(root)/android/import-contacts
» Revision
22
: 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 22
- compare with another revision
- compare with revision 117
- download diff
- download tarball
- view history from revision 22
- Committer: edam
- Date: 2010-12-11 23:57:07 UTC
- Revision ID: edam@waxworlds.org-20101211235707-czyw48tt3hcopuwf
- read vCard files in as raw bytes now and convert to string only tentatively to check for version no.s and property names and params
- ASCII is now (correctly) the default charset
- added conversion from 8-bit ASCII to UTF-8 (not used on 7-bit US-ASCII) which works on raw bytes, not chars
- unencode quoted-printable now works on raw bytes, not chars
- ASCII is now (correctly) the default charset
- added conversion from 8-bit ASCII to UTF-8 (not used on 7-bit US-ASCII) which works on raw bytes, not chars
- unencode quoted-printable now works on raw bytes, not chars
added
removed
src/org/waxworlds/edam/importcontacts/VCFImporter.java
25
25
26
26
import java.io.BufferedReader;
27
27
import java.io.File;
28
import java.io.FileInputStream;
28
29
import java.io.FileNotFoundException;
29
30
import java.io.FileReader;
30
31
import java.io.FilenameFilter;
31
32
import java.io.IOException;
32
33
import java.io.UnsupportedEncodingException;
34
import java.nio.ByteBuffer;
33
35
import java.util.Arrays;
34
36
import java.util.HashSet;
35
37
import java.util.List;
148
150
149
151
private void importVCardFile( File file ) throws AbortImportException
150
152
{
153
// check file is good
154
if( !file.exists() )
155
showError( getText( R.string.error_filenotfound ) +
156
file.getName() );
157
if( file.length() == 0 )
158
showError( getText( R.string.error_fileisempty ) +
159
file.getName() );
160
151
161
try
152
162
{
153
// open file
154
BufferedReader reader = new BufferedReader(
155
new FileReader( file ) );
156
157
// read
158
StringBuffer content = new StringBuffer();
159
String line;
160
while( ( line = reader.readLine() ) != null )
161
content.append( line ).append( "\n" );
162
163
importVCardFileContent( content.toString(), file.getName() );
163
// open/read file
164
FileInputStream istream = new FileInputStream( file );
165
byte[] content = new byte[ (int)file.length() ];
166
istream.read( content );
167
168
// import
169
importVCardFileContent( content, file.getName() );
164
170
}
165
171
catch( FileNotFoundException e ) {
166
172
showError( getText( R.string.error_filenotfound ) +
171
177
}
172
178
}
173
179
174
private void importVCardFileContent( String content, String fileName )
180
private void importVCardFileContent( byte[] content, String fileName )
175
181
throws AbortImportException
176
182
{
177
// get lines and parse them
178
String[] lines = content.split( "\n" );
183
ByteBuffer buffers[] = getLinesFromContent( content );
184
185
// go through lines
179
186
VCard vCard = null;
180
for( int i = 0; i < lines.length; i++ )
187
for( int i = 0; i < buffers.length; i++ )
181
188
{
182
String line = lines[ i ];
189
// get a US-ASCII version of the line for processing
190
String line;
191
try {
192
line = new String( buffers[ i ].array(), buffers[ i ].position(),
193
buffers[ i ].limit() - buffers[ i ].position(), "US-ASCII" );
194
}
195
catch( UnsupportedEncodingException e ) {
196
// we know US-ASCII is supported, so appease the compiler...
