4
* Copyright (C) 2009 to 2011 Tim Marston <tim@ed.am>
4
* Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
6
6
* This file is part of the Import Contacts program (hereafter referred
7
* to as "this program"). For more information, see
7
* to as "this program"). For more information, see
8
8
* http://ed.am/dev/android/import-contacts
10
10
* This program is free software: you can redistribute it and/or modify
45
45
import java.util.regex.Matcher;
46
46
import java.util.regex.Pattern;
48
import android.annotation.SuppressLint;
49
48
import android.content.SharedPreferences;
49
import android.os.Environment;
51
51
public class VcardImporter extends Importer
71
70
File[] files = null;
73
// check SD card is mounted
74
String state = Environment.getExternalStorageState();
75
if( !Environment.MEDIA_MOUNTED.equals( state ) &&
76
!Environment.MEDIA_MOUNTED_READ_ONLY.equals( state ) )
78
showError( R.string.error_nosdcard );
75
String path = "/sdcard" + prefs.getString( "location", "/" );
76
File file = new File( path );
82
File file = new File( Environment.getExternalStorageDirectory(),
83
prefs.getString( "location", "/" ) );
77
84
if( !file.exists() )
78
85
showError( R.string.error_locationnotfound );
84
91
class VCardFilter implements FilenameFilter {
85
92
public boolean accept( File dir, String name ) {
86
return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
93
return name.toLowerCase( Locale.ENGLISH )
89
97
files = file.listFiles( new VCardFilter() );
133
141
boolean in_vcard = false;
134
142
while( ( line = reader.readLine() ) != null )
137
146
// look for vcard beginning
138
if( line.matches( "^BEGIN:VCARD" ) ) {
147
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
151
// check for vMsg files
152
else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
153
showError( getText( R.string.error_vcf_vmsgfile )
143
else if( line.matches( "^END:VCARD" ) )
157
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
144
158
in_vcard = false;
148
163
catch( FileNotFoundException e ) {
170
185
FileInputStream istream = new FileInputStream( file );
171
186
byte[] content = new byte[ (int)file.length() ];
172
187
istream.read( content );
176
191
importVCardFileContent( content, file.getName() );
193
catch( OutOfMemoryError e ) {
194
showError( R.string.error_outofmemory );
178
196
catch( FileNotFoundException e ) {
179
197
showError( getText( R.string.error_filenotfound ) +
180
198
file.getName() );
193
211
ContentLineIterator cli = new ContentLineIterator( content );
194
212
while( cli.hasNext() )
196
ByteBuffer buffer = cli.next();
214
ContentLine content_line = cli.next();
198
// get a US-ASCII version of the line for processing
201
line = new String( buffer.array(), buffer.position(),
202
buffer.limit() - buffer.position(), "US-ASCII" );
204
catch( UnsupportedEncodingException e ) {
205
// we know US-ASCII is supported, so appease the compiler...
216
// get a US-ASCII version of the string, for processing
217
String line = content_line.getUsAsciiLine();
209
219
if( vcard == null ) {
210
220
// look for vcard beginning
211
if( line.matches( "^BEGIN:VCARD" ) ) {
221
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
212
222
setProgress( _progress++ );
213
223
vcard = new Vcard();
214
224
vcard_start_line = cli.getLineNumber();
226
236
importContact( vcard );
228
238
catch( Vcard.ParseException e ) {
230
240
getText( R.string.error_vcf_parse ).toString()
232
242
getText( R.string.error_vcf_parse_line ).toString()
233
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
235
finish( ACTION_ABORT );
243
+ cli.getLineNumber() + ":\n" + e.getMessage() );
240
246
catch( ContactData.ContactNotIdentifiableException e ) {
242
248
getText( R.string.error_vcf_parse ).toString()
244
250
getText( R.string.error_vcf_parse_line ).toString()
245
251
+ vcard_start_line + ":\n" + getText(
246
R.string.error_vcf_notenoughinfo ).toString()
249
finish( ACTION_ABORT );
252
R.string.error_vcf_notenoughinfo ).toString() );
255
256
// discard this vcard
260
261
// try giving the line to the vcard
262
vcard.parseLine( buffer, line,
263
cli.doesNextLineLookFolded() );
263
vcard.parseLine( content_line );
265
265
catch( Vcard.ParseException e ) {
268
268
getText( R.string.error_vcf_parse ).toString()
270
270
getText( R.string.error_vcf_parse_line ).toString()
271
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
273
finish( ACTION_ABORT );
271
+ cli.getLineNumber() + "\n" + e.getMessage() );
276
// although we're continuing, we still need to abort
277
// this vCard. Further lines will be ignored until we
273
// Although we're continuing, we still need to abort
274
// this vCard. Further lines will be ignored until we
278
275
// get to another BEGIN:VCARD line.
