138
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 ) +
143
BufferedReader reader = new BufferedReader(
144
new FileReader( file ) );
147
StringBuffer content = new StringBuffer();
149
while( ( line = reader.readLine() ) != null )
150
content.append( line ).append( "\n" );
152
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() );
154
174
catch( FileNotFoundException e ) {
155
showError( getText( R.string.error_filenotfound ) + file.getName() );
175
showError( getText( R.string.error_filenotfound ) +
157
178
catch( IOException e ) {
158
179
showError( getText( R.string.error_ioerror ) + file.getName() );
162
private void importVCardFileContent( String content, String fileName )
163
throws AbortImportException
183
private void importVCardFileContent( byte[] content, String fileName )
184
throws AbortImportException
165
// unfold RFC2425 section 5.8.1 folded lines, except that we must also
166
// handle embedded Quoted-Printable encodings that have a trailing '='.
167
// So we remove these first before doing RFC2425 unfolding.
168
content = content.replaceAll( "=\n[ \\t]", "" )
169
.replaceAll( "\n[ \\t]", "" );
171
// get lines and parse them
172
String[] lines = content.split( "\n" );
173
187
VCard vCard = null;
174
for( int i = 0; i < lines.length; i++ )
188
ContentLineIterator cli = new ContentLineIterator( content );
189
while( cli.hasNext() )
176
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...
178
204
if( vCard == null ) {
179
205
// look for vcard beginning
180
if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
206
if( line.matches( "^BEGIN:VCARD" ) ) {
181
207
setProgress( ++_progress );
182
208
vCard = new VCard();
186
212
// look for vcard content or ending
187
if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
213
if( line.matches( "^END:VCARD" ) )
189
215
// store vcard and do away with it
262
387
@SuppressWarnings("serial")
263
388
protected class SkipContactException extends Exception { }
265
public void parseLine( String line )
266
throws ParseException, SkipContactException,
269
// get property halves
270
String[] props = line.split( ":" );
271
for( int i = 0; i < props.length; i++ )
272
props[ i ] = props[ i ].trim();
273
if( props.length < 2 ||
274
props[ 0 ].length() < 1 || props[ 1 ].length() < 1 )
275
throw new ParseException( R.string.error_vcf_malformed );
390
private String extractCollonPartFromLine( ByteBuffer buffer,
391
String line, boolean former )
395
// get a US-ASCII version of the line for processing, unless we were
399
line = new String( buffer.array(), buffer.position(),
400
buffer.limit() - buffer.position(), "US-ASCII" );
402
catch( UnsupportedEncodingException e ) {
403
// we know US-ASCII is supported, so appease the compiler...
408
// split line into name and value parts and check to make sure we
409
// only got 2 parts and that the first part is not zero in length
410
String[] parts = line.split( ":", 2 );
411
if( parts.length == 2 && parts[ 0 ].length() > 0 )
412
ret = parts[ former? 0 : 1 ];
417
private String extractNameAndParamsFromLine( ByteBuffer buffer,
420
return extractCollonPartFromLine( buffer, line, true );
423
private String extractValueFromLine( ByteBuffer buffer, String line )
425
return extractCollonPartFromLine( buffer, line, false );
428
public void parseLine( ByteBuffer buffer, String line,
429
boolean next_line_looks_folded )
430
throws ParseException, SkipContactException,
433
// do we have a version yet?
277
434
if( _version == null )
279
if( props[ 0 ].equals( "VERSION" ) )
436
// tentatively get name and params from line
437
String name_and_params =
438
extractNameAndParamsFromLine( buffer, line );
440
// is it a version line?
