(root)/android/import-contacts
» Revision
36
(compared to revision 105)
: 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 36
- compare with another revision
- download diff
- download tarball
- view history from revision 36
- Committer: edam
- Date: 2011-03-19 20:33:09 UTC
- Revision ID: edam@waxworlds.org-20110319203309-5dzfyqrxwk94jtin
- formatting: removed some double-indents on overrunning lines
- updated TODO and NEWS
- rewrote central logic of parser so it makes more sense, looks nicer and has a small optimisation (getting name and params from line only when necessary)
- optimised unnecessary mutliple converting of lines to US-ASCII
- re-wrote line extraction from vcards so that we can lookahead for v3 folded lines
- added support for v3 folded lines
- updated TODO and NEWS
- rewrote central logic of parser so it makes more sense, looks nicer and has a small optimisation (getting name and params from line only when necessary)
- optimised unnecessary mutliple converting of lines to US-ASCII
- re-wrote line extraction from vcards so that we can lookahead for v3 folded lines
- added support for v3 folded lines
- files added:
- TODO
- gen/org
- gen/org/waxworlds
- gen/org/waxworlds/edam
- gen/org/waxworlds/edam/importcontacts
- src/org
- src/org/waxworlds
- src/org/waxworlds/edam
- files removed:
- gen/am
- gen/am/ed
- gen/am/ed/importcontacts
- src/am
- src/am/ed
- files renamed:
- src/am/ed/importcontacts/ => src/org/waxworlds/edam/importcontacts/
- src/am/ed/importcontacts/VcardImporter.java => src/org/waxworlds/edam/importcontacts/VCFImporter.java
- files modified:
- .bzrignore
added
removed
src/org/waxworlds/edam/importcontacts/VCFImporter.java
1
1
/*
2
2
* VCFImporter.java
3
3
*
4
* Copyright (C) 2009 to 2013 Tim Marston <tim@ed.am>
4
* Copyright (C) 2009 Tim Marston <edam@waxworlds.org>
5
5
*
6
6
* This file is part of the Import Contacts program (hereafter referred
7
* to as "this program"). For more information, see
8
* http://ed.am/dev/android/import-contacts
7
* to as "this program"). For more information, see
8
* http://www.waxworlds.org/edam/software/android/import-contacts
9
9
*
10
10
* This program is free software: you can redistribute it and/or modify
11
11
* it under the terms of the GNU General Public License as published by
21
21
* along with this program. If not, see <http://www.gnu.org/licenses/>.
22
22
*/
23
23
24
package am.ed.importcontacts;
24
package org.waxworlds.edam.importcontacts;
25
25
26
26
import java.io.BufferedReader;
27
27
import java.io.File;
32
32
import java.io.IOException;
33
33
import java.io.UnsupportedEncodingException;
34
34
import java.nio.ByteBuffer;
35
import java.util.ArrayList;
36
35
import java.util.Arrays;
37
import java.util.HashMap;
38
36
import java.util.HashSet;
39
37
import java.util.Iterator;
40
38
import java.util.List;
41
import java.util.Locale;
42
import java.util.NoSuchElementException;
43
39
import java.util.Set;
44
40
import java.util.Vector;
45
41
import java.util.regex.Matcher;
46
42
import java.util.regex.Pattern;
43
import java.util.NoSuchElementException;
44
import java.lang.UnsupportedOperationException;
47
45
48
46
import android.content.SharedPreferences;
49
import android.os.Environment;
47
import android.provider.Contacts;
48
import android.provider.Contacts.PhonesColumns;
50
49
51
public class VcardImporter extends Importer
50
public class VCFImporter extends Importer
52
51
{
53
private int _vcard_count = 0;
52
private int _vCardCount = 0;
54
53
private int _progress = 0;
55
54
56
public VcardImporter( Doit doit )
55
public VCFImporter( Doit doit )
57
56
{
58
57
super( doit );
59
58
}
70
69
File[] files = null;
71
70
try
72
71
{
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 ) )
77
{
78
showError( R.string.error_nosdcard );
79
}
80
81
72
// open directory
82
File file = new File( Environment.getExternalStorageDirectory(),
83
prefs.getString( "location", "/" ) );
73
String path = "/sdcard" + prefs.getString( "location", "/" );
74
File file = new File( path );
84
75
if( !file.exists() )
85
76
showError( R.string.error_locationnotfound );
86
77
90
81
// get files
91
82
class VCardFilter implements FilenameFilter {
92
83
public boolean accept( File dir, String name ) {
93
return name.toLowerCase( Locale.US ).endsWith( ".vcf" );
84
return name.toLowerCase().endsWith( ".vcf" );
94
85
}
95
86
}
96
87
files = file.listFiles( new VCardFilter() );
118
109
countVCardFile( files[ i ] );
119
110
setTmpProgress( i );
120
111
}
121
setProgressMax( _vcard_count ); // will also update tmp progress
112
setProgressMax( _vCardCount ); // will also update tmp progress
122
113
123
114
// import them
124
115
setProgress( 0 );
125
116
for( int i = 0; i < files.length; i++ )
126
117
importVCardFile( files[ i ] );
127
setProgress( _vcard_count );
128
118
}
129
119
130
120
private void countVCardFile( File file ) throws AbortImportException
137
127
138
128
// read
139
129
String line;
140
boolean in_vcard = false;
130
boolean inVCard = false;
141
131
while( ( line = reader.readLine() ) != null )
142
132
{
