140
154
private void importVCardFile( File file ) throws AbortImportException
156
// check file is good
158
showError( getText( R.string.error_filenotfound ) +
160
if( file.length() == 0 )
161
showError( getText( R.string.error_fileisempty ) +
145
BufferedReader reader = new BufferedReader(
146
new FileReader( file ) );
149
StringBuffer content = new StringBuffer();
151
while( ( line = reader.readLine() ) != null )
152
content.append( line ).append( "\n" );
154
importVCardFileContent( content.toString(), file.getName() );
167
FileInputStream istream = new FileInputStream( file );
168
byte[] content = new byte[ (int)file.length() ];
169
istream.read( content );
172
importVCardFileContent( content, file.getName() );
156
174
catch( FileNotFoundException e ) {
157
showError( getText( R.string.error_filenotfound ) + file.getName() );
175
showError( getText( R.string.error_filenotfound ) +
159
178
catch( IOException e ) {
160
179
showError( getText( R.string.error_ioerror ) + file.getName() );
164
private void importVCardFileContent( String content, String fileName )
165
throws AbortImportException
183
private void importVCardFileContent( byte[] content, String fileName )
184
throws AbortImportException
167
// unfold RFC2425 section 5.8.1 folded lines, except that we must also
168
// handle embedded Quoted-Printable encodings that have a trailing '='.
169
// So we remove these first before doing RFC2425 unfolding.
170
content = content.replaceAll( "=\n[ \\t]", "" )
171
.replaceAll( "\n[ \\t]", "" );
173
// get lines and parse them
174
String[] lines = content.split( "\n" );
175
187
VCard vCard = null;
176
for( int i = 0; i < lines.length; i++ )
188
ContentLineIterator cli = new ContentLineIterator( content );
189
while( cli.hasNext() )
178
String line = lines[ i ];
191
ByteBuffer buffer = cli.next();
193
// get a US-ASCII version of the line for processing
196
line = new String( buffer.array(), buffer.position(),
197
buffer.limit() - buffer.position(), "US-ASCII" );
199
catch( UnsupportedEncodingException e ) {
200
// we know US-ASCII is supported, so appease the compiler...
180
204
if( vCard == null ) {
181
205
// look for vcard beginning
182
if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
206
if( line.matches( "^BEGIN:VCARD" ) ) {
183
207
setProgress( ++_progress );
184
208
vCard = new VCard();
188
212
// look for vcard content or ending
189
if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
213
if( line.matches( "^END:VCARD" ) )
191
215
// store vcard and do away with it
383
@SuppressWarnings("serial")
262
384
protected class SkipContactException extends Exception { }
264
public void parseLine( String line )
265
throws ParseException, SkipContactException,
268
// get property halves
269
String[] props = line.split( ":" );
270
for( int i = 0; i < props.length; i++ )
271
props[ i ] = props[ i ].trim();
272
if( props.length < 2 ||
273
props[ 0 ].length() < 1 || props[ 1 ].length() < 1 )
274
throw new ParseException( R.string.error_vcf_malformed );
386
private String extractCollonPartFromLine( ByteBuffer buffer,
387
String line, boolean former )
391
// get a US-ASCII version of the line for processing, unless we were
395
line = new String( buffer.array(), buffer.position(),
396
buffer.limit() - buffer.position(), "US-ASCII" );
398
catch( UnsupportedEncodingException e ) {
399
// we know US-ASCII is supported, so appease the compiler...
404
// split line into name and value parts and check to make sure we
405
// only got 2 parts and that the first part is not zero in length
406
String[] parts = line.split( ":", 2 );
407
if( parts.length == 2 && parts[ 0 ].length() > 0 )
408
ret = parts[ former? 0 : 1 ];
413
private String extractNameAndParamsFromLine( ByteBuffer buffer,
416
return extractCollonPartFromLine( buffer, line, true );
419
private String extractValueFromLine( ByteBuffer buffer, String line )
421
return extractCollonPartFromLine( buffer, line, false );
424
public void parseLine( ByteBuffer buffer, String line,
425
boolean next_line_looks_folded )
426
throws ParseException, SkipContactException,
429
// do we have a version yet?