441
if( name_and_params != null &&
442
name_and_params.equals( "VERSION" ) )
282
if( !props[ 1 ].equals( "2.1" ) &&
283
!props[ 1 ].equals( "3.0" ) )
445
String value = extractValueFromLine( buffer, line );
446
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
284
447
throw new ParseException( R.string.error_vcf_version );
285
_version = props[ 1 ];
287
// parse any other lines we've accumulated so far
289
for( int i = 0; i < _lines.size(); i++ )
290
parseLine( _lines.get( i ) );
450
// parse any buffers we've been accumulating while we waited
452
if( _buffers != null )
453
for( int i = 0; i < _buffers.size(); i++ )
454
parseLine( _buffers.get( i ), null,
455
i + 1 < _buffers.size() &&
456
_buffers.get( i + 1 ).hasRemaining() &&
457
_buffers.get( i + 1 ).get(
458
_buffers.get( i + 1 ).position() ) == ' ' );
295
// stash this line till we have a version
297
_lines = new Vector< String >();
463
// no, so stash this line till we get a version
464
if( _buffers == null )
465
_buffers = new Vector< ByteBuffer >();
466
_buffers.add( buffer );
471
// name and params and the position in the buffer where the
472
// "value" part of the line start
473
String name_and_params;
476
if( _parser_multiline_state != MULTILINE_NONE )
478
// if we're currently in a multi-line value, use the stored
479
// property name and parameters
480
name_and_params = _parser_current_name_and_params;
482
// skip some initial line characters, depending on the type
483
// of multi-line we're handling
484
pos = buffer.position();
485
switch( _parser_multiline_state )
487
case MULTILINE_FOLDED:
490
case MULTILINE_ENCODED:
491
while( pos < buffer.limit() && (
492
buffer.get( pos ) == ' ' ||
493
buffer.get( pos ) == '\t' ) )
502
// take us out of multi-line so that we can re-detect that
503
// this line is a multi-line or not
504
_parser_multiline_state = MULTILINE_NONE;
508
// get name and params from line, and since we're not
509
// parsing a subsequent line in a multi-line, this should
510
// not fail, or it's an error
512
extractNameAndParamsFromLine( buffer, line );
513
if( name_and_params == null )
514
throw new ParseException(
515
R.string.error_vcf_malformed );
517
// calculate how many chars to skip from beginning of line
518
// so we skip the property "name:" part
519
pos = buffer.position() + name_and_params.length() + 1;
521
// reset the saved multi-line state
522
_parser_current_name_and_params = name_and_params;
523
_parser_buffered_value_so_far = "";
526
// get value from buffer, as raw bytes
528
value = ByteBuffer.wrap( buffer.array(), pos,
529
buffer.limit() - pos );
303
531
// get parameter parts
304
String[] params = props[ 0 ].split( ";" );
305
for( int i = 0; i < params.length; i++ )
306
params[ i ] = params[ i ].trim();
532
String[] name_param_parts = name_and_params.split( ";", -1 );
533
for( int i = 0; i < name_param_parts.length; i++ )
534
name_param_parts[ i ] = name_param_parts[ i ].trim();
536
// parse encoding parameter
537
String encoding = checkParam( name_param_parts, "ENCODING" );
538
if( encoding != null ) encoding = encoding.toUpperCase();
539
if( encoding != null && !encoding.equals( "8BIT" ) &&
540
!encoding.equals( "QUOTED-PRINTABLE" ) )
541
//&& !encoding.equals( "BASE64" ) )
543
throw new ParseException( R.string.error_vcf_encoding );
546
// parse charset parameter
547
String charset = checkParam( name_param_parts, "CHARSET" );
548
if( charset != null ) charset = charset.toUpperCase();
549
if( charset != null && !charset.equals( "US-ASCII" ) &&
550
!charset.equals( "ASCII" ) &&
551
!charset.equals( "UTF-8" ) )
553
throw new ParseException( R.string.error_vcf_charset );
556
// do unencoding (or default to a fake unencoding result with
558
UnencodeResult unencoding_result = null;
559
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
560
unencoding_result = unencodeQuotedPrintable( value );
561
// else if( encoding != null && encoding.equals( "BASE64" ) )
562
// unencoding_result = unencodeBase64( props[ 1 ], charset );
563
if( unencoding_result != null ) {
564
value = unencoding_result.getBuffer();
565
if( unencoding_result.isAnotherLineRequired() )
566
_parser_multiline_state = MULTILINE_ENCODED;
569
// convert 8-bit ASCII charset to US-ASCII
570
if( charset == null || charset.equals( "ASCII" ) ) {
571
value = transcodeAsciiToUtf8( value );
578
string_value = new String( value.array(), value.position(),
579
value.limit() - value.position(), charset );
580
} catch( UnsupportedEncodingException e ) {
581
throw new ParseException( R.string.error_vcf_charset );
584