143
if( !in_vcard )
144
{
133
if( !inVCard ) {
145
134
// look for vcard beginning
146
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
147
in_vcard = true;
148
_vcard_count++;
149
}
150
// check for vMsg files
151
else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
152
showError( getText( R.string.error_vcf_vmsgfile )
153
+ file.getName() );
135
if( line.matches( "^BEGIN:VCARD" ) ) {
136
inVCard = true;
137
_vCardCount++;
154
138
}
155
139
}
156
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
157
in_vcard = false;
140
else if( line.matches( "^END:VCARD" ) )
141
inVCard = false;
158
142
}
159
reader.close();
160
143
161
144
}
162
145
catch( FileNotFoundException e ) {
184
167
FileInputStream istream = new FileInputStream( file );
185
168
byte[] content = new byte[ (int)file.length() ];
186
169
istream.read( content );
187
istream.close();
188
170
189
171
// import
190
172
importVCardFileContent( content, file.getName() );
191
173
}
192
catch( OutOfMemoryError e ) {
193
showError( R.string.error_outofmemory );
194
}
195
174
catch( FileNotFoundException e ) {
196
175
showError( getText( R.string.error_filenotfound ) +
197
176
file.getName() );
205
184
throws AbortImportException
206
185
{
207
186
// go through lines
208
Vcard vcard = null;
209
int vcard_start_line = 0;
187
VCard vCard = null;
210
188
ContentLineIterator cli = new ContentLineIterator( content );
211
189
while( cli.hasNext() )
212
190
{
213
ContentLine content_line = cli.next();
214
215
// get a US-ASCII version of the string, for processing
216
String line = content_line.getUsAsciiLine();
217
218
if( vcard == null ) {
191
ByteBuffer buffer = cli.next();
192
193
// get a US-ASCII version of the line for processing
194
String line;
195
try {
196
line = new String( buffer.array(), buffer.position(),
197
buffer.limit() - buffer.position(), "US-ASCII" );
198
}
199
catch( UnsupportedEncodingException e ) {
200
// we know US-ASCII is supported, so appease the compiler...
201
line = "";
202
}
203
204
if( vCard == null ) {
219
205
// look for vcard beginning
220
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
221
setProgress( _progress++ );
222
vcard = new Vcard();
223
vcard_start_line = cli.getLineNumber();
206
if( line.matches( "^BEGIN:VCARD" ) ) {
207
setProgress( ++_progress );
208
vCard = new VCard();
224
209
}
225
210
}
226
211
else {
227
212
// look for vcard content or ending
228
if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
213
if( line.matches( "^END:VCARD" ) )
229
214
{
230
// finalise the vcard/contact
215
// store vcard and do away with it
231
216
try {
232
vcard.finaliseVcard();
233
234
// pass the finalised contact to the importer
235
importContact( vcard );
236
}
237
catch( Vcard.ParseException e ) {
238
if( !showContinue(
239
getText( R.string.error_vcf_parse ).toString()
240
+ fileName +
241
getText( R.string.error_vcf_parse_line ).toString()
242
+ cli.getLineNumber() + ":\n" + e.getMessage() ) )
243
{
244
finish( ACTION_ABORT );
245
}
246
else
247
skipContact();
248
}
249
catch( ContactData.ContactNotIdentifiableException e ) {
250
if( !showContinue(
251
getText( R.string.error_vcf_parse ).toString()
252
+ fileName +
253
getText( R.string.error_vcf_parse_line ).toString()
254
+ vcard_start_line + ":\n" + getText(
255
R.string.error_vcf_notenoughinfo ).toString()
256
) )
257
{
258
finish( ACTION_ABORT );
259
}
260
else
261
skipContact();
262
}
263
264
// discard this vcard
265
vcard = null;
217
vCard.finaliseParsing();
218
importContact( vCard );
219
}
220
catch( VCard.ParseException e ) {
221
skipContact();
222
if( !showContinue(
223
getText( R.string.error_vcf_parse ).toString()
224
+ fileName + "\n" + e.getMessage() ) )
225
{
226
finish( ACTION_ABORT );
227
}
228
}
229
catch( VCard.SkipContactException e ) {
230
skipContact();
231
// do nothing
232
}
233
vCard = null;
266
234
}
267
235
else
268
236
{
269
237
// try giving the line to the vcard
270
238
try {
271
vcard.parseLine( content_line );
239
vCard.parseLine( buffer, line,
240
cli.doesNextLineLookFolded() );
272
241
}
273
catch( Vcard.ParseException e ) {
242
catch( VCard.ParseException e ) {
274
243
skipContact();
275
244
if( !showContinue(
276
245
getText( R.string.error_vcf_parse ).toString()
277
+ fileName +
278
getText( R.string.error_vcf_parse_line ).toString()
279
+ cli.getLineNumber() + "\n" + e.getMessage() ) )
246
+ fileName + "\n" + e.getMessage() ) )
280
247
{
281
248
finish( ACTION_ABORT );
282
249
}
283
250
284
// Although we're continuing, we still need to abort
285
// this vCard. Further lines will be ignored until we
251
// although we're continuing, we still need to abort
252
// this vCard. Further lines will be ignored until we
286
253
// get to another BEGIN:VCARD line.
287
vcard = null;
254
vCard = null;
288
255
}
289
catch( Vcard.SkipImportException e ) {
256
catch( VCard.SkipContactException e ) {
290
257
skipContact();
291
// Abort this vCard. Further lines will be ignored until
258
// abort this vCard. Further lines will be ignored until
292
259
// we get to another BEGIN:VCARD line.