197
line = "";
198
}
183
199
184
200
if( vCard == null ) {
185
201
// look for vcard beginning
214
230
{
215
231
// try giving the line to the vcard
216
232
try {
217
vCard.parseLine( line );
233
vCard.parseLine( buffers[ i ] );
218
234
}
219
235
catch( VCard.ParseException e ) {
220
236
skipContact();
239
255
}
240
256
}
241
257
258
private ByteBuffer[] getLinesFromContent( byte[] content )
259
{
260
// count lines in data
261
int num_lines = 1;
262
for( int a = 0; a < content.length; a++ )
263
if( content[ a ] == '\n' )
264
num_lines++;
265
266
// get lines, removing \r's and \n's as we go
267
ByteBuffer lines[] = new ByteBuffer[ num_lines ];
268
int last = 0;
269
for( int a = 0, b = 0; a < content.length; a++ )
270
if( content[ a ] == '\n' ) {
271
int to = ( a > 0 && content[ a - 1 ] == '\r' &&
272
a - 1 >= last )? a - 1 : a;
273
lines[ b++ ] = ByteBuffer.wrap( content, last, to - last );
274
last = a + 1;
275
}
276
lines[ lines.length - 1 ] = ByteBuffer.wrap( content, last,
277
content.length - last );
278
279
return lines;
280
}
281
242
282
private class VCard extends ContactData
243
283
{
244
284
private final static int NAMELEVEL_NONE = 0;
247
287
private final static int NAMELEVEL_N = 3;
248
288
249
289
private String _version = null;
250
private Vector< String > _lines = null;
290
private Vector< ByteBuffer > _buffers = null;
251
291
private int _name_level = NAMELEVEL_NONE;
252
292
private boolean _parser_in_multiline = false;
253
293
private String _parser_current_name_and_params = null;
256
296
protected class UnencodeResult
257
297
{
258
298
private boolean _another_line_required;
259
private byte[] _bytes;
260
private int _num_bytes;
299
private ByteBuffer _buffer;
261
300
262
public UnencodeResult( boolean another_line_required, byte[] bytes,
263
int num_bytes )
301
public UnencodeResult( boolean another_line_required,
302
ByteBuffer buffer )
264
303
{
265
304
_another_line_required = another_line_required;
266
_bytes = bytes;
267
_num_bytes = num_bytes;
305
_buffer = buffer;
268
306
}
269
307
270
308
public boolean isAnotherLineRequired()
272
310
return _another_line_required;
273
311
}
274
312
275
public byte[] getBytes()
276
{
277
return _bytes;
278
}
279
280
public int getNumBytes()
281
{
282
return _num_bytes;
313
public ByteBuffer getBuffer()
314
{
315
return _buffer;
283
316
}
284
317
}
285
318
301
334
@SuppressWarnings("serial")
302
335
protected class SkipContactException extends Exception { }
303
336
304
public void parseLine( String line )
337
public void parseLine( ByteBuffer buffer )
305
338
throws ParseException, SkipContactException,
306
339
AbortImportException
307
340
{
341
// get a US-ASCII version of the line for processing
342
String line;
343
try {
344
line = new String( buffer.array(), buffer.position(),
345
buffer.limit() - buffer.position(), "US-ASCII" );
346
}
347
catch( UnsupportedEncodingException e ) {
348
// we know US-ASCII is supported, so appease the compiler...
349
line = "";
350
}
351
308
352
// ignore empty lines
309
353
if( line.trim() == "" ) return;
310
354
311
355
// split line into name and value parts (this may turn out to be
312
356
// unwanted if the line is a subsequent line in a multi-line
313
357
// value, but we have to do this now to check for and handle VCF
314
// versions first)
315
String[] props = line.split( ":", 2 );
316
for( int i = 0; i < props.length; i++ )
317
props[ i ] = props[ i ].trim();
358
// versions first). Also, the value part is only created tentatively
359
// because it may have an encoding/charset. Since we're treating it
360
// as UTF-8 (which is compatible with 7-bit US-ASCII) this is ok
361
// though so long as we later use the raw bytes. ALso we check for
362
// malformed property:name pairs.