// for some entries that have semicolon-separated value parts,
585
// check to see if the value ends in an escape character, which
586
// indicates that we have a multi-line value
587
if( ( name_param_parts[ 0 ].equals( "N" ) ||
588
name_param_parts[ 0 ].equals( "ORG" ) ||
589
name_param_parts[ 0 ].equals( "ADR" ) ) &&
590
doesStringEndInAnEscapeChar( string_value ) )
592
_parser_multiline_state = MULTILINE_ESCAPED;
593
string_value = string_value.substring( 0,
594
string_value.length() - 1 );
597
// now we know whether we're in an encoding multi-line,
598
// determine if we're in a v3 folded multi-line or not
599
if( _parser_multiline_state == MULTILINE_NONE &&
600
_version.equals( "3.0" ) && next_line_looks_folded )
602
_parser_multiline_state = MULTILINE_FOLDED;
605
// handle multi-lines by buffering them and parsing them when we
606
// are processing the last line in a multi-line sequence
607
if( _parser_multiline_state != MULTILINE_NONE ) {
608
_parser_buffered_value_so_far += string_value;
611
String complete_value =
612
( _parser_buffered_value_so_far + string_value ).trim();
614
// ignore empty values
615
if( complete_value.length() < 1 ) return;
308
617
// parse some properties
309
if( params[ 0 ].equals( "N" ) )
310
parseN( params, props[ 1 ] );
311
else if( params[ 0 ].equals( "FN" ) )
312
parseFN( params, props[ 1 ] );
313
else if( params[ 0 ].equals( "ORG" ) )
314
parseORG( params, props[ 1 ] );
315
else if( params[ 0 ].equals( "TEL" ) )
316
parseTEL( params, props[ 1 ] );
317
else if( params[ 0 ].equals( "EMAIL" ) )
318
parseEMAIL( params, props[ 1 ] );
618
if( name_param_parts[ 0 ].equals( "N" ) )
619
parseN( name_param_parts, complete_value );
620
else if( name_param_parts[ 0 ].equals( "FN" ) )
621
parseFN( name_param_parts, complete_value );
622
else if( name_param_parts[ 0 ].equals( "ORG" ) )
623
parseORG( name_param_parts, complete_value );
624
else if( name_param_parts[ 0 ].equals( "TEL" ) )
625
parseTEL( name_param_parts, complete_value );
626
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
627
parseEMAIL( name_param_parts, complete_value );
628
else if( name_param_parts[ 0 ].equals( "ADR" ) )
629
parseADR( name_param_parts, complete_value );
633
private boolean doesStringEndInAnEscapeChar( String string )
635
// count the number of backslashes at the end of the string
637
for( int a = string.length() - 1; a >= 0; a-- )
638
if( string.charAt( a ) == '\\' )
643
// if there are an even number of backslashes then the final one
645
return ( count & 1 ) == 1;
648
private String[] splitValueBySemicolon( String value )
650
// split string in to parts by semicolon
651
ArrayList< String > parts = new ArrayList< String >(
652
Arrays.asList( value.split( ";" ) ) );
655
for( int a = 0; a < parts.size(); a++ )
657
String str = parts.get( a );
659
// look for parts that end in an escape character, but ignore
660
// the final part. We've already detected escape chars at the
661
// end of the final part in parseLine() and handled multi-lines
663
if( a < parts.size() - 1 &&
664
doesStringEndInAnEscapeChar( str ) )
666
// join the next part to this part and remove the next part
667
parts.set( a, str.substring( 0, str.length() - 1 ) +
668
';' + parts.get( a + 1 ) );
669
parts.remove( a + 1 );
671
// re-visit this part
676
// trim and replace string
681
String[] ret = new String[ parts.size() ];
682
return parts.toArray( ret );
322
685
private void parseN( String[] params, String value )
323
throws ParseException, SkipContactException,
686
throws ParseException, SkipContactException,
326
689
// already got a better name?
327
if( _nameLevel >= NAMELEVEL_N ) return;
690
if( _name_level >= NAMELEVEL_N ) return;
329
692
// get name parts
330
String[] nameparts = value.split( ";" );
331
for( int i = 0; i < nameparts.length; i++ )
332
nameparts[ i ] = nameparts[ i ].trim();
693
String[] name_parts = splitValueBySemicolon( value );
336
if( nameparts.length > 1 && nameparts[ 1 ].length() > 0 )
337
value += nameparts[ 1 ];
338
if( nameparts[ 0 ].length() > 0 )
339
value += ( value.length() == 0? "" : " " ) + nameparts[ 0 ];
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 ];
342
setName( undoCharsetAndEncoding( params, value ) );
343
_nameLevel = NAMELEVEL_N;
704
_name_level = NAMELEVEL_N;
345
706
// check now to see if we need to import this contact (to avoid
346
707
// parsing the rest of the vCard unnecessarily)
351
712
private void parseFN( String[] params, String value )
352
throws ParseException, SkipContactException
713
throws ParseException, SkipContactException
354
715
// already got a better name?
