlds/edam/importcontacts/ContactsCache.java
- files modified:
- NEWS
added
removed
src/org/waxworlds/edam/importcontacts/VCFImporter.java
1
1
/*
2
2
* VCFImporter.java
3
3
*
4
* Copyright (C) 2009 to 2011 Tim Marston <edam@waxworlds.org>
4
* Copyright (C) 2009 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
32
32
import java.io.IOException;
33
33
import java.io.UnsupportedEncodingException;
34
34
import java.nio.ByteBuffer;
35
import java.util.ArrayList;
36
35
import java.util.Arrays;
37
import java.util.HashMap;
38
36
import java.util.HashSet;
39
37
import java.util.Iterator;
40
38
import java.util.List;
41
import java.util.NoSuchElementException;
42
39
import java.util.Set;
43
40
import java.util.Vector;
44
41
import java.util.regex.Matcher;
45
42
import java.util.regex.Pattern;
43
import java.util.NoSuchElementException;
44
import java.lang.UnsupportedOperationException;
46
45
47
46
import android.content.SharedPreferences;
48
47
import android.provider.Contacts;
50
49
51
50
public class VCFImporter extends Importer
52
51
{
53
private int _vcard_count = 0;
52
private int _vCardCount = 0;
54
53
private int _progress = 0;
55
54
56
55
public VCFImporter( Doit doit )
110
109
countVCardFile( files[ i ] );
111
110
setTmpProgress( i );
112
111
}
113
setProgressMax( _vcard_count ); // will also update tmp progress
112
setProgressMax( _vCardCount ); // will also update tmp progress
114
113
115
114
// import them
116
115
setProgress( 0 );
128
127
129
128
// read
130
129
String line;
131
boolean in_vcard = false;
130
boolean inVCard = false;
132
131
while( ( line = reader.readLine() ) != null )
133
132
{
134
if( !in_vcard ) {
133
if( !inVCard ) {
135
134
// look for vcard beginning
136
135
if( line.matches( "^BEGIN:VCARD" ) ) {
137
in_vcard = true;
138
_vcard_count++;
136
inVCard = true;
137
_vCardCount++;
139
138
}
140
139
}
141
140
else if( line.matches( "^END:VCARD" ) )
142
in_vcard = false;
141
inVCard = false;
143
142
}
144
143
145
144
}
185
184
throws AbortImportException
186
185
{
187
186
// go through lines
188
VCard vcard = null;
187
VCard vCard = null;
189
188
ContentLineIterator cli = new ContentLineIterator( content );
190
189
while( cli.hasNext() )
191
190
{
202
201
line = "";
203
202
}
204
203
205
if( vcard == null ) {
204
if( vCard == null ) {
206
205
// look for vcard beginning
207
206
if( line.matches( "^BEGIN:VCARD" ) ) {
208
207
setProgress( ++_progress );
209
vcard = new VCard();
208
vCard = new VCard();
210
209
}
211
210
}
212
211
else {
215
214
{
216
215
// store vcard and do away with it
217
216
try {
218
vcard.finaliseParsing();
219
importContact( vcard );
217
vCard.finaliseParsing();
218
importContact( vCard );
220
219
}
221
220
catch( VCard.ParseException e ) {
222
221
skipContact();
231
230
skipContact();
232
231
// do nothing
233
232
}
234
vcard = null;
233
vCard = null;
235
234
}
236
235
else
237
236
{
238
237
// try giving the line to the vcard
239
238
try {
240
vcard.parseLine( buffer, line,
239
vCard.parseLine( buffer, line,
241
240
cli.doesNextLineLookFolded() );
242
241
}
243
242
catch( VCard.ParseException e ) {
252
251
// although we're continuing, we still need to abort
253
252
// this vCard. Further lines will be ignored until we
254
253
// get to another BEGIN:VCARD line.
255
vcard = null;
254
vCard = null;
256
255
}
257
256
catch( VCard.SkipContactException e ) {
258
257
skipContact();
259
258
// abort this vCard. Further lines will be ignored until
260
259
// we get to another BEGIN:VCARD line.
