4
* Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
4
* Copyright (C) 2009 to 2011 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;
48
49
import android.content.SharedPreferences;
49
import android.os.Environment;
51
51
public class VcardImporter extends Importer
70
71
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 );
82
File file = new File( Environment.getExternalStorageDirectory(),
83
prefs.getString( "location", "/" ) );
75
String path = "/sdcard" + prefs.getString( "location", "/" );
76
File file = new File( path );
84
77
if( !file.exists() )
85
78
showError( R.string.error_locationnotfound );
91
84
class VCardFilter implements FilenameFilter {
92
85
public boolean accept( File dir, String name ) {
93
return name.toLowerCase( Locale.ENGLISH )
86
return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
97
89
files = file.listFiles( new VCardFilter() );
141
133
boolean in_vcard = false;
142
134
while( ( line = reader.readLine() ) != null )
146
137
// look for vcard beginning
147
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
138
if( line.matches( "^BEGIN:VCARD" ) ) {
151
// check for vMsg files
152
else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
153
showError( getText( R.string.error_vcf_vmsgfile )
157
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
143
else if( line.matches( "^END:VCARD" ) )
158
144
in_vcard = false;
163
148
catch( FileNotFoundException e ) {
185
170
FileInputStream istream = new FileInputStream( file );
186
171
byte[] content = new byte[ (int)file.length() ];
187
172
istream.read( content );
191
176
importVCardFileContent( content, file.getName() );
193
catch( OutOfMemoryError e ) {
194
showError( R.string.error_outofmemory );
196
178
catch( FileNotFoundException e ) {
197
179
showError( getText( R.string.error_filenotfound ) +
198
180
file.getName() );
211
193
ContentLineIterator cli = new ContentLineIterator( content );
212
194
while( cli.hasNext() )
214
ContentLine content_line = cli.next();
196
ByteBuffer buffer = cli.next();
216
// get a US-ASCII version of the string, for processing
217
String line = content_line.getUsAsciiLine();
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...
219
209
if( vcard == null ) {
220
210
// look for vcard beginning
221
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
211
if( line.matches( "^BEGIN:VCARD" ) ) {
222
212
setProgress( _progress++ );
223
213
vcard = new Vcard();
224
214
vcard_start_line = cli.getLineNumber();
236
226
importContact( vcard );
238
228
catch( Vcard.ParseException e ) {
240
230
getText( R.string.error_vcf_parse ).toString()
242
232
getText( R.string.error_vcf_parse_line ).toString()
243
+ cli.getLineNumber() + ":\n" + e.getMessage() );
233
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
235
finish( ACTION_ABORT );
246
240
catch( ContactData.ContactNotIdentifiableException e ) {
248
242
getText( R.string.error_vcf_parse ).toString()
250
244
getText( R.string.error_vcf_parse_line ).toString()
251
245
+ vcard_start_line + ":\n" + getText(
252
R.string.error_vcf_notenoughinfo ).toString() );
246
R.string.error_vcf_notenoughinfo ).toString()
249
finish( ACTION_ABORT );
256
255
// discard this vcard
261
260
// try giving the line to the vcard
263
vcard.parseLine( content_line );
262
vcard.parseLine( buffer, line,
263
cli.doesNextLineLookFolded() );
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() );
271
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
273
finish( ACTION_ABORT );
273
// Although we're continuing, we still need to abort
274
// this vCard. Further lines will be ignored until we
276
// although we're continuing, we still need to abort
277
// this vCard. Further lines will be ignored until we
275
278
// get to another BEGIN:VCARD line.
278
281
catch( Vcard.SkipImportException e ) {
280
// Abort this vCard. Further lines will be ignored until
283
// abort this vCard. Further lines will be ignored until
281
284
// we get to another BEGIN:VCARD line.
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 >
292
class ContentLineIterator implements Iterator< ByteBuffer >
333
294
protected byte[] _content = null;
334
295
protected int _pos = 0;
359
320
_pos > initial_pos )? _pos - 1 : _pos;
362
return new ContentLine(
363
ByteBuffer.wrap( _content, initial_pos,
365
doesNextLineLookFolded() );
323
return ByteBuffer.wrap( _content, initial_pos,
368
327
// we didn't find one, but were there bytes left?