355
if( _nameLevel >= NAMELEVEL_FN ) return;
716
if( _name_level >= NAMELEVEL_FN ) return;
358
setName( undoCharsetAndEncoding( params, value ) );
359
_nameLevel = NAMELEVEL_FN;
720
_name_level = NAMELEVEL_FN;
362
723
private void parseORG( String[] params, String value )
363
throws ParseException, SkipContactException
724
throws ParseException, SkipContactException
365
726
// already got a better name?
366
if( _nameLevel >= NAMELEVEL_ORG ) return;
727
if( _name_level >= NAMELEVEL_ORG ) return;
369
String[] orgparts = value.split( ";" );
370
for( int i = 0; i < orgparts.length; i++ )
371
orgparts[ i ] = orgparts[ i ].trim();
730
String[] org_parts = splitValueBySemicolon( value );
374
if( orgparts[ 0 ].length() == 0 && orgparts.length > 1 )
375
value = orgparts[ 1 ];
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 ];
377
value = orgparts[ 0 ];
738
value = org_parts[ 0 ];
380
setName( undoCharsetAndEncoding( params, value ) );
381
_nameLevel = NAMELEVEL_ORG;
742
_name_level = NAMELEVEL_ORG;
384
745
private void parseTEL( String[] params, String value )
385
throws ParseException
746
throws ParseException
387
748
if( value.length() == 0 ) return;
389
750
Set< String > types = extractTypes( params, Arrays.asList(
390
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
391
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
751
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
752
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
393
754
// here's the logic...
394
755
boolean preferred = types.contains( "PREF" );
756
int type = PhonesColumns.TYPE_MOBILE;
395
757
if( types.contains( "VOICE" ) )
396
758
if( types.contains( "WORK" ) )
397
addPhone( value, PhonesColumns.TYPE_WORK, preferred );
759
type = PhonesColumns.TYPE_WORK;
399
addPhone( value, PhonesColumns.TYPE_HOME, preferred );
761
type = PhonesColumns.TYPE_HOME;
400
762
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
401
addPhone( value, PhonesColumns.TYPE_MOBILE, preferred );
763
type = PhonesColumns.TYPE_MOBILE;
402
764
if( types.contains( "FAX" ) )
403
765
if( types.contains( "HOME" ) )
404
addPhone( value, PhonesColumns.TYPE_FAX_HOME, preferred );
766
type = PhonesColumns.TYPE_FAX_HOME;
406
addPhone( value, PhonesColumns.TYPE_FAX_WORK, preferred );
768
type = PhonesColumns.TYPE_FAX_WORK;
407
769
if( types.contains( "PAGER" ) )
408
addPhone( value, PhonesColumns.TYPE_PAGER, preferred );
770
type = PhonesColumns.TYPE_PAGER;
773
addPhone( value, type, preferred );
411
776
public void parseEMAIL( String[] params, String value )
777
throws ParseException
413
779
if( value.length() == 0 ) return;
415
781
Set< String > types = extractTypes( params, Arrays.asList(
416
"PREF", "WORK", "HOME", "INTERNET" ) );
782
"PREF", "WORK", "HOME", "INTERNET" ) );
418
// here's the logic...
419
785
boolean preferred = types.contains( "PREF" );
420
786
if( types.contains( "WORK" ) )
421
787
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
423
789
addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
792
private void parseADR( String[] params, String value )
793
throws ParseException, SkipContactException
796
String[] adr_parts = splitValueBySemicolon( value );
800
for( int a = 0; a < adr_parts.length; a++ ) {
801
if( value.length() > 0 ) value += "\n";
802
value += adr_parts[ a ].trim();
805
Set< String > types = extractTypes( params, Arrays.asList(
806
"PREF", "WORK", "HOME", "INTERNET" ) );
809
if( types.contains( "WORK" ) )
810
addAddress( value, Contacts.ContactMethods.TYPE_WORK );
812
addAddress( value, Contacts.ContactMethods.TYPE_HOME);
426
815
public void finaliseParsing()
427
throws ParseException, SkipContactException,
816
throws ParseException, SkipContactException,
430
819
// missing version (and data is present)
431
if( _version == null && _lines != null )
820
if( _version == null && _buffers != null )
432
821
throw new ParseException( R.string.error_vcf_malformed );
434
823
// missing name properties?
435
if( _nameLevel == NAMELEVEL_NONE )
824
if( _name_level == NAMELEVEL_NONE )
436
825
throw new ParseException( R.string.error_vcf_noname );
438
827
// check if we should import this one? If we've already got an 'N'-
439
828
// type name, this will already have been done by parseN() so we
440
829
// mustn't do this here (or it could prompt twice!)