276
430
if( _version == null )
278
if( props[ 0 ].equals( "VERSION" ) )
432
// tentatively get name and params from line
433
String name_and_params =
434
extractNameAndParamsFromLine( buffer, line );
436
// is it a version line?
437
if( name_and_params != null &&
438
name_and_params.equals( "VERSION" ) )
281
if( !props[ 1 ].equals( "2.1" ) &&
282
!props[ 1 ].equals( "3.0" ) )
441
String value = extractValueFromLine( buffer, line );
442
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
283
443
throw new ParseException( R.string.error_vcf_version );
284
_version = props[ 1 ];
286
// parse any other lines we've accumulated so far
288
for( int i = 0; i < _lines.size(); i++ )
289
parseLine( _lines.get( i ) );
446
// parse any buffers we've been accumulating while we waited
448
if( _buffers != null )
449
for( int i = 0; i < _buffers.size(); i++ )
450
parseLine( _buffers.get( i ), null,
451
i + 1 < _buffers.size() &&
452
_buffers.get( i + 1 ).hasRemaining() &&
453
_buffers.get( i + 1 ).get(
454
_buffers.get( i + 1 ).position() ) == ' ' );
294
// stash this line till we have a version
296
_lines = new Vector< String >();
459
// no, so stash this line till we get a version
460
if( _buffers == null )
461
_buffers = new Vector< ByteBuffer >();
462
_buffers.add( buffer );
467
// name and params and the position in the buffer where the
468
// "value" part of the line start
469
String name_and_params;
472
if( _parser_in_encoded_multiline ||
473
_parser_in_folded_multiline )
475
// if we're currently in a multi-line value, use the stored
476
// property name and parameters
477
name_and_params = _parser_current_name_and_params;
479
pos = buffer.position();
481
// for folded multi-lines, skip the single space at the
482
// start of the next line
483
if( _parser_in_folded_multiline )
486
// else, this must be an encoded multi-line, so skip any
487
// whitespace we find at the start of the next line
489
while( pos < buffer.limit() && (
490
buffer.get( pos ) == ' ' ||
491
buffer.get( pos ) == '\t' ) )
498
// get name and params from line, and since we're not
499
// parsing a subsequent line in a multi-line, this should
500
// not fail, or it's an error
502
extractNameAndParamsFromLine( buffer, line );
503
if( name_and_params == null )
504
throw new ParseException(
505
R.string.error_vcf_malformed );
507
// calculate how many chars to skip from beginning of line
508
// so we skip the property "name:" part
509
pos = buffer.position() + name_and_params.length() + 1;
511
// reset the saved multi-line state
512
_parser_current_name_and_params = name_and_params;
513
_parser_buffered_value_so_far = "";
516
// get value from buffer, as raw bytes
518
value = ByteBuffer.wrap( buffer.array(), pos,
519
buffer.limit() - pos );
302
521
// get parameter parts
303
String[] params = props[ 0 ].split( ";" );
304
for( int i = 0; i < params.length; i++ )
305
params[ i ] = params[ i ].trim();
522
String[] name_param_parts = name_and_params.split( ";", -1 );
523
for( int i = 0; i < name_param_parts.length; i++ )
524
name_param_parts[ i ] = name_param_parts[ i ].trim();
526
// parse encoding parameter
527
String encoding = checkParam( name_param_parts, "ENCODING" );
528
if( encoding != null ) encoding = encoding.toUpperCase();
529
if( encoding != null && !encoding.equals( "8BIT" ) &&
530
!encoding.equals( "QUOTED-PRINTABLE" ) )
531
//&& !encoding.equals( "BASE64" ) )
533
throw new ParseException( R.string.error_vcf_encoding );
536
// parse charset parameter
537
String charset = checkParam( name_param_parts, "CHARSET" );
538
if( charset != null ) charset = charset.toUpperCase();
539
if( charset != null && !charset.equals( "US-ASCII" ) &&
540
!charset.equals( "ASCII" ) &&
541
!charset.equals( "UTF-8" ) )
543
throw new ParseException( R.string.error_vcf_charset );
546