373
return new ContentLine(
374
ByteBuffer.wrap( _content, initial_pos,
376
doesNextLineLookFolded() );
332
return ByteBuffer.wrap( _content, initial_pos,
391
348
* onto the end of this one?
394
private boolean doesNextLineLookFolded()
351
public boolean doesNextLineLookFolded()
396
353
return _pos > 0 && _pos < _content.length &&
397
_content[ _pos - 1 ] == '\n' &&
398
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
354
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
401
357
public int getLineNumber()
413
369
private final static int MULTILINE_NONE = 0;
414
370
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
415
371
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
416
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
372
private final static int MULTILINE_FOLDED = 3; // v3.0 folding
418
374
private String _version = null;
419
private Vector< ContentLine > _content_lines = null;
375
private Vector< ByteBuffer > _buffers = null;
420
376
private int _name_level = NAMELEVEL_NONE;
421
377
private int _parser_multiline_state = MULTILINE_NONE;
422
378
private String _parser_current_name_and_params = null;
465
421
@SuppressWarnings("serial")
466
422
protected class SkipImportException extends Exception { }
468
private String extractCollonPartFromLine( ContentLine content_line,
424
private String extractCollonPartFromLine( ByteBuffer buffer,
425
String line, boolean former )
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...
471
442
// split line into name and value parts and check to make sure we
472
443
// only got 2 parts and that the first part is not zero in length
473
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
444
String[] parts = line.split( ":", 2 );
474
445
if( parts.length == 2 && parts[ 0 ].length() > 0 )
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 )
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 )
491
464
throws ParseException, SkipImportException,
492
465
AbortImportException
497
470
// tentatively get name and params from line
498
471
String name_and_params =
499
extractNameAndParamsFromLine( content_line );
472
extractNameAndParamsFromLine( buffer, line );
501
474
// is it a version line?
502
475
if( name_and_params != null &&
503
name_and_params.equalsIgnoreCase( "VERSION" ) )
476
name_and_params.equals( "VERSION" ) )
506
String value = extractValueFromLine( content_line );
507
if( value == null || (
508
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
479
String value = extractValueFromLine( buffer, line );
480
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
510
481
throw new ParseException( R.string.error_vcf_version );
512
482
_version = value;
514
484
// parse any buffers we've been accumulating while we waited
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;
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() ) == ' ' );
523
497
// no, so stash this line till we get a version
524
if( _content_lines == null )
525
_content_lines = new Vector< ContentLine >();
526
_content_lines.add( content_line );
498
if( _buffers == null )
499
_buffers = new Vector< ByteBuffer >();
500
_buffers.add( buffer );
531
505
// name and params and the position in the buffer where the
532
// "value" part of the line starts
506
// "value" part of the line start
533
507
String name_and_params;
542
516
// skip some initial line characters, depending on the type
543
517
// of multi-line we're handling
544
pos = content_line.getBuffer().position();
518
pos = buffer.position();
545
519
switch( _parser_multiline_state )
547
521
case MULTILINE_FOLDED:
550
524
case MULTILINE_ENCODED:
551
while( pos < content_line.getBuffer().limit() && (
552
content_line.getBuffer().get( pos ) == ' ' ||
553
content_line.getBuffer().get( pos ) == '\t' ) )
525
while( pos < buffer.limit() && (
526
buffer.get( pos ) == ' ' ||
527
buffer.get( pos ) == '\t' ) )
569
if( content_line.getUsAsciiLine().trim().length() == 0 )
572
542
// get name and params from line, and since we're not
573
543
// parsing a subsequent line in a multi-line, this should
574
544
// not fail, or it's an error
575
545
name_and_params =
576
extractNameAndParamsFromLine( content_line );
546
extractNameAndParamsFromLine( buffer, line );
577
547
if( name_and_params == null )
578
548
throw new ParseException(
579
549
R.string.error_vcf_malformed );
581
551
// calculate how many chars to skip from beginning of line
582
552
// so we skip the property "name:" part
583
pos = content_line.getBuffer().position() +