261
vcard = null;
260
vCard = null;
262
261
}
263
262
}
264
263
}
331
330
private class VCard extends ContactData
332
331
{
333
332
private final static int NAMELEVEL_NONE = 0;
334
private final static int NAMELEVEL_FN = 1;
335
private final static int NAMELEVEL_N = 2;
336
337
private final static int MULTILINE_NONE = 0;
338
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
339
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
340
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
333
private final static int NAMELEVEL_ORG = 1;
334
private final static int NAMELEVEL_FN = 2;
335
private final static int NAMELEVEL_N = 3;
341
336
342
337
private String _version = null;
343
338
private Vector< ByteBuffer > _buffers = null;
344
339
private int _name_level = NAMELEVEL_NONE;
345
private int _parser_multiline_state = MULTILINE_NONE;
340
private boolean _parser_in_encoded_multiline = false;
341
private boolean _parser_in_folded_multiline = false;
346
342
private String _parser_current_name_and_params = null;
347
343
private String _parser_buffered_value_so_far = "";
348
private String _cached_organisation = null;
349
private String _cached_title = null;
350
344
351
345
protected class UnencodeResult
352
346
{
475
469
String name_and_params;
476
470
int pos;
477
471
478
if( _parser_multiline_state != MULTILINE_NONE )
472
if( _parser_in_encoded_multiline ||
473
_parser_in_folded_multiline )
479
474
{
480
475
// if we're currently in a multi-line value, use the stored
481
476
// property name and parameters
482
477
name_and_params = _parser_current_name_and_params;
483
478
484
// skip some initial line characters, depending on the type
485
// of multi-line we're handling
486
479
pos = buffer.position();
487
switch( _parser_multiline_state )
488
{
489
case MULTILINE_FOLDED:
480
481
// for folded multi-lines, skip the single space at the
482
// start of the next line
483
if( _parser_in_folded_multiline )
490
484
pos++;
491
break;
492
case MULTILINE_ENCODED:
485
486
// else, this must be an encoded multi-line, so skip any
487
// whitespace we find at the start of the next line
488
else
493
489
while( pos < buffer.limit() && (
494
490
buffer.get( pos ) == ' ' ||
495
491
buffer.get( pos ) == '\t' ) )
496
492
{
497
493
pos++;
498
494
}
499
break;
500
default:
501
// do nothing
502
}
503
504
// take us out of multi-line so that we can re-detect that
505
// this line is a multi-line or not
506
_parser_multiline_state = MULTILINE_NONE;
507
495
}
508
496
else
509
497
{
564
552
// unencoding_result = unencodeBase64( props[ 1 ], charset );
565
553
if( unencoding_result != null ) {
566
554
value = unencoding_result.getBuffer();
567
if( unencoding_result.isAnotherLineRequired() )
568
_parser_multiline_state = MULTILINE_ENCODED;
555
_parser_in_encoded_multiline =
556
unencoding_result.isAnotherLineRequired();
569
557
}
570
558
571
559
// convert 8-bit ASCII charset to US-ASCII
583
571
throw new ParseException( R.string.error_vcf_charset );
584
572
}
585
573
586
// for some entries that have semicolon-separated value parts,
587
// check to see if the value ends in an escape character, which
588
// indicates that we have a multi-line value
589
if( ( name_param_parts[ 0 ].equals( "N" ) ||
590
name_param_parts[ 0 ].equals( "ORG" ) ||
591
name_param_parts[ 0 ].equals( "ADR" ) ) &&
592
doesStringEndInAnEscapeChar( string_value ) )
593
{
594
_parser_multiline_state = MULTILINE_ESCAPED;
595
string_value = string_value.substring( 0,
596
string_value.length() - 1 );
597
}
598
599
574
// now we know whether we're in an encoding multi-line,
600
575
// determine if we're in a v3 folded multi-line or not
601
if( _parser_multiline_state == MULTILINE_NONE &&
602
_version.equals( "3.0" ) && next_line_looks_folded )
576
_parser_in_folded_multiline = !_parser_in_encoded_multiline &&
577
_version.equals( "3.0" ) && next_line_looks_folded;
578
579
// handle multi-line requests
580
if( _parser_in_encoded_multiline ||
581
_parser_in_folded_multiline )
603
582
{
604
_parser_multiline_state = MULTILINE_FOLDED;
605
}
606
607
// handle multi-lines by buffering them and parsing them when we
608
// are processing the last line in a multi-line sequence
609
if( _parser_multiline_state != MULTILINE_NONE ) {
610
583
_parser_buffered_value_so_far += string_value;
611
584
return;
612
585
}
586
587
// add on buffered multi-line content
613
588
String complete_value =
614
( _parser_buffered_value_so_far + string_value ).trim();
589
_parser_buffered_value_so_far + string_value;
615
590
616
591
// ignore empty values
617
592
if( complete_value.length() < 1 ) return;
623
598
parseFN( name_param_parts, complete_value );
624
599
else if( name_param_parts[ 0 ].equals( "ORG" ) )
625
600
parseORG( name_param_parts, complete_value );
626
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
627
parseTITLE( name_param_parts, complete_value );
628
601
else if( name_param_parts[ 0 ].equals( "TEL" ) )
629
602
parseTEL( name_param_parts, complete_value );
630
603
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
631
604
parseEMAIL( name_param_parts, complete_value );
632
else if( name_param_parts[ 0 ].equals( "ADR" ) )
633
parseADR( name_param_parts, complete_value );
634
}
635
}
636
637
private boolean doesStringEndInAnEscapeChar( String string )
638
{
639
// count the number of backslashes at the end of the string
640
int count = 0;
641
for( int a = string.length() - 1; a >= 0; a-- )
642
if( string.charAt( a ) == '\\' )
643
count++;
644
else
645
break;
646
647
// if there are an even number of backslashes then the final one
648
// doesn't count
649
return ( count & 1 ) == 1;
650
}
651
652
private String[] splitValueBySemicolon( String value )
653
{
654
// split string in to parts by semicolon
655
ArrayList< String > parts = new ArrayList< String >(
656
Arrays.asList( value.split( ";" ) ) );
657
658
// go through parts
659
for( int a = 0; a < parts.size(); a++ )
660
{
661
String str = parts.get( a );
662
663
// look for parts that end in an escape character, but ignore
664
// the final part. We've already detected escape chars at the
665
// end of the final part in parseLine() and handled multi-lines
666
// accordingly.