363
String name_and_params, string_value;
364
{
365
String[] bits = line.split( ":", 2 );
366
if( bits.length == 2 ) {
367
name_and_params = bits[ 0 ].trim();
368
string_value = bits[ 1 ].trim();
369
if( name_and_params.length() == 0 )
370
throw new ParseException( R.string.error_vcf_malformed );
371
}
372
else
373
{
374
if( !_parser_in_multiline )
375
throw new ParseException( R.string.error_vcf_malformed );
376
name_and_params = null;
377
string_value = null;
378
}
379
}
318
380
319
381
// if we haven't yet got a version, we won't be paring anything!
320
382
if( _version == null )
321
383
{
322
384
// is this a version?
323
if( props.length == 2 && props[ 0 ].equals( "VERSION" ) )
385
if( name_and_params.equals( "VERSION" ) )
324
386
{
325
387
// yes, check/store it
326
if( !props[ 1 ].equals( "2.1" ) &&
327
!props[ 1 ].equals( "3.0" ) )
388
if( !string_value.equals( "2.1" ) &&
389
!string_value.equals( "3.0" ) )
328
390
throw new ParseException( R.string.error_vcf_version );
329
_version = props[ 1 ];
391
_version = string_value;
330
392
331
// parse any other lines we've accumulated so far
332
if( _lines != null )
333
for( int i = 0; i < _lines.size(); i++ )
334
parseLine( _lines.get( i ) );
335
_lines = null;
393
// parse any other buffers we've accumulated so far
394
if( _buffers != null )
395
for( int i = 0; i < _buffers.size(); i++ )
396
parseLine( _buffers.get( i ) );
397
_buffers = null;
336
398
}
337
399
else
338
400
{
339
// no, so stash this line till we have a version
340
if( _lines == null )
341
_lines = new Vector< String >();
342
_lines.add( line );
401
// no, so stash this buffer till we have a version
402
if( _buffers == null )
403
_buffers = new Vector< ByteBuffer >();
404
_buffers.add( buffer );
343
405
}
344
406
}
345
407
else
346
408
{
409
// value bytes, for processing
410
ByteBuffer value;
411
347
412
if( _parser_in_multiline )
348
413
{
349
414
// if we're currently in a multi-line value, use the stored
350
415
// property name and parameters
351
props = new String[ 2 ];
352
props[ 0 ] = _parser_current_name_and_params;
353
props[ 1 ] = line.trim();
416
name_and_params = _parser_current_name_and_params;
417
418
// find start of string (skip spaces/tabs)
419
int pos = buffer.position();
420
byte[] buffer_array = buffer.array();
421
while( pos < buffer.limit() && (
422
buffer_array[ pos ] == ' ' ||
423
buffer_array[ pos ] == '\t' ) )
424
{
425
pos++;
426
}
427
428
// get value from buffer
429
value = ByteBuffer.wrap( buffer.array(), pos,
430
buffer.limit() - pos );
354
431
}
355
432
else
356
433
{
357
// for normal lines, check the property name/value bits
358
if( props.length < 2 || props[ 0 ].length() == 0 )
359
throw new ParseException(
360
R.string.error_vcf_malformed );
361
362
// ignore empty properties
363
if( props[ 1 ].length() < 1 )
364
return;
434
// ignore empty values
435
if( string_value.length() < 1 ) return;
436
437
// calculate how many chars to skip from beginning of line
438
// so we skip the property "name:" part
439
int pos = buffer.position() + name_and_params.length() + 1;
440
441
// get value from buffer
442
value = ByteBuffer.wrap( buffer.array(), pos,
443
buffer.limit() - pos );
365
444
366
445
// reset the saved multi-line state
367
_parser_current_name_and_params = props[ 0 ];
446
_parser_current_name_and_params = name_and_params;
368
447
_parser_buffered_value_so_far = "";
369
448
}
370
449
371
450
// get parameter parts
372
String[] params = props[ 0 ].split( ";" );
373
for( int i = 0; i < params.length; i++ )
374
params[ i ] = params[ i ].trim();
451
String[] name_and_param_bits = name_and_params.split( ";" );
452
for( int i = 0; i < name_and_param_bits.length; i++ )
453
name_and_param_bits[ i ] = name_and_param_bits[ i ].trim();
375
454
376
// parse charset and encoding parameters
377
String charset, encoding;
378
if( ( charset = checkParam( params, "CHARSET" ) ) != null &&
379
!charset.equals( "UTF-8" ) && !charset.equals( "UTF-16" ) )
380
{
381