441
if( _nameLevel < NAMELEVEL_N && !isImportRequired( getName() ) )
830
if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
442
831
throw new SkipContactException();
445
private String undoCharsetAndEncoding( String[] params, String value )
446
throws ParseException
448
// check encoding/charset
449
String charset, encoding;
450
if( ( charset = checkParam( params, "CHARSET" ) ) != null &&
451
!charset.equals( "UTF-8" ) && !charset.equals( "UTF-16" ) )
452
throw new ParseException( R.string.error_vcf_charset );
453
if( ( encoding = checkParam( params, "ENCODING" ) ) != null &&
454
!encoding.equals( "QUOTED-PRINTABLE" ) )
455
throw new ParseException( R.string.error_vcf_encoding );
458
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
459
return unencodeQuotedPrintable( value, charset );
465
834
private String checkParam( String[] params, String name )
467
Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
836
Pattern p = Pattern.compile(
837
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
468
838
for( int i = 0; i < params.length; i++ ) {
469
839
Matcher m = p.matcher( params[ i ] );
470
840
if( m.matches() )
476
846
private Set< String > extractTypes( String[] params,
477
List< String > validTypes )
847
List< String > valid_types )
479
849
HashSet< String > types = new HashSet< String >();
481
851
// get 3.0-style TYPE= param
483
if( ( typeParam = checkParam( params, "TYPE" ) ) != null ) {
484
String[] bits = typeParam.split( "," );
485
for( int i = 0; i < bits.length; i++ )
486
if( validTypes.contains( bits[ i ] ) )
487
types.add( bits[ i ] );
853
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
854
String[] parts = type_param.split( "," );
855
for( int i = 0; i < parts.length; i++ )
856
if( valid_types.contains( parts[ i ] ) )
857
types.add( parts[ i ] );
490
860
// get 2.1-style type param
491
861
if( _version.equals( "2.1" ) ) {
492
862
for( int i = 1; i < params.length; i++ )
493
if( validTypes.contains( params[ i ] ) )
863
if( valid_types.contains( params[ i ] ) )
494
864
types.add( params[ i ] );
500
private String unencodeQuotedPrintable( String str, String charset )
870
private UnencodeResult unencodeQuotedPrintable( ByteBuffer in )
502
// default encoding scheme
503
if( charset == null ) charset = "UTF-8";
872
boolean another = false;
505
// unencode quoted-pritable encoding, as per RFC1521 section 5.1
506
byte[] bytes = new byte[ str.length() ];
874
// unencode quoted-printable encoding, as per RFC1521 section 5.1
875
byte[] out = new byte[ in.limit() - in.position() ];
508
for( int i = 0; i < str.length(); i++, j++ ) {
509
char ch = str.charAt( i );
510
if( ch == '=' && i < str.length() - 2 ) {
512
Character.digit( str.charAt( i + 1 ), 16 ) * 16 +
513
Character.digit( str.charAt( i + 2 ), 16 ) );
877
for( int i = in.position(); i < in.limit(); i++ )
879
// get next char and process...
880
byte ch = in.array()[ i ];
881
if( ch == '=' && i < in.limit() - 2 )
883
// we found a =XX format byte, add it
885
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
886
Character.digit( in.array()[ i + 2 ], 16 ) );
889
else if( ch == '=' && i == in.limit() - 1 )
891
// we found a '=' at the end of a line signifying a multi-
892
// line string, so we don't add it.
517
bytes[ j ] = (byte)ch;
520
return new String( bytes, 0, j, charset );
521
} catch( UnsupportedEncodingException e ) { }
897
// just a normal char...
902
return new UnencodeResult( another, ByteBuffer.wrap( out, 0, j ) );
905
private ByteBuffer transcodeAsciiToUtf8( ByteBuffer in )
908
byte[] out = new byte[ ( in.limit() - in.position() ) * 2 ];
910
for( int a = in.position(); a < in.limit(); a++ )
912
// if char is < 127, keep it as-is
913
if( in.array()[ a ] >= 0 )
914
out[ j++ ] = in.array()[ a ];
916
// else, convert it to UTF-8
918
int b = 0xff & (int)in.array()[ a ];
919
out[ j++ ] = (byte)( 0xc0 | ( b >> 6 ) );
920
out[ j++ ] = (byte)( 0x80 | ( b & 0x3f ) );
924
return ByteBuffer.wrap( out, 0, j );