667
if( a < parts.size() - 1 &&
668
doesStringEndInAnEscapeChar( str ) )
669
{
670
// join the next part to this part and remove the next part
671
parts.set( a, str.substring( 0, str.length() - 1 ) +
672
';' + parts.get( a + 1 ) );
673
parts.remove( a + 1 );
674
675
// re-visit this part
676
a--;
677
continue;
678
}
679
680
// trim and replace string
681
str = str.trim();
682
parts.set( a, str );
683
}
684
685
String[] ret = new String[ parts.size() ];
686
return parts.toArray( ret );
605
}
687
606
}
688
607
689
608
private void parseN( String[] params, String value )
609
throws ParseException, SkipContactException,
610
AbortImportException
690
611
{
691
612
// already got a better name?
692
613
if( _name_level >= NAMELEVEL_N ) return;
693
614
694
615
// get name parts
695
String[] name_parts = splitValueBySemicolon( value );
616
String[] name_parts = value.split( ";" );
617
for( int i = 0; i < name_parts.length; i++ )
618
name_parts[ i ] = name_parts[ i ].trim();
696
619
697
620
// build name
698
621
value = "";
704
627
// set name
705
628
setName( value );
706
629
_name_level = NAMELEVEL_N;
630
631
// check now to see if we need to import this contact (to avoid
632
// parsing the rest of the vCard unnecessarily)
633
if( !isImportRequired( getName() ) )
634
throw new SkipContactException();
707
635
}
708
636
709
637
private void parseFN( String[] params, String value )
638
throws ParseException, SkipContactException
710
639
{
711
640
// already got a better name?
712
641
if( _name_level >= NAMELEVEL_FN ) return;
717
646
}
718
647
719
648
private void parseORG( String[] params, String value )
649
throws ParseException, SkipContactException
720
650
{
651
// already got a better name?
652
if( _name_level >= NAMELEVEL_ORG ) return;
653
721
654
// get org parts
722
String[] org_parts = splitValueBySemicolon( value );
723
if( org_parts == null || org_parts.length < 1 ) return;
724
725
// build organisation name
726
StringBuilder builder = new StringBuilder(
727
String.valueOf( org_parts[ 0 ] ) );
728
for( int a = 1; a < org_parts.length; a++ )
729
builder.append( ", " ).append( org_parts[ a ] );
730
String organisation = builder.toString();
731
732
// set organisation name (using a title we've previously found)
733
addOrganisation( organisation, _cached_title, true );
734
735
// if we've not previously found a title, store this organisation
736
// name (we'll need it when we find a title to update the
737
// organisation, by name), else if we *have* previously found a
738
// title, clear it (since we just used it)
739
if( _cached_title == null )
740
_cached_organisation = organisation;
741
else
742
_cached_title = null;
743
}
744
745
private void parseTITLE( String[] params, String value )
746
{
747
// if we previously had an organisation, look it up and append this
748
// title to it
749
if( _cached_organisation != null && hasOrganisations() ) {
750
HashMap< String, ExtraDetail > datas = getOrganisations();
751
ExtraDetail detail = datas.get( _cached_organisation );
752
if( detail != null )
753
detail.setExtra( value );
754
}
755
756
// same as when handling organisation, if we've not previously found
757
// an organisation we store this title, else we clear it (since we
758
// just appended this title to it)
759
if( _cached_organisation == null )
760
_cached_title = value;
761
else
762
_cached_organisation = null;
655
String[] org_parts = value.split( ";" );
656
for( int i = 0; i < org_parts.length; i++ )
657
org_parts[ i ] = org_parts[ i ].trim();
658
659
// build name
660
if( org_parts.length > 1 && org_parts[ 0 ].length() == 0 )
661
value = org_parts[ 1 ];
662
else
663
value = org_parts[ 0 ];
664
665
// set name
666
setName( value );
667
_name_level = NAMELEVEL_ORG;
763
668
}
764
669
765
670
private void parseTEL( String[] params, String value )
671
throws ParseException
766
672
{
767
673
if( value.length() == 0 ) return;
768
674
771
677
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
772
678
773
679
// here's the logic...