throw new ParseException( R.string.error_vcf_charset );
382
}
383
if( ( encoding = checkParam( params, "ENCODING" ) ) != null &&
384
!encoding.equals( "QUOTED-PRINTABLE" ) &&
385
!encoding.equals( "8BIT" ) )
455
// parse encoding parameter
456
String encoding = checkParam( name_and_param_bits, "ENCODING" );
457
if( encoding != null ) encoding = encoding.toUpperCase();
458
if( encoding != null && !encoding.equals( "8BIT" ) &&
459
!encoding.equals( "QUOTED-PRINTABLE" ) )
386
460
//&& !encoding.equals( "BASE64" ) )
387
461
{
388
462
throw new ParseException( R.string.error_vcf_encoding );
389
463
}
390
464
465
// parse charset parameter
466
String charset = checkParam( name_and_param_bits, "CHARSET" );
467
if( charset != null ) charset = charset.toUpperCase();
468
if( charset != null && !charset.equals( "US-ASCII" ) &&
469
!charset.equals( "ASCII" ) && !charset.equals( "UTF-8" ) )
470
{
471
throw new ParseException( R.string.error_vcf_charset );
472
}
473
391
474
// do unencoding (or default to a fake unencoding result with
392
475
// the raw string)
393
UnencodeResult result;
476
UnencodeResult unencoding_result = null;
394
477
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
395
result = unencodeQuotedPrintable( props[ 1 ], charset );
478
unencoding_result = unencodeQuotedPrintable( value );
396
479
// else if( encoding != null && encoding.equals( "BASE64" ) )
397
480
// result = unencodeBase64( props[ 1 ], charset );
398
else
399
result = new UnencodeResult( false, props[ 1 ].getBytes(),
400
props[ 1 ].getBytes().length );
481
if( unencoding_result != null ) {
482
value = unencoding_result.getBuffer();
483
_parser_in_multiline =
484
unencoding_result.isAnotherLineRequired();
485
}
486
487
// convert 8-bit ASCII charset to US-ASCII
488
if( charset == null || charset == "ASCII" ) {
489
value = transcodeAsciiToUtf8( value );
490
charset = "UTF-8";
491
}
401
492
402
493
// process charset
403
494
try {
404
props[ 1 ] = new String( result.getBytes(), 0,
405
result.getNumBytes(),
406
charset == null? "UTF-8" : charset );
495
string_value =
496
new String( value.array(), value.position(),
497
value.limit() - value.position(), charset );
407
498
} catch( UnsupportedEncodingException e ) {
408
499
throw new ParseException( R.string.error_vcf_charset );
409
500
}
410
501
411
502
// handle multi-line requests
412
_parser_in_multiline = result.isAnotherLineRequired();
413
503
if( _parser_in_multiline ) {
414
_parser_buffered_value_so_far += props[ 1 ];
504
_parser_buffered_value_so_far += string_value;
415
505
return;
416
506
}
417
507
418
508
// add on buffered multi-line content
419
String value = _parser_buffered_value_so_far + props[ 1 ];
509
String complete_value =
510
_parser_buffered_value_so_far + string_value;
420
511
421
512
// parse some properties
422
if( params[ 0 ].equals( "N" ) )
423
parseN( params, value );
424
else if( params[ 0 ].equals( "FN" ) )
425
parseFN( params, value );
426
else if( params[ 0 ].equals( "ORG" ) )
427
parseORG( params, value );
428
else if( params[ 0 ].equals( "TEL" ) )
429
parseTEL( params, value );
430
else if( params[ 0 ].equals( "EMAIL" ) )
431
parseEMAIL( params, value );
513
if( name_and_param_bits[ 0 ].equals( "N" ) )
514
parseN( name_and_param_bits, complete_value );
515
else if( name_and_param_bits[ 0 ].equals( "FN" ) )
516
parseFN( name_and_param_bits, complete_value );
517
else if( name_and_param_bits[ 0 ].equals( "ORG" ) )
518
parseORG( name_and_param_bits, complete_value );
519
else if( name_and_param_bits[ 0 ].equals( "TEL" ) )
520
parseTEL( name_and_param_bits, complete_value );
521
else if( name_and_param_bits[ 0 ].equals( "EMAIL" ) )
522
parseEMAIL( name_and_param_bits, complete_value );
432
523
}
433
524
}
434
525
542
633
AbortImportException
543
634
{
544
635
// missing version (and data is present)
545
if( _version == null && _lines != null )
636
if( _version == null && _buffers != null )
546
637
throw new ParseException( R.string.error_vcf_malformed );
547
638
548
639
// missing name properties?