584
name_and_params.length() + 1;
553
pos = buffer.position() + name_and_params.length() + 1;
586
555
// reset the saved multi-line state
587
556
_parser_current_name_and_params = name_and_params;
591
560
// get value from buffer, as raw bytes
592
561
ByteBuffer value;
593
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
594
content_line.getBuffer().limit() - pos );
562
value = ByteBuffer.wrap( buffer.array(), pos,
563
buffer.limit() - pos );
596
565
// get parameter parts
597
566
String[] name_param_parts = name_and_params.split( ";", -1 );
604
573
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
606
575
boolean is_interesting_field =
607
interesting_fields.contains(
608
name_param_parts[ 0 ].toUpperCase( Locale.ENGLISH ) );
576
interesting_fields.contains( name_param_parts[ 0 ] );
610
578
// parse encoding parameter
611
579
String encoding = checkParam( name_param_parts, "ENCODING" );
612
580
if( encoding != null )
613
encoding = encoding.toUpperCase( Locale.ENGLISH );
581
encoding = encoding.toUpperCase( Locale.US );
614
582
if( is_interesting_field && encoding != null &&
615
!encoding.equalsIgnoreCase( "8BIT" ) &&
616
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
617
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
583
!encoding.equals( "8BIT" ) &&
584
!encoding.equals( "QUOTED-PRINTABLE" ) )
585
//&& !encoding.equals( "BASE64" ) )
619
587
throw new ParseException( R.string.error_vcf_encoding );
622
590
// parse charset parameter
623
591
String charset = checkParam( name_param_parts, "CHARSET" );
624
592
if( charset != null )
625
charset = charset.toUpperCase( Locale.ENGLISH );
593
charset = charset.toUpperCase( Locale.US );
626
594
if( charset != null &&
627
!charset.equalsIgnoreCase( "US-ASCII" ) &&
628
!charset.equalsIgnoreCase( "ASCII" ) &&
629
!charset.equalsIgnoreCase( "UTF-8" ) )
595
!charset.equals( "US-ASCII" ) &&
596
!charset.equals( "ASCII" ) &&
597
!charset.equals( "UTF-8" ) )
631
599
throw new ParseException( R.string.error_vcf_charset );
634
602
// do unencoding (or default to a fake unencoding result with
635
603
// the raw string)
636
604
UnencodeResult unencoding_result = null;
637
if( encoding != null &&
638
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
605
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
640
606
unencoding_result = unencodeQuotedPrintable( value );
642
// else if( encoding != null &&
643
// encoding.equalsIgnoreCase( "BASE64" ) )
607
// else if( encoding != null && encoding.equals( "BASE64" ) )
645
608
// unencoding_result = unencodeBase64( props[ 1 ], charset );
647
609
if( unencoding_result != null ) {
648
610
value = unencoding_result.getBuffer();
649
611
if( unencoding_result.isAnotherLineRequired() )
654
616
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
655
617
if( ( charset == null && _version.equals( "2.1" ) ) ||
656
618
( charset != null && (
657
charset.equalsIgnoreCase( "ASCII" ) ||
658
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
619
charset.equals( "ASCII" ) ||
620
charset.equals( "US-ASCII" ) ) ) )
660
622
value = transcodeAsciiToUtf8( value );
672
634
// for some entries that have semicolon-separated value parts,
673
635
// check to see if the value ends in an escape character, which
674
636
// indicates that we have a multi-line value
675
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
676
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
677
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
637
if( ( name_param_parts[ 0 ].equals( "N" ) ||
638
name_param_parts[ 0 ].equals( "ORG" ) ||
639
name_param_parts[ 0 ].equals( "ADR" ) ) &&
678
640
doesStringEndInAnEscapeChar( string_value ) )
680
642
_parser_multiline_state = MULTILINE_ESCAPED;
682
644
string_value.length() - 1 );
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
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
687
649
if( _parser_multiline_state == MULTILINE_NONE &&
688
content_line.doesNextLineLookFolded() )
650
_version.equals( "3.0" ) && next_line_looks_folded )
690
652
_parser_multiline_state = MULTILINE_FOLDED;
703
665
if( complete_value.length() < 1 ) return;
705
667
// parse some properties
706
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
668
if( name_param_parts[ 0 ].equals( "N" ) )
707
669
parseN( name_param_parts, complete_value );
708
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
670
else if( name_param_parts[ 0 ].equals( "FN" ) )
709
671
parseFN( name_param_parts, complete_value );