// do unencoding (or default to a fake unencoding result with
548
UnencodeResult unencoding_result = null;
549
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
550
unencoding_result = unencodeQuotedPrintable( value );
551
// else if( encoding != null && encoding.equals( "BASE64" ) )
552
// unencoding_result = unencodeBase64( props[ 1 ], charset );
553
if( unencoding_result != null ) {
554
value = unencoding_result.getBuffer();
555
_parser_in_encoded_multiline =
556
unencoding_result.isAnotherLineRequired();
559
// convert 8-bit ASCII charset to US-ASCII
560
if( charset == null || charset.equals( "ASCII" ) ) {
561
value = transcodeAsciiToUtf8( value );
568
string_value = new String( value.array(), value.position(),
569
value.limit() - value.position(), charset );
570
} catch( UnsupportedEncodingException e ) {
571
throw new ParseException( R.string.error_vcf_charset );
574
// now we know whether we're in an encoding multi-line,
575
// determine if we're in a v3 folded multi-line or not
576
_parser_in_folded_multiline = !_parser_in_encoded_multiline &&
577
_version.equals( "3.0" ) && next_line_looks_folded;
579
// handle multi-line requests
580
if( _parser_in_encoded_multiline ||
581
_parser_in_folded_multiline )
583
_parser_buffered_value_so_far += string_value;
587
// add on buffered multi-line content
588
String complete_value =
589
_parser_buffered_value_so_far + string_value;
591
// ignore empty values
592
if( complete_value.length() < 1 ) return;
307
594
// parse some properties
308
if( params[ 0 ].equals( "N" ) )
309
parseN( params, props[ 1 ] );
310
else if( params[ 0 ].equals( "FN" ) )
311
parseFN( params, props[ 1 ] );
312
else if( params[ 0 ].equals( "ORG" ) )
313
parseORG( params, props[ 1 ] );
314
else if( params[ 0 ].equals( "TEL" ) )
315
parseTEL( params, props[ 1 ] );
316
else if( params[ 0 ].equals( "EMAIL" ) )
317
parseEMAIL( params, props[ 1 ] );
595
if( name_param_parts[ 0 ].equals( "N" ) )
596
parseN( name_param_parts, complete_value );
597
else if( name_param_parts[ 0 ].equals( "FN" ) )
598
parseFN( name_param_parts, complete_value );
599
else if( name_param_parts[ 0 ].equals( "ORG" ) )
600
parseORG( name_param_parts, complete_value );
601
else if( name_param_parts[ 0 ].equals( "TEL" ) )
602
parseTEL( name_param_parts, complete_value );
603
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
604
parseEMAIL( name_param_parts, complete_value );
321
608
private void parseN( String[] params, String value )
322
throws ParseException, SkipContactException,
609
throws ParseException, SkipContactException,
325
612
// already got a better name?
326
if( _nameLevel >= NAMELEVEL_N ) return;
613
if( _name_level >= NAMELEVEL_N ) return;
328
615
// get name parts
329
String[] nameparts = value.split( ";" );
330
for( int i = 0; i < nameparts.length; i++ )
331
nameparts[ i ] = nameparts[ i ].trim();
616
String[] name_parts = value.split( ";" );
617
for( int i = 0; i < name_parts.length; i++ )
618
name_parts[ i ] = name_parts[ i ].trim();
335
if( nameparts.length > 1 && nameparts[ 1 ].length() > 0 )
336
value += nameparts[ 1 ];
337
if( nameparts[ 0 ].length() > 0 )
338
value += ( value.length() == 0? "" : " " ) + nameparts[ 0 ];
622
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
623
value += name_parts[ 1 ];
624
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
625
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
341
setName( undoCharsetAndEncoding( params, value ) );
342
_nameLevel = NAMELEVEL_N;
629
_name_level = NAMELEVEL_N;
344
631
// check now to see if we need to import this contact (to avoid
345
632
// parsing the rest of the vCard unnecessarily)
350
637
private void parseFN( String[] params, String value )
351
throws ParseException, SkipContactException
638
throws ParseException, SkipContactException
353
640
// already got a better name?