293
vcard = null;
260
vCard = null;
294
261
}
295
262
}
296
263
}
297
264
}
298
265
}
299
266
300
class ContentLine
301
{
302
private ByteBuffer _buffer;
303
private boolean _folded_next;
304
private String _line;
305
306
public ContentLine( ByteBuffer buffer, boolean folded_next )
307
{
308
_buffer = buffer;
309
_folded_next = folded_next;
310
_line = null;
311
}
312
313
public ByteBuffer getBuffer()
314
{
315
return _buffer;
316
}
317
318
public boolean doesNextLineLookFolded()
319
{
320
return _folded_next;
321
}
322
323
public String getUsAsciiLine()
324
{
325
// generated line and cache it
326
if( _line == null ) {
327
try {
328
_line = new String( _buffer.array(), _buffer.position(),
329
_buffer.limit() - _buffer.position(), "US-ASCII" );
330
}
331
catch( UnsupportedEncodingException e ) {
332
// we know US-ASCII *is* supported, so appease the
333
// compiler...
334
}
335
}
336
337
// return cached line
338
return _line;
339
}
340
}
341
342
class ContentLineIterator implements Iterator< ContentLine >
267
class ContentLineIterator implements Iterator< ByteBuffer >
343
268
{
344
269
protected byte[] _content = null;
345
270
protected int _pos = 0;
346
protected int _line = 0;
347
271
348
272
public ContentLineIterator( byte[] content )
349
273
{
357
281
}
358
282
359
283
@Override
360
public ContentLine next()
284
public ByteBuffer next()
361
285
{
362
286
int initial_pos = _pos;
363
287
369
293
int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
370
294
_pos > initial_pos )? _pos - 1 : _pos;
371
295
_pos++;
372
_line++;
373
return new ContentLine(
374
ByteBuffer.wrap( _content, initial_pos,
375
to - initial_pos ),
376
doesNextLineLookFolded() );
296
return ByteBuffer.wrap( _content, initial_pos,
297
to - initial_pos );
377
298
}
378
299
379
300
// we didn't find one, but were there bytes left?
380
301
if( _pos != initial_pos ) {
381
302
int to = _pos;
382
303
_pos++;
383
_line++;
384
return new ContentLine(
385
ByteBuffer.wrap( _content, initial_pos,
386
to - initial_pos ),
387
doesNextLineLookFolded() );
304
return ByteBuffer.wrap( _content, initial_pos,
305
to - initial_pos );
388
306
}
389
307
390
308
// no bytes left
402
320
* onto the end of this one?
403
321
* @return
404
322
*/
405
private boolean doesNextLineLookFolded()
323
public boolean doesNextLineLookFolded()
406
324
{
407
325
return _pos > 0 && _pos < _content.length &&
408
_content[ _pos - 1 ] == '\n' &&
409
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
410
}
411
412
public int getLineNumber()
413
{
414
return _line;
326
_content[ _pos - 1 ] == '\n' && _content[ _pos ] == ' ';
415
327
}
416
328
}
417
329
418
private class Vcard extends ContactData
330
private class VCard extends ContactData
419
331
{
420
332
private final static int NAMELEVEL_NONE = 0;
421
private final static int NAMELEVEL_N = 1;
333
private final static int NAMELEVEL_ORG = 1;
422
334
private final static int NAMELEVEL_FN = 2;
423
424
private final static int MULTILINE_NONE = 0;
425
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
426
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
427
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
335
private final static int NAMELEVEL_N = 3;
428
336
429
337
private String _version = null;
430
private Vector< ContentLine > _content_lines = null;
338
private Vector< ByteBuffer > _buffers = null;
431
339
private int _name_level = NAMELEVEL_NONE;
432
private int _parser_multiline_state = MULTILINE_NONE;
340
private boolean _parser_in_encoded_multiline = false;
341
private boolean _parser_in_folded_multiline = false;
433
342
private String _parser_current_name_and_params = null;
434
343
private String _parser_buffered_value_so_far = "";
435
private String _cached_organisation = null;
436
private String _cached_title = null;
437
344
438
345
protected class UnencodeResult
439
346
{
469
376
470
377
public ParseException( int res )
471
378
{
472
super( VcardImporter.this.getText( res ).toString() );
379
super( VCFImporter.this.getText( res ).toString() );
473
380
}
474
381
}
475
382
476
383
@SuppressWarnings("serial")
477
protected class SkipImportException extends Exception { }
384
protected class SkipContactException extends Exception { }
478
385
479
private String extractCollonPartFromLine( ContentLine content_line,
480
boolean former )
386
private String extractCollonPartFromLine( ByteBuffer buffer,
387
String line, boolean former )
481
388
{
389
String ret = null;
390
391
// get a US-ASCII version of the line for processing, unless we were
392
// supplied with one
393
if( line == null ) {
394
try {
395
line = new String( buffer.array(), buffer.position(),
396
buffer.limit() - buffer.position(), "US-ASCII" );
397
}
398
catch( UnsupportedEncodingException e ) {
399
// we know US-ASCII is supported, so appease the compiler...