281
278
catch( Vcard.SkipImportException e ) {
283
// abort this vCard. Further lines will be ignored until
280
// Abort this vCard. Further lines will be ignored until
284
281
// we get to another BEGIN:VCARD line.
292
class ContentLineIterator implements Iterator< ByteBuffer >
291
private ByteBuffer _buffer;
292
private boolean _folded_next;
293
private String _line;
295
public ContentLine( ByteBuffer buffer, boolean folded_next )
298
_folded_next = folded_next;
302
public ByteBuffer getBuffer()
307
public boolean doesNextLineLookFolded()
312
public String getUsAsciiLine()
314
// generated line and cache it
315
if( _line == null ) {
317
_line = new String( _buffer.array(), _buffer.position(),
318
_buffer.limit() - _buffer.position(), "US-ASCII" );
320
catch( UnsupportedEncodingException e ) {
321
// we know US-ASCII *is* supported, so appease the
326
// return cached line
331
class ContentLineIterator implements Iterator< ContentLine >
294
333
protected byte[] _content = null;
295
334
protected int _pos = 0;
320
359
_pos > initial_pos )? _pos - 1 : _pos;
323
return ByteBuffer.wrap( _content, initial_pos,
362
return new ContentLine(
363
ByteBuffer.wrap( _content, initial_pos,
365
doesNextLineLookFolded() );
327
368
// we didn't find one, but were there bytes left?
332
return ByteBuffer.wrap( _content, initial_pos,
373
return new ContentLine(
374
ByteBuffer.wrap( _content, initial_pos,
376
doesNextLineLookFolded() );
348
391
* onto the end of this one?
351
public boolean doesNextLineLookFolded()
394
private boolean doesNextLineLookFolded()
353
396
return _pos > 0 && _pos < _content.length &&
354
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
397
_content[ _pos - 1 ] == '\n' &&
398
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
357
401
public int getLineNumber()
369
413
private final static int MULTILINE_NONE = 0;
370
414
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
371
415
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
372
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
416
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
374
418
private String _version = null;
375
private Vector< ByteBuffer > _buffers = null;
419
private Vector< ContentLine > _content_lines = null;
376
420
private int _name_level = NAMELEVEL_NONE;
377
421
private int _parser_multiline_state = MULTILINE_NONE;
378
422
private String _parser_current_name_and_params = null;
421
465
@SuppressWarnings("serial")
422
466
protected class SkipImportException extends Exception { }
424
private String extractCollonPartFromLine( ByteBuffer buffer,
425
String line, boolean former )
468
private String extractCollonPartFromLine( ContentLine content_line,
429
// get a US-ASCII version of the line for processing, unless we were
433
line = new String( buffer.array(), buffer.position(),
434
buffer.limit() - buffer.position(), "US-ASCII" );
436
catch( UnsupportedEncodingException e ) {
437
// we know US-ASCII is supported, so appease the compiler...
442
471
// split line into name and value parts and check to make sure we
443
472
// only got 2 parts and that the first part is not zero in length
444
String[] parts = line.split( ":", 2 );
473
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
445
474
if( parts.length == 2 && parts[ 0 ].length() > 0 )
446
ret = parts[ former? 0 : 1 ];
451
private String extractNameAndParamsFromLine( ByteBuffer buffer,
454
return extractCollonPartFromLine( buffer, line, true );
457
private String extractValueFromLine( ByteBuffer buffer, String line )
459
return extractCollonPartFromLine( buffer, line, false );
462
public void parseLine( ByteBuffer buffer, String line,
463
boolean next_line_looks_folded )
475
return parts[ former? 0 : 1 ].trim();
480
private String extractNameAndParamsFromLine( ContentLine content_line )
482
return extractCollonPartFromLine( content_line, true );
485
private String extractValueFromLine( ContentLine content_line )
487
return extractCollonPartFromLine( content_line, false );
490
public void parseLine( ContentLine content_line )
464
491
throws ParseException, SkipImportException,
465
492
AbortImportException
470
497
// tentatively get name and params from line
471
498
String name_and_params =
472
extractNameAndParamsFromLine( buffer, line );
499
extractNameAndParamsFromLine( content_line );
474
501
// is it a version line?