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
672
else if( name_param_parts[ 0 ].equals( "ORG" ) )
711
673
parseORG( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
674
else if( name_param_parts[ 0 ].equals( "TITLE" ) )
713
675
parseTITLE( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
676
else if( name_param_parts[ 0 ].equals( "TEL" ) )
715
677
parseTEL( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
678
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
717
679
parseEMAIL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
680
else if( name_param_parts[ 0 ].equals( "ADR" ) )
719
681
parseADR( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
682
else if( name_param_parts[ 0 ].equals( "LABEL" ) )
721
683
parseLABEL( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
723
parseNOTE( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
725
parseBDAY( name_param_parts, complete_value );
753
711
String str = parts.get( a );
755
// Look for parts that end in an escape character, but ignore
756
// the final part. We've already detected escape chars at the
713
// look for parts that end in an escape character, but ignore
714
// the final part. We've already detected escape chars at the
757
715
// end of the final part in parseLine() and handled multi-lines
759
717
if( a < parts.size() - 1 &&
849
801
for( int b = 0; b < name_part_parts.length; b++ )
850
802
if( name_part_parts[ b ].length() > 0 )
852
if( value.length() > 0 ) value += " ";
804
if( value.length() == 0 ) value += " ";
853
805
value += name_part_parts[ b ];
974
926
for( int a = 0; a < adr_parts.length; a++ )
975
927
if( adr_parts[ a ].length() > 0 )
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 ];
929
// split this part in to it's comma-separated bits
930
String[] adr_part_parts =
931
splitValueByCharacter( adr_parts[ a ], ',' );
932
for( int b = 0; b < adr_part_parts.length; b++ )
933
if( adr_part_parts[ b ].length() > 0 )
935
if( value.length() > 0 ) value += "\n";
936
value += adr_part_parts[ b ];
999
940
Set< String > types = extractTypes( params, Arrays.asList(
1024
965
addAddress( unescapeValue( value ), type );
1027
private void parseNOTE( String[] params, String value )
1029
addNote( unescapeValue( value ) );
1032
private void parseBDAY( String[] params, String value )
1034
setBirthday( value );
1037
968
public void finaliseVcard()
1038
969
throws ParseException, ContactNotIdentifiableException
1040
971
// missing version (and data is present)
1041
if( _version == null && _content_lines != null )
972
if( _version == null && _buffers != null )
1042
973
throw new ParseException( R.string.error_vcf_malformed );
1044
975
// finalise the parent class
1071
1000
HashSet< String > ret = new HashSet< String >();
1073
1002
Pattern p = Pattern.compile(
1074
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1075
Pattern.CASE_INSENSITIVE );
1003
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1076
1004
for( int i = 0; i < params.length; i++ ) {
1077
1005
Matcher m = p.matcher( params[ i ] );
1078
1006
if( m.matches() )
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
1014
* Amongst the params, return any type values present. For v2.1 vCards,
1015
* those types are just parameters. For v3.0, they are prefixed with
1016
* "TYPE=". There may also be multiple type parameters.
1018
* @param a list of type values to look for
1092
1019
* @return a set of present type values
1094
1021
private Set< String > extractTypes( String[] params,
1100
1027
String type_params[] = checkParams( params, "TYPE" );
1101
1028
for( int a = 0; a < type_params.length; a++ )
1103
// check for a comma-separated list of types (why? I don't think
1104
// this is in the specs!)
1030
// check for a comma-separated list of types (why? this isn't in
1105
1032
String[] parts = type_params[ a ].split( "," );
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 );
1033
for( int i = 0; i < parts.length; i++ )
1034
if( valid_types.contains( parts[ i ] ) )
1035
types.add( parts[ i ] );
1113
1038
// get 2.1-style type param
1114
1039
if( _version.equals( "2.1" ) ) {
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 );
1040
for( int i = 1; i < params.length; i++ )
1041
if( valid_types.contains( params[ i ] ) )
1042
types.add( params[ i ] );
1144
1067
else if( ch == '=' && i == in.limit() - 1 )
1146
1069
// we found a '=' at the end of a line signifying a multi-
1147
// line string, so we don't add it
1070
// line string, so we don't add it.
1148
1071
another = true;