354
if( _nameLevel >= NAMELEVEL_FN ) return;
641
if( _name_level >= NAMELEVEL_FN ) return;
357
setName( undoCharsetAndEncoding( params, value ) );
358
_nameLevel = NAMELEVEL_FN;
645
_name_level = NAMELEVEL_FN;
361
648
private void parseORG( String[] params, String value )
362
throws ParseException, SkipContactException
649
throws ParseException, SkipContactException
364
651
// already got a better name?
365
if( _nameLevel >= NAMELEVEL_ORG ) return;
652
if( _name_level >= NAMELEVEL_ORG ) return;
368
String[] orgparts = value.split( ";" );
369
for( int i = 0; i < orgparts.length; i++ )
370
orgparts[ i ] = orgparts[ i ].trim();
655
String[] org_parts = value.split( ";" );
656
for( int i = 0; i < org_parts.length; i++ )
657
org_parts[ i ] = org_parts[ i ].trim();
373
if( orgparts[ 0 ].length() == 0 && orgparts.length > 1 )
374
value = orgparts[ 1 ];
660
if( org_parts.length > 1 && org_parts[ 0 ].length() == 0 )
661
value = org_parts[ 1 ];
376
value = orgparts[ 0 ];
663
value = org_parts[ 0 ];
379
setName( undoCharsetAndEncoding( params, value ) );
380
_nameLevel = NAMELEVEL_ORG;
667
_name_level = NAMELEVEL_ORG;
383
670
private void parseTEL( String[] params, String value )
384
throws ParseException
671
throws ParseException
386
673
if( value.length() == 0 ) return;
388
675
Set< String > types = extractTypes( params, Arrays.asList(
389
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
390
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
676
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
677
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
392
679
// here's the logic...
393
680
boolean preferred = types.contains( "PREF" );
681
int type = PhonesColumns.TYPE_MOBILE;
394
682
if( types.contains( "VOICE" ) )
395
683
if( types.contains( "WORK" ) )
396
addPhone( value, PhonesColumns.TYPE_WORK, preferred );
684
type = PhonesColumns.TYPE_WORK;
398
addPhone( value, PhonesColumns.TYPE_HOME, preferred );
686
type = PhonesColumns.TYPE_HOME;
399
687
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
400
addPhone( value, PhonesColumns.TYPE_MOBILE, preferred );
688
type = PhonesColumns.TYPE_MOBILE;
401
689
if( types.contains( "FAX" ) )
402
690
if( types.contains( "HOME" ) )
403
addPhone( value, PhonesColumns.TYPE_FAX_HOME, preferred );
691
type = PhonesColumns.TYPE_FAX_HOME;
405
addPhone( value, PhonesColumns.TYPE_FAX_WORK, preferred );
693
type = PhonesColumns.TYPE_FAX_WORK;
406
694
if( types.contains( "PAGER" ) )
407
addPhone( value, PhonesColumns.TYPE_PAGER, preferred );
695
type = PhonesColumns.TYPE_PAGER;
698
addPhone( value, type, preferred );
410
701
public void parseEMAIL( String[] params, String value )
702
throws ParseException
412
704
if( value.length() == 0 ) return;
414
706
Set< String > types = extractTypes( params, Arrays.asList(
415
"PREF", "WORK", "HOME", "INTERNET" ) );
707
"PREF", "WORK", "HOME", "INTERNET" ) );
417
709
// here's the logic...
418
710
boolean preferred = types.contains( "PREF" );
425
717
public void finaliseParsing()
426
throws ParseException, SkipContactException,
718
throws ParseException, SkipContactException,
429
721
// missing version (and data is present)
430
if( _version == null && _lines != null )
722
if( _version == null && _buffers != null )
431
723
throw new ParseException( R.string.error_vcf_malformed );
433
725
// missing name properties?
434
if( _nameLevel == NAMELEVEL_NONE )
726
if( _name_level == NAMELEVEL_NONE )
435
727
throw new ParseException( R.string.error_vcf_noname );
437
729
// check if we should import this one? If we've already got an 'N'-
438
730
// type name, this will already have been done by parseN() so we
439
731
// mustn't do this here (or it could prompt twice!)