475
502
if( name_and_params != null &&
476
name_and_params.equals( "VERSION" ) )
503
name_and_params.equalsIgnoreCase( "VERSION" ) )
479
String value = extractValueFromLine( buffer, line );
480
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
506
String value = extractValueFromLine( content_line );
507
if( value == null || (
508
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
481
510
throw new ParseException( R.string.error_vcf_version );
482
512
_version = value;
484
514
// parse any buffers we've been accumulating while we waited
486
if( _buffers != null )
487
for( int i = 0; i < _buffers.size(); i++ )
488
parseLine( _buffers.get( i ), null,
489
i + 1 < _buffers.size() &&
490
_buffers.get( i + 1 ).hasRemaining() &&
491
_buffers.get( i + 1 ).get(
492
_buffers.get( i + 1 ).position() ) == ' ' );
516
if( _content_lines != null )
517
for( int i = 0; i < _content_lines.size(); i++ )
518
parseLine( _content_lines.get( i ) );
519
_content_lines = null;
497
523
// no, so stash this line till we get a version
498
if( _buffers == null )
499
_buffers = new Vector< ByteBuffer >();
500
_buffers.add( buffer );
524
if( _content_lines == null )
525
_content_lines = new Vector< ContentLine >();
526
_content_lines.add( content_line );
505
531
// name and params and the position in the buffer where the
506
// "value" part of the line start
532
// "value" part of the line starts
507
533
String name_and_params;
516
542
// skip some initial line characters, depending on the type
517
543
// of multi-line we're handling
518
pos = buffer.position();
544
pos = content_line.getBuffer().position();
519
545
switch( _parser_multiline_state )
521
547
case MULTILINE_FOLDED:
524
550
case MULTILINE_ENCODED:
525
while( pos < buffer.limit() && (
526
buffer.get( pos ) == ' ' ||
527
buffer.get( pos ) == '\t' ) )
551
while( pos < content_line.getBuffer().limit() && (
552
content_line.getBuffer().get( pos ) == ' ' ||
553
content_line.getBuffer().get( pos ) == '\t' ) )
569
if( content_line.getUsAsciiLine().trim().length() == 0 )
542
572
// get name and params from line, and since we're not
543
573
// parsing a subsequent line in a multi-line, this should
544
574
// not fail, or it's an error
545
575
name_and_params =
546
extractNameAndParamsFromLine( buffer, line );
576
extractNameAndParamsFromLine( content_line );
547
577
if( name_and_params == null )
548
578
throw new ParseException(
549
579
R.string.error_vcf_malformed );
551
581
// calculate how many chars to skip from beginning of line
552
582
// so we skip the property "name:" part
553
pos = buffer.position() + name_and_params.length() + 1;
583
pos = content_line.getBuffer().position() +
584
name_and_params.length() + 1;
555
586
// reset the saved multi-line state
556
587
_parser_current_name_and_params = name_and_params;
560
591
// get value from buffer, as raw bytes
561
592
ByteBuffer value;
562
value = ByteBuffer.wrap( buffer.array(), pos,
563
buffer.limit() - pos );
593
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
594
content_line.getBuffer().limit() - pos );
565
596
// get parameter parts
566
597
String[] name_param_parts = name_and_params.split( ";", -1 );
573
604
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
575
606
boolean is_interesting_field =
576
interesting_fields.contains( name_param_parts[ 0 ] );
607
interesting_fields.contains(
608
name_param_parts[ 0 ].toUpperCase( Locale.ENGLISH ) );
578
610
// parse encoding parameter
579
611
String encoding = checkParam( name_param_parts, "ENCODING" );
580
612
if( encoding != null )
581
encoding = encoding.toUpperCase( Locale.US );
613
encoding = encoding.toUpperCase( Locale.ENGLISH );
582
614
if( is_interesting_field && encoding != null &&
583
!encoding.equals( "8BIT" ) &&
584
!encoding.equals( "QUOTED-PRINTABLE" ) )
585
//&& !encoding.equals( "BASE64" ) )
615
!encoding.equalsIgnoreCase( "8BIT" ) &&
616
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
617
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
587
619
throw new ParseException( R.string.error_vcf_encoding );
590
622
// parse charset parameter
591