400
line = "";
401
}
402
}
403
482
404
// split line into name and value parts and check to make sure we
483
405
// only got 2 parts and that the first part is not zero in length
484
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
406
String[] parts = line.split( ":", 2 );
485
407
if( parts.length == 2 && parts[ 0 ].length() > 0 )
486
return parts[ former? 0 : 1 ].trim();
487
488
return null;
489
}
490
491
private String extractNameAndParamsFromLine( ContentLine content_line )
492
{
493
return extractCollonPartFromLine( content_line, true );
494
}
495
496
private String extractValueFromLine( ContentLine content_line )
497
{
498
return extractCollonPartFromLine( content_line, false );
499
}
500
501
public void parseLine( ContentLine content_line )
502
throws ParseException, SkipImportException,
408
ret = parts[ former? 0 : 1 ];
409
410
return ret;
411
}
412
413
private String extractNameAndParamsFromLine( ByteBuffer buffer,
414
String line )
415
{
416
return extractCollonPartFromLine( buffer, line, true );
417
}
418
419
private String extractValueFromLine( ByteBuffer buffer, String line )
420
{
421
return extractCollonPartFromLine( buffer, line, false );
422
}
423
424
public void parseLine( ByteBuffer buffer, String line,
425
boolean next_line_looks_folded )
426
throws ParseException, SkipContactException,
503
427
AbortImportException
504
428
{
505
429
// do we have a version yet?
507
431
{
508
432
// tentatively get name and params from line
509
433
String name_and_params =
510
extractNameAndParamsFromLine( content_line );
434
extractNameAndParamsFromLine( buffer, line );
511
435
512
436
// is it a version line?
513
437
if( name_and_params != null &&
514
name_and_params.equalsIgnoreCase( "VERSION" ) )
438
name_and_params.equals( "VERSION" ) )
515
439
{
516
440
// yes, get it!
517
String value = extractValueFromLine( content_line );
518
if( value == null || (
519
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
520
{
441
String value = extractValueFromLine( buffer, line );
442
if( !value.equals( "2.1" ) && !value.equals( "3.0" ) )
521
443
throw new ParseException( R.string.error_vcf_version );
522
}
523
444
_version = value;
524
445
525
446
// parse any buffers we've been accumulating while we waited
526
447
// for a version
527
if( _content_lines != null )
528
for( int i = 0; i < _content_lines.size(); i++ )
529
parseLine( _content_lines.get( i ) );
530
_content_lines = null;
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() ) == ' ' );
455
_buffers = null;
531
456
}
532
457
else
533
458
{
534
459
// no, so stash this line till we get a version
535
if( _content_lines == null )
536
_content_lines = new Vector< ContentLine >();
537
_content_lines.add( content_line );
460
if( _buffers == null )
461
_buffers = new Vector< ByteBuffer >();
462
_buffers.add( buffer );
538
463
}
539
464
}
540
465
else
541
466
{
542
467
// name and params and the position in the buffer where the
543
// "value" part of the line starts
468
// "value" part of the line start
544
469
String name_and_params;
545
470
int pos;
546
471
547
if( _parser_multiline_state != MULTILINE_NONE )
472
if( _parser_in_encoded_multiline ||
473
_parser_in_folded_multiline )
548
474
{
549
475
// if we're currently in a multi-line value, use the stored
550
476
// property name and parameters
551
477
name_and_params = _parser_current_name_and_params;
552
478
553
// skip some initial line characters, depending on the type
554
// of multi-line we're handling
555
pos = content_line.getBuffer().position();
556
switch( _parser_multiline_state )
557
{
558
case MULTILINE_FOLDED:
479
pos = buffer.position();
480
481
// for folded multi-lines, skip the single space at the
482
// start of the next line
483
if( _parser_in_folded_multiline )
559
484
pos++;
560
break;
561
case MULTILINE_ENCODED:
562
while( pos < content_line.getBuffer().limit() && (
563
content_line.getBuffer().get( pos ) == ' ' ||
564
content_line.getBuffer().get( pos ) == '\t' ) )
485
486
// else, this must be an encoded multi-line, so skip any
487
// whitespace we find at the start of the next line
488
else
489
while( pos < buffer.limit() && (
490
buffer.get( pos ) == ' ' ||
491
buffer.get( pos ) == '\t' ) )
565
492
{
566
493
pos++;
567
494
}
568
break;
569
default:
570
// do nothing
571
}
572
573
// take us out of multi-line so that we can re-detect that
574
// this line is a multi-line or not
575
_parser_multiline_state = MULTILINE_NONE;
576
495
}
577
496
else
578
497
{
579
// skip empty lines
580
if( content_line.getUsAsciiLine().trim().length() == 0 )
581
return;
582
583
498
// get name and params from line, and since we're not
584
499
// parsing a subsequent line in a multi-line, this should
585
500
// not fail, or it's an error
586
501
name_and_params =
587
extractNameAndParamsFromLine( content_line );
502
extractNameAndParamsFromLine( buffer, line );
588
503
if( name_and_params == null )
589
504
throw new ParseException(
590
505
R.string.error_vcf_malformed );
591
506
592
507
// calculate how many chars to skip from beginning of line
593
508
// so we skip the property "name:" part
594
pos = content_line.getBuffer().position() +
595
name_and_params.length() + 1;
509
pos = buffer.position() + name_and_params.length() + 1;
596
510
597
511
// reset the saved multi-line state
598
512
_parser_current_name_and_params = name_and_params;
601
515
602
516
// get value from buffer, as raw bytes
603
517
ByteBuffer value;
604
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
605