591
682
return types;
592
683
}
593
684
594
private UnencodeResult unencodeQuotedPrintable( String str, String charset )
685
private UnencodeResult unencodeQuotedPrintable( ByteBuffer in )
595
686
{
596
687
boolean another = false;
597
688
598
// default encoding scheme
599
if( charset == null ) charset = "UTF-8";
600
601
689
// unencode quoted-pritable encoding, as per RFC1521 section 5.1
602
byte[] bytes = new byte[ str.length() ];
690
byte[] out = new byte[ in.limit() - in.position() ];
603
691
int j = 0;
604
for( int i = 0; i < str.length(); i++ )
692
for( int i = in.position(); i < in.limit(); i++ )
605
693
{
606
694
// get next char and process...
607
char ch = str.charAt( i );
608
if( ch == '=' && i < str.length() - 2 )
695
byte ch = in.array()[ i ];
696
if( ch == '=' && i < in.limit() - 2 )
609
697
{
610
698
// we found a =XX format byte, add it
611
bytes[ j ] = (byte)(
612
Character.digit( str.charAt( i + 1 ), 16 ) * 16 +
613
Character.digit( str.charAt( i + 2 ), 16 ) );
699
out[ j ] = (byte)(
700
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
701
Character.digit( in.array()[ i + 2 ], 16 ) );
614
702
i += 2;
615
703
}
616
else if( ch == '=' && i == str.length() - 1 )
704
else if( ch == '=' && i == in.limit() - 1 )
617
705
{
618
706
// we found a '=' at the end of a line signifying a multi-
619
707
// line string, so we don't add it.
622
710
}
623
711
else
624
712
// just a normal char...
625
bytes[ j ] = (byte)ch;
713
out[ j ] = (byte)ch;
626
714
j++;
627
715
}
628
716
629
return new UnencodeResult( another, bytes, j );
717
return new UnencodeResult( another, ByteBuffer.wrap( out, 0, j ) );
718
}
719
720
private ByteBuffer transcodeAsciiToUtf8( ByteBuffer in )
721
{
722
// transcode
723
byte[] out = new byte[ ( in.limit() - in.position() ) * 2 ];
724
int j = 0;
725
for( int a = in.position(); a < in.limit(); a++ )
726
{
727
// if char is < 127, keep it as-is
728
if( in.array()[ a ] >= 0 )
729
out[ j++ ] = in.array()[ a ];
730
731
// else, convert it to UTF-8
732
else {
733
int b = 0xff & (int)in.array()[ a ];
734
out[ j++ ] = (byte)( 0xc0 | ( b >> 6 ) );
735
out[ j++ ] = (byte)( 0x80 | ( b & 0x3f ) );
736
}
737
}
738
739
return ByteBuffer.wrap( out, 0, j );
630
740
}
631
741
}
632
742
}
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480