774
boolean is_preferred = types.contains( "PREF" );
775
int type;
680
boolean preferred = types.contains( "PREF" );
681
int type = PhonesColumns.TYPE_MOBILE;
682
if( types.contains( "VOICE" ) )
683
if( types.contains( "WORK" ) )
684
type = PhonesColumns.TYPE_WORK;
685
else
686
type = PhonesColumns.TYPE_HOME;
687
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
688
type = PhonesColumns.TYPE_MOBILE;
776
689
if( types.contains( "FAX" ) )
777
690
if( types.contains( "HOME" ) )
778
691
type = PhonesColumns.TYPE_FAX_HOME;
779
692
else
780
693
type = PhonesColumns.TYPE_FAX_WORK;
781
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
782
type = PhonesColumns.TYPE_MOBILE;
783
else if( types.contains( "PAGER" ) )
694
if( types.contains( "PAGER" ) )
784
695
type = PhonesColumns.TYPE_PAGER;
785
else if( types.contains( "WORK" ) )
786
type = PhonesColumns.TYPE_WORK;
787
else
788
type = PhonesColumns.TYPE_HOME;
789
696
790
697
// add phone number
791
addNumber( value, type, is_preferred );
698
addPhone( value, type, preferred );
792
699
}
793
700
794
701
public void parseEMAIL( String[] params, String value )
702
throws ParseException
795
703
{
796
704
if( value.length() == 0 ) return;
797
705
798
706
Set< String > types = extractTypes( params, Arrays.asList(
799
707
"PREF", "WORK", "HOME", "INTERNET" ) );
800
708
801
// add email address
802
boolean is_preferred = types.contains( "PREF" );
803
int type;
804
if( types.contains( "WORK" ) )
805
type = Contacts.ContactMethods.TYPE_WORK;
806
else
807
type = Contacts.ContactMethods.TYPE_HOME;
808
809
addEmail( value, type, is_preferred );
810
}
811
812
private void parseADR( String[] params, String value )
813
{
814
// get address parts
815
String[] adr_parts = splitValueBySemicolon( value );
816
817
// build address
818
value = "";
819
for( int a = 0; a < adr_parts.length; a++ ) {
820
if( value.length() > 0 ) value += "\n";
821
value += adr_parts[ a ].trim();
822
}
823
824
Set< String > types = extractTypes( params, Arrays.asList(
825
"PREF", "WORK", "HOME", "INTERNET" ) );
826
827
// add address
828
int type;
829
if( types.contains( "WORK" ) )
830
type = Contacts.ContactMethods.TYPE_WORK;
831
else
832
type = Contacts.ContactMethods.TYPE_HOME;
833
834
addAddress( value, type );
709
// here's the logic...
710
boolean preferred = types.contains( "PREF" );
711
if( types.contains( "WORK" ) )
712
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
713
else
714
addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
835
715
}
836
716
837
717
public void finaliseParsing()
842
722
if( _version == null && _buffers != null )
843
723
throw new ParseException( R.string.error_vcf_malformed );
844
724
845
// check if we should import this contact
846
try {
847
if( !isImportRequired( this ) )
848
throw new SkipContactException();
849
}
850
catch( ContactNeedsMoreInfoException e ) {
851
throw new ParseException( R.string.error_vcf_notenoughinfo );
852
}
725
// missing name properties?
726
if( _name_level == NAMELEVEL_NONE )
727
throw new ParseException( R.string.error_vcf_noname );
728
729
// check if we should import this one? If we've already got an 'N'-
730
// type name, this will already have been done by parseN() so we
731
// mustn't do this here (or it could prompt twice!)
732
if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
733
throw new SkipContactException();
853
734
}
854
735
855
736
private String checkParam( String[] params, String name )
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480