content_line.getBuffer().limit() - pos );
518
value = ByteBuffer.wrap( buffer.array(), pos,
519
buffer.limit() - pos );
606
520
607
521
// get parameter parts
608
522
String[] name_param_parts = name_and_params.split( ";", -1 );
609
523
for( int i = 0; i < name_param_parts.length; i++ )
610
524
name_param_parts[ i ] = name_param_parts[ i ].trim();
611
525
612
// determine whether we care about this entry
613
final HashSet< String > interesting_fields =
614
new HashSet< String >( Arrays.asList( new String[] { "N",
615
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
616
) );
617
boolean is_interesting_field =
618
interesting_fields.contains(
619
name_param_parts[ 0 ].toUpperCase( Locale.US ) );
620
621
526
// parse encoding parameter
622
527
String encoding = checkParam( name_param_parts, "ENCODING" );
623
if( encoding != null )
624
encoding = encoding.toUpperCase( Locale.US );
625
if( is_interesting_field && encoding != null &&
626
!encoding.equalsIgnoreCase( "8BIT" ) &&
627
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
628
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
528
if( encoding != null ) encoding = encoding.toUpperCase();
529
if( encoding != null && !encoding.equals( "8BIT" ) &&
530
!encoding.equals( "QUOTED-PRINTABLE" ) )
531
//&& !encoding.equals( "BASE64" ) )
629
532
{
630
533
throw new ParseException( R.string.error_vcf_encoding );
631
534
}
632
535
633
536
// parse charset parameter
634
537
String charset = checkParam( name_param_parts, "CHARSET" );
635
if( charset != null )
636
charset = charset.toUpperCase( Locale.US );
637
if( charset != null &&
638
!charset.equalsIgnoreCase( "US-ASCII" ) &&
639
!charset.equalsIgnoreCase( "ASCII" ) &&
640
!charset.equalsIgnoreCase( "UTF-8" ) )
538
if( charset != null ) charset = charset.toUpperCase();
539
if( charset != null && !charset.equals( "US-ASCII" ) &&
540
!charset.equals( "ASCII" ) &&
541
!charset.equals( "UTF-8" ) )
641
542
{
642
543
throw new ParseException( R.string.error_vcf_charset );
643
544
}
645
546
// do unencoding (or default to a fake unencoding result with
646
547
// the raw string)
647
548
UnencodeResult unencoding_result = null;
648
if( encoding != null &&
649
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
650
{
549
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
651
550
unencoding_result = unencodeQuotedPrintable( value );
652
}
653
// else if( encoding != null &&
654
// encoding.equalsIgnoreCase( "BASE64" ) )
655
// {
551
// else if( encoding != null && encoding.equals( "BASE64" ) )
656
552
// unencoding_result = unencodeBase64( props[ 1 ], charset );
657
// }
658
553
if( unencoding_result != null ) {
659
554
value = unencoding_result.getBuffer();
660
if( unencoding_result.isAnotherLineRequired() )
661
_parser_multiline_state = MULTILINE_ENCODED;
555
_parser_in_encoded_multiline =
556
unencoding_result.isAnotherLineRequired();
662
557
}
663
558
664
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
665
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
666
if( ( charset == null && _version.equals( "2.1" ) ) ||
667
( charset != null && (
668
charset.equalsIgnoreCase( "ASCII" ) ||
669
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
670
{
559
// convert 8-bit ASCII charset to US-ASCII
560
if( charset == null || charset.equals( "ASCII" ) ) {
671
561
value = transcodeAsciiToUtf8( value );
562
charset = "UTF-8";
672
563
}
673
564
674
// process charset (value is now in UTF-8)
565
// process charset
675
566
String string_value;
676
567
try {
677
568
string_value = new String( value.array(), value.position(),
678
value.limit() - value.position(), "UTF-8" );
569
value.limit() - value.position(), charset );
679
570
} catch( UnsupportedEncodingException e ) {
680
571
throw new ParseException( R.string.error_vcf_charset );
681
572
}
682
573
683
// for some entries that have semicolon-separated value parts,
684
// check to see if the value ends in an escape character, which
685
// indicates that we have a multi-line value
686
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
687
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
688
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
689
doesStringEndInAnEscapeChar( string_value ) )
690
{
691
_parser_multiline_state = MULTILINE_ESCAPED;
692
string_value = string_value.substring( 0,
693
string_value.length() - 1 );
694
}
695
696
// if we know we're not in an encoding-based multi-line, check
697
// to see if we're in a folded multi-line
698
if( _parser_multiline_state == MULTILINE_NONE &&
699
content_line.doesNextLineLookFolded() )
700
{
701
_parser_multiline_state = MULTILINE_FOLDED;
702
}
703
704
// handle multi-lines by buffering them and parsing them when we
705
// are processing the last line in a multi-line sequence
706
if( _parser_multiline_state != MULTILINE_NONE ) {
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;
578
579
// handle multi-line requests
580
if( _parser_in_encoded_multiline ||
581
_parser_in_folded_multiline )
582
{
707
583
_parser_buffered_value_so_far += string_value;
708
584
return;
709
585
}
586
587
// add on buffered multi-line content
710
588
String complete_value =
711
( _parser_buffered_value_so_far + string_value ).trim();
589
_parser_buffered_value_so_far + string_value;
712
590
713
591
// ignore empty values
714
592
if( complete_value.length() < 1 ) return;
715
593
716
594
// parse some properties
717
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
595
if( name_param_parts[ 0 ].equals( "N" ) )