623
String charset = checkParam( name_param_parts, "CHARSET" );
592
624
if( charset != null )
593
charset = charset.toUpperCase( Locale.US );
625
charset = charset.toUpperCase( Locale.ENGLISH );
594
626
if( charset != null &&
595
!charset.equals( "US-ASCII" ) &&
596
!charset.equals( "ASCII" ) &&
597
!charset.equals( "UTF-8" ) )
627
!charset.equalsIgnoreCase( "US-ASCII" ) &&
628
!charset.equalsIgnoreCase( "ASCII" ) &&
629
!charset.equalsIgnoreCase( "UTF-8" ) )
599
631
throw new ParseException( R.string.error_vcf_charset );
602
634
// do unencoding (or default to a fake unencoding result with
603
635
// the raw string)
604
636
UnencodeResult unencoding_result = null;
605
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
637
if( encoding != null &&
638
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
606
640
unencoding_result = unencodeQuotedPrintable( value );
607
// else if( encoding != null && encoding.equals( "BASE64" ) )
642
// else if( encoding != null &&
643
// encoding.equalsIgnoreCase( "BASE64" ) )
608
645
// unencoding_result = unencodeBase64( props[ 1 ], charset );
609
647
if( unencoding_result != null ) {
610
648
value = unencoding_result.getBuffer();
611
649
if( unencoding_result.isAnotherLineRequired() )
616
654
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
617
655
if( ( charset == null && _version.equals( "2.1" ) ) ||
618
656
( charset != null && (
619
charset.equals( "ASCII" ) ||
620
charset.equals( "US-ASCII" ) ) ) )
657
charset.equalsIgnoreCase( "ASCII" ) ||
658
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
622
660
value = transcodeAsciiToUtf8( value );
634
672
// for some entries that have semicolon-separated value parts,
635
673
// check to see if the value ends in an escape character, which
636
674
// indicates that we have a multi-line value
637
if( ( name_param_parts[ 0 ].equals( "N" ) ||
638
name_param_parts[ 0 ].equals( "ORG" ) ||
639
name_param_parts[ 0 ].equals( "ADR" ) ) &&
675
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
676
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
677
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
640
678
doesStringEndInAnEscapeChar( string_value ) )
642
680
_parser_multiline_state = MULTILINE_ESCAPED;
644
682
string_value.length() - 1 );
647
// now we know whether we're in an encoding multi-line,
648
// determine if we're in a v3 folded multi-line or not
685
// if we know we're not in an encoding-based multi-line, check
686
// to see if we're in a folded multi-line
649
687
if( _parser_multiline_state == MULTILINE_NONE &&
650
_version.equals( "3.0" ) && next_line_looks_folded )
688
content_line.doesNextLineLookFolded() )
652
690
_parser_multiline_state = MULTILINE_FOLDED;
665
703
if( complete_value.length() < 1 ) return;
667
705
// parse some properties
668
if( name_param_parts[ 0 ].equals( "N" ) )
706
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
669
707
parseN( name_param_parts, complete_value );
670
else if( name_param_parts[ 0 ].equals( "FN" ) )
708
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
671
709
parseFN( name_param_parts, complete_value );
672
else if( name_param_parts[ 0 ].equals( "ORG" ) )
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
673
711
parseORG( name_param_parts, complete_value );
674
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
675
713
parseTITLE( name_param_parts, complete_value );
676
else if( name_param_parts[ 0 ].equals( "TEL" ) )
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
677
715
parseTEL( name_param_parts, complete_value );
678
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
679
717
parseEMAIL( name_param_parts, complete_value );
680
else if( name_param_parts[ 0 ].equals( "ADR" ) )
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
681
719
parseADR( name_param_parts, complete_value );
682
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
683
721
parseLABEL( name_param_parts, complete_value );
684
else if( name_param_parts[ 0 ].equals( "NOTE" ) )
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
685
723
parseNOTE( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