440
if( _nameLevel < NAMELEVEL_N && !isImportRequired( getName() ) )
732
if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
441
733
throw new SkipContactException();
444
private String undoCharsetAndEncoding( String[] params, String value )
445
throws ParseException
447
// check encoding/charset
448
String charset, encoding;
449
if( ( charset = checkParam( params, "CHARSET" ) ) != null &&
450
!charset.equals( "UTF-8" ) && !charset.equals( "UTF-16" ) )
451
throw new ParseException( R.string.error_vcf_charset );
452
if( ( encoding = checkParam( params, "ENCODING" ) ) != null &&
453
!encoding.equals( "QUOTED-PRINTABLE" ) )
454
throw new ParseException( R.string.error_vcf_encoding );
457
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
458
return unencodeQuotedPrintable( value, charset );
464
736
private String checkParam( String[] params, String name )
466
Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
738
Pattern p = Pattern.compile(
739
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
467
740
for( int i = 0; i < params.length; i++ ) {
468
741
Matcher m = p.matcher( params[ i ] );
469
742
if( m.matches() )
475
748
private Set< String > extractTypes( String[] params,
476
List< String > validTypes )
749
List< String > valid_types )
478
751
HashSet< String > types = new HashSet< String >();
480
753
// get 3.0-style TYPE= param
482
if( ( typeParam = checkParam( params, "TYPE" ) ) != null ) {
483
String[] bits = typeParam.split( "," );
484
for( int i = 0; i < bits.length; i++ )
485
if( validTypes.contains( bits[ i ] ) )
486
types.add( bits[ i ] );
755
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
756
String[] parts = type_param.split( "," );
757
for( int i = 0; i < parts.length; i++ )
758
if( valid_types.contains( parts[ i ] ) )
759
types.add( parts[ i ] );
489
762
// get 2.1-style type param
490
763
if( _version.equals( "2.1" ) ) {
491
764
for( int i = 1; i < params.length; i++ )
492
if( validTypes.contains( params[ i ] ) )
765
if( valid_types.contains( params[ i ] ) )
493
766
types.add( params[ i ] );
499
private String unencodeQuotedPrintable( String str, String charset )
772
private UnencodeResult unencodeQuotedPrintable( ByteBuffer in )
501
// default encoding scheme
502
if( charset == null ) charset = "UTF-8";
774
boolean another = false;
504
// unencode quoted-pritable encoding, as per RFC1521 section 5.1
505
byte[] bytes = new byte[ str.length() ];
776
// unencode quoted-printable encoding, as per RFC1521 section 5.1
777
byte[] out = new byte[ in.limit() - in.position() ];
507
for( int i = 0; i < str.length(); i++, j++ ) {
508
char ch = str.charAt( i );
509
if( ch == '=' && i < str.length() - 2 ) {
511
Character.digit( str.charAt( i + 1 ), 16 ) * 16 +
512
Character.digit( str.charAt( i + 2 ), 16 ) );
779
for( int i = in.position(); i < in.limit(); i++ )
781
// get next char and process...
782
byte ch = in.array()[ i ];
783
if( ch == '=' && i < in.limit() - 2 )
785
// we found a =XX format byte, add it
787
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
788
Character.digit( in.array()[ i + 2 ], 16 ) );
791
else if( ch == '=' && i == in.limit() - 1 )
793
// we found a '=' at the end of a line signifying a multi-
794
// line string, so we don't add it.
516
bytes[ j ] = (byte)ch;
519
return new String( bytes, 0, j, charset );
520
} catch( UnsupportedEncodingException e ) { }
799
// just a normal char...
804
return new UnencodeResult( another, ByteBuffer.wrap( out, 0, j ) );
807
private ByteBuffer transcodeAsciiToUtf8( ByteBuffer in )
810
byte[] out = new byte[ ( in.limit() - in.position() ) * 2 ];
812
for( int a = in.position(); a < in.limit(); a++ )
814
// if char is < 127, keep it as-is
815
if( in.array()[ a ] >= 0 )
816
out[ j++ ] = in.array()[ a ];
818
// else, convert it to UTF-8
820
int b = 0xff & (int)in.array()[ a ];
821
out[ j++ ] = (byte)( 0xc0 | ( b >> 6 ) );
822
out[ j++ ] = (byte)( 0x80 | ( b & 0x3f ) );
826
return ByteBuffer.wrap( out, 0, j );