718
596
parseN( name_param_parts, complete_value );
719
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
597
else if( name_param_parts[ 0 ].equals( "FN" ) )
720
598
parseFN( name_param_parts, complete_value );
721
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
599
else if( name_param_parts[ 0 ].equals( "ORG" ) )
722
600
parseORG( name_param_parts, complete_value );
723
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
724
parseTITLE( name_param_parts, complete_value );
725
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
601
else if( name_param_parts[ 0 ].equals( "TEL" ) )
726
602
parseTEL( name_param_parts, complete_value );
727
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
603
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
728
604
parseEMAIL( name_param_parts, complete_value );
729
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
730
parseADR( name_param_parts, complete_value );
731
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
732
parseLABEL( name_param_parts, complete_value );
733
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
734
parseNOTE( name_param_parts, complete_value );
735
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
736
parseBDAY( name_param_parts, complete_value );
737
}
738
}
739
740
private boolean doesStringEndInAnEscapeChar( String string )
741
{
742
// count the number of backslashes at the end of the string
743
int count = 0;
744
for( int a = string.length() - 1; a >= 0; a-- )
745
if( string.charAt( a ) == '\\' )
746
count++;
747
else
748
break;
749
750
// if there are an even number of backslashes then the final one
751
// doesn't count
752
return ( count & 1 ) == 1;
753
}
754
755
private String[] splitValueByCharacter( String value, char character )
756
{
757
// split string in to parts by specified character
758
ArrayList< String > parts = new ArrayList< String >(
759
Arrays.asList( value.split( "" + character ) ) );
760
761
// go through parts
762
for( int a = 0; a < parts.size(); a++ )
763
{
764
String str = parts.get( a );
765
766
// Look for parts that end in an escape character, but ignore
767
// the final part. We've already detected escape chars at the
768
// end of the final part in parseLine() and handled multi-lines
769
// accordingly.
770
if( a < parts.size() - 1 &&
771
doesStringEndInAnEscapeChar( str ) )
772
{
773
// append the escaped character, join the next part to this
774
// part and remove the next part
775
parts.set( a, str.substring( 0, str.length() - 1 ) +
776
character + parts.get( a + 1 ) );
777
parts.remove( a + 1 );
778
779
// re-visit this part
780
a--;
781
continue;
782
}
783
784
// trim and replace string
785
str = str.trim();
786
parts.set( a, str );
787
}
788
789
String[] ret = new String[ parts.size() ];
790
return parts.toArray( ret );
791
}
792
793
private String unescapeValue( String value )
794
{
795
StringBuilder ret = new StringBuilder( value.length() );
796
boolean in_escape = false;
797
for( int a = 0; a < value.length(); a++ )
798
{
799
int c = value.codePointAt( a );
800
801
// process a normal character
802
if( !in_escape ) {
803
if( c == '\\' )
804
in_escape = true;
805
else
806
ret.append( Character.toChars( c ) );
807
continue;
808
}
809
810
// process an escape sequence
811
in_escape = false;
812
switch( c )
813
{
814
case 'T':
815
case 't':
816
// add tab (invalid/non-standard, but accepted)
817
ret.append( '\t' );
818
break;
819
case 'N':
820
case 'n':
821
// add newline
822
ret.append( '\n' );
823
break;
824
case '\\':
825
case ',':
826
case ';':
827
// add escaped character
828
ret.append( Character.toChars( c ) );
829
break;
830
default:
831
// unknown escape sequence, so add it unescaped
832
// (invalid/non-standard, but accepted)
833
ret.append( "\\" );
834
ret.append( Character.toChars( c ) );
835
break;
836
}
837
}
838
839
return ret.toString();
605
}
840
606
}
841
607
842
608
private void parseN( String[] params, String value )
609
throws ParseException, SkipContactException,
610
AbortImportException
843
611
{
844
612
// already got a better name?
845
613
if( _name_level >= NAMELEVEL_N ) return;
846
614
847
615
// get name parts
848
String[] name_parts = splitValueByCharacter( value, ';' );
616
String[] name_parts = value.split( ";" );
617
for( int i = 0; i < name_parts.length; i++ )
618
name_parts[ i ] = name_parts[ i ].trim();
849
619
850
620
// build name
851
621
value = "";
852
final int[] part_order = { 3, 1, 2, 0, 4 };
853
for( int a = 0; a < part_order.length; a++ )
854
if( name_parts.length > part_order[ a ] &&
855
name_parts[ part_order[ a ] ].length() > 0 )
856
{
857
// split this part in to it's comma-separated bits
858
String[] name_part_parts = splitValueByCharacter(
859
name_parts[ part_order[ a ] ], ',' );
860
for( int b = 0; b < name_part_parts.length; b++ )
861
if( name_part_parts[ b ].length() > 0 )
862
{
863
if( value.length() > 0 ) value += " ";
864
value += name_part_parts[ b ];
865
}
866
}
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 ];
867
626
868
627
// set name
869
setName( unescapeValue( value ) );
628
setName( value );
870
629
_name_level = NAMELEVEL_N;
630
631
// check now to see if we need to import this contact (to avoid
632
// parsing the rest of the vCard unnecessarily)
633
if( !isImportRequired( getName() ) )
634
throw new SkipContactException();
871
635
}
872
636
873
637
private void parseFN( String[] params, String value )
638
throws ParseException, SkipContactException
874
639
{
875
640
// already got a better name?