725
parseBDAY( name_param_parts, complete_value );
713
753
String str = parts.get( a );
715
// look for parts that end in an escape character, but ignore
716
// the final part. We've already detected escape chars at the
755
// Look for parts that end in an escape character, but ignore
756
// the final part. We've already detected escape chars at the
717
757
// end of the final part in parseLine() and handled multi-lines
719
759
if( a < parts.size() - 1 &&
809
849
for( int b = 0; b < name_part_parts.length; b++ )
810
850
if( name_part_parts[ b ].length() > 0 )
812
if( value.length() == 0 ) value += " ";
852
if( value.length() > 0 ) value += " ";
813
853
value += name_part_parts[ b ];
934
974
for( int a = 0; a < adr_parts.length; a++ )
935
975
if( adr_parts[ a ].length() > 0 )
937
// split this part in to it's comma-separated bits
938
String[] adr_part_parts =
939
splitValueByCharacter( adr_parts[ a ], ',' );
940
for( int b = 0; b < adr_part_parts.length; b++ )
941
if( adr_part_parts[ b ].length() > 0 )
943
if( value.length() > 0 ) value += "\n";
944
value += adr_part_parts[ b ];
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 ];
948
999
Set< String > types = extractTypes( params, Arrays.asList(
978
1029
addNote( unescapeValue( value ) );
1032
private void parseBDAY( String[] params, String value )
1034
setBirthday( value );
981
1037
public void finaliseVcard()
982
1038
throws ParseException, ContactNotIdentifiableException
984
1040
// missing version (and data is present)
985
if( _version == null && _buffers != null )
1041
if( _version == null && _content_lines != null )
986
1042
throw new ParseException( R.string.error_vcf_malformed );
988
1044
// finalise the parent class
1013
1071
HashSet< String > ret = new HashSet< String >();
1015
1073
Pattern p = Pattern.compile(
1016
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1074
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1075
Pattern.CASE_INSENSITIVE );
1017
1076
for( int i = 0; i < params.length; i++ ) {
1018
1077
Matcher m = p.matcher( params[ i ] );
1019
1078
if( m.matches() )
1027
* Amongst the params, return any type values present. For v2.1 vCards,
1028
* those types are just parameters. For v3.0, they are prefixed with
1029
* "TYPE=". There may also be multiple type parameters.
1031
* @param a list of type values to look for
1086
* Amongst the params, return any type values present. For v2.1 vCards,
1087
* those types are just parameters. For v3.0, they are prefixed with
1088
* "TYPE=". There may also be multiple type parameters.
1090
* @param params an array of params to look for types in
1091
* @param valid_types an list of upper-case type values to look for
1032
1092
* @return a set of present type values
1034
1094
private Set< String > extractTypes( String[] params,
1040
1100
String type_params[] = checkParams( params, "TYPE" );
1041
1101
for( int a = 0; a < type_params.length; a++ )
1043
// check for a comma-separated list of types (why? this isn't in
1103
// check for a comma-separated list of types (why? I don't think
1104
// this is in the specs!)
1045
1105
String[] parts = type_params[ a ].split( "," );
1046
for( int i = 0; i < parts.length; i++ )
1047
if( valid_types.contains( parts[ i ] ) )
1048
types.add( parts[ i ] );
1106
for( int i = 0; i < parts.length; i++ ) {
1107
String ucpart = parts[ i ].toUpperCase( Locale.ENGLISH );
1108
if( valid_types.contains( ucpart ) )
1109
types.add( ucpart );
1051
1113
// get 2.1-style type param
1052
1114
if( _version.equals( "2.1" ) ) {
1053
for( int i = 1; i < params.length; i++ )
1054
if( valid_types.contains( params[ i ] ) )
1055
types.add( params[ i ] );
1115
for( int i = 1; i < params.length; i++ ) {
1116
String ucparam = params[ i ].toUpperCase( Locale.ENGLISH );
1117
if( valid_types.contains( ucparam ) )
1118
types.add( ucparam );
1080
1144
else if( ch == '=' && i == in.limit() - 1 )
1082
1146
// we found a '=' at the end of a line signifying a multi-
1083
// line string, so we don't add it.
1147
// line string, so we don't add it
1084
1148
another = true;