876
641
if( _name_level >= NAMELEVEL_FN ) return;
877
642
878
643
// set name
879
setName( unescapeValue( value ) );
644
setName( value );
880
645
_name_level = NAMELEVEL_FN;
881
646
}
882
647
883
648
private void parseORG( String[] params, String value )
649
throws ParseException, SkipContactException
884
650
{
651
// already got a better name?
652
if( _name_level >= NAMELEVEL_ORG ) return;
653
885
654
// get org parts
886
String[] org_parts = splitValueByCharacter( value, ';' );
887
if( org_parts == null || org_parts.length < 1 ) return;
888
889
// build organisation name
890
StringBuilder builder = new StringBuilder(
891
String.valueOf( org_parts[ 0 ] ) );
892
for( int a = 1; a < org_parts.length; a++ )
893
builder.append( ", " ).append( org_parts[ a ] );
894
String organisation = unescapeValue( builder.toString() );
895
896
// set organisation name (using a title we've previously found)
897
addOrganisation( organisation, _cached_title, true );
898
899
// if we've not previously found a title, store this organisation
900
// name (we'll need it when we find a title to update the
901
// organisation, by name), else if we *have* previously found a
902
// title, clear it (since we just used it)
903
if( _cached_title == null )
904
_cached_organisation = organisation;
905
else
906
_cached_title = null;
907
}
908
909
private void parseTITLE( String[] params, String value )
910
{
911
value = unescapeValue( value );
912
913
// if we previously had an organisation, look it up and append this
914
// title to it
915
if( _cached_organisation != null && hasOrganisations() ) {
916
HashMap< String, ExtraDetail > datas = getOrganisations();
917
ExtraDetail detail = datas.get( _cached_organisation );
918
if( detail != null )
919
detail.setExtra( value );
920
}
921
922
// same as when handling organisation, if we've not previously found
923
// an organisation we store this title, else we clear it (since we
924
// just appended this title to it)
925
if( _cached_organisation == null )
926
_cached_title = value;
927
else
928
_cached_organisation = null;
655
String[] org_parts = value.split( ";" );
656
for( int i = 0; i < org_parts.length; i++ )
657
org_parts[ i ] = org_parts[ i ].trim();
658
659
// build name
660
if( org_parts.length > 1 && org_parts[ 0 ].length() == 0 )
661
value = org_parts[ 1 ];
662
else
663
value = org_parts[ 0 ];
664
665
// set name
666
setName( value );
667
_name_level = NAMELEVEL_ORG;
929
668
}
930
669
931
670
private void parseTEL( String[] params, String value )
671
throws ParseException
932
672
{
933
673
if( value.length() == 0 ) return;
934
674
937
677
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
938
678
939
679
// here's the logic...
940
boolean is_preferred = types.contains( "PREF" );
941
int type;
680
boolean preferred = types.contains( "PREF" );
681
int type = PhonesColumns.TYPE_MOBILE;
682
if( types.contains( "VOICE" ) )
683
if( types.contains( "WORK" ) )
684
type = PhonesColumns.TYPE_WORK;
685
else
686
type = PhonesColumns.TYPE_HOME;
687
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
688
type = PhonesColumns.TYPE_MOBILE;
942
689
if( types.contains( "FAX" ) )
943
690
if( types.contains( "HOME" ) )
944
type = TYPE_FAX_HOME;
691
type = PhonesColumns.TYPE_FAX_HOME;
945
692
else
946
type = TYPE_FAX_WORK;
947
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
948
type = TYPE_MOBILE;
949
else if( types.contains( "PAGER" ) )
950
type = TYPE_PAGER;
951
else if( types.contains( "WORK" ) )
952
type = TYPE_WORK;
953
else
954
type = TYPE_HOME;
693
type = PhonesColumns.TYPE_FAX_WORK;
694
if( types.contains( "PAGER" ) )
695
type = PhonesColumns.TYPE_PAGER;
955
696
956
697
// add phone number
957
addNumber( value, type, is_preferred );
698
addPhone( value, type, preferred );
958
699
}
959
700
960
701
public void parseEMAIL( String[] params, String value )
702
throws ParseException
961
703
{
962
704
if( value.length() == 0 ) return;
963
705
964
706
Set< String > types = extractTypes( params, Arrays.asList(
965
707
"PREF", "WORK", "HOME", "INTERNET" ) );
966
708
967
// add email address
968
boolean is_preferred = types.contains( "PREF" );
969
int type;
970
if( types.contains( "WORK" ) )
971
type = TYPE_WORK;
972
else
973
type = TYPE_HOME;
974
975
addEmail( unescapeValue( value ), type, is_preferred );
976
}
977
978
private void parseADR( String[] params, String value )
979
{
980
// get address parts
981
String[] adr_parts = splitValueByCharacter( value, ';' );
982
983
// build address
984
value = "";
985
for( int a = 0; a < adr_parts.length; a++ )
986
if( adr_parts[ a ].length() > 0 )
987
{
988
// version 3.0 vCards allow further splitting by comma
989
if( _version.equals( "3.0" ) )
990
{
991
// split this part in to it's comma-separated bits and
992
// add them on individual lines
993
String[] adr_part_parts =
994
splitValueByCharacter( adr_parts[ a ], ',' );
995
for( int b = 0; b < adr_part_parts.length; b++ )
996
if( adr_part_parts[ b ].length() > 0 )
997
{
998
if( value.length() > 0 ) value += "\n";
999
value += adr_part_parts[ b ];
1000
}
1001
}
1002
else
1003
{
1004
// add this part on an individual line
1005
if( value.length() > 0 ) value += "\n";
1006
value += adr_parts[ a ];
1007
}
1008
}
1009
1010
Set< String > types = extractTypes( params, Arrays.asList(
1011
"PREF", "WORK", "HOME" ) );
1012
1013
// add address
1014
int type;
1015
if( types.contains( "WORK" ) )
1016
type = TYPE_WORK;
1017
else
1018
type = TYPE_HOME;
1019
1020
addAddress( unescapeValue( value ), type );
1021
}
1022
1023
private void parseLABEL( String[] params, String value )
1024
{
1025
Set< String > types = extractTypes( params, Arrays.asList(
1026
"PREF", "WORK", "HOME" ) );
1027
1028
// add address
1029
int type;
1030
if( types.contains( "WORK" ) )
1031
type = TYPE_WORK;
1032
else
1033
type = TYPE_HOME;
1034
1035
addAddress( unescapeValue( value ), type );
1036
}
1037
1038
private void parseNOTE( String[] params, String value )
1039
{
1040
addNote( unescapeValue( value ) );
1041
}
1042
1043
private void parseBDAY( String[] params, String value )
1044
{
1045
setBirthday( value );
1046
}
1047
1048
public void finaliseVcard()
1049
throws ParseException, ContactNotIdentifiableException
709
// here's the logic...
710
boolean preferred = types.contains( "PREF" );
711
if( types.contains( "WORK" ) )
712
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
713
else
714
addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
715
}
716
717
public void finaliseParsing()
718
throws ParseException, SkipContactException,
719
AbortImportException
1050
720
{
1051
721
// missing version (and data is present)
1052
if( _version == null && _content_lines != null )
722
if( _version == null && _buffers != null )
1053
723
throw new ParseException( R.string.error_vcf_malformed );
1054
724
1055
// finalise the parent class
1056
finalise();
725
// missing name properties?
726
if( _name_level == NAMELEVEL_NONE )
727
throw new ParseException( R.string.error_vcf_noname );
728
729
// check if we should import this one? If we've already got an 'N'-
730
// type name, this will already have been done by parseN() so we
731
// mustn't do this here (or it could prompt twice!)
732
if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
733
throw new SkipContactException();
1057
734
}
1058
735
1059
/**
1060
* Amongst the params, find the value of the first, only, of any with
1061
* the specified name.
1062
*
1063
* @param params
1064
* @param name
1065
* @return a value, or null
1066
*/
1067
736
private String checkParam( String[] params, String name )
1068
737
{
1069
String[] res = checkParams( params, name );
1070
return res.length > 0? res[ 0 ] : null;
1071
}
1072
1073
/**
1074
* Amongst the params, find the values of any with the specified name.
1075
*
1076
* @param params
1077
* @param name
1078
* @return an array of values, or null
1079
*/
1080
private String[] checkParams( String[] params, String name )
1081
{
1082
HashSet< String > ret = new HashSet< String >();
1083
1084
738
Pattern p = Pattern.compile(
1085
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1086
Pattern.CASE_INSENSITIVE );
739
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$" );
1087
740
for( int i = 0; i < params.length; i++ ) {
1088
741
Matcher m = p.matcher( params[ i ] );
1089
742
if( m.matches() )
1090
ret.add( m.group( 2 ) );
743
return m.group( 2 );
1091
744
}
1092
1093
return (String[]) ret.toArray( new String[ ret.size() ] );
745
return null;
1094
746
}
1095
747
1096
/**
1097
* Amongst the params, return any type values present. For v2.1 vCards,
1098
* those types are just parameters. For v3.0, they are prefixed with
1099
* "TYPE=". There may also be multiple type parameters.
1100
*
1101
* @param params an array of params to look for types in
1102
* @param valid_types an list of upper-case type values to look for
1103
* @return a set of present type values
1104
*/
1105
748
private Set< String > extractTypes( String[] params,
1106
749
List< String > valid_types )
1107
750
{
1108
751
HashSet< String > types = new HashSet< String >();
1109
752
1110
753
// get 3.0-style TYPE= param
1111
String type_params[] = checkParams( params, "TYPE" );
1112
for( int a = 0; a < type_params.length; a++ )
1113
{
1114
// check for a comma-separated list of types (why? I don't think
1115
// this is in the specs!)
1116
String[] parts = type_params[ a ].split( "," );
1117
for( int i = 0; i < parts.length; i++ ) {
1118
String ucpart = parts[ i ].toUpperCase( Locale.US );
1119
if( valid_types.contains( ucpart ) )
1120
types.add( ucpart );
1121
}
754
String type_param;
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 ] );
1122
760
}
1123
761
1124
762
// get 2.1-style type param
1125
763
if( _version.equals( "2.1" ) ) {
1126
for( int i = 1; i < params.length; i++ ) {
1127
String ucparam = params[ i ].toUpperCase( Locale.US );
1128
if( valid_types.contains( ucparam ) )
1129
types.add( ucparam );
1130
}
764
for( int i = 1; i < params.length; i++ )
765
if( valid_types.contains( params[ i ] ) )
766
types.add( params[ i ] );
1131
767
}
1132
768
1133
769
return types;
1155
791
else if( ch == '=' && i == in.limit() - 1 )
1156
792
{
1157
793
// we found a '=' at the end of a line signifying a multi-
1158
// line string, so we don't add it
794
// line string, so we don't add it.
1159
795
another = true;
1160
796
continue;
1161
797
}
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480