(root)/android/import-contacts
» Revision
25
(compared to revision 116)
: 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 25
- compare with another revision
- download diff
- download tarball
- view history from revision 25
- Committer: edam
- Date: 2010-12-14 13:41:21 UTC
- Revision ID: edam@waxworlds.org-20101214134121-hz3v1pq2umlkompk
- fixed bug where parts[0] was assumed to exists after calling split()
- renamed "bits" to "parts" and check all usage of split()
- renamed "bits" to "parts" and check all usage of split()
- 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
import java.util.Iterator;
40
37
import java.util.List;
41
import java.util.Locale;
42
import java.util.NoSuchElementException;
43
38
import java.util.Set;
44
39
import java.util.Vector;
45
40
import java.util.regex.Matcher;
46
41
import java.util.regex.Pattern;
47
42
48
43
import android.content.SharedPreferences;
49
import android.os.Environment;
44
import android.provider.Contacts;
45
import android.provider.Contacts.PhonesColumns;
50
46
51
public class VcardImporter extends Importer
47
public class VCFImporter extends Importer
52
48
{
53
private int _vcard_count = 0;
49
private int _vCardCount = 0;
54
50
private int _progress = 0;
55
51
56
public VcardImporter( Doit doit )
52
public VCFImporter( Doit doit )
57
53
{
58
54
super( doit );
59
55
}
70
66
File[] files = null;
71
67
try
72
68
{
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
69
// open directory
82
File file = new File( Environment.getExternalStorageDirectory(),
83
prefs.getString( "location", "/" ) );
70
String path = "/sdcard" + prefs.getString( "location", "/" );
71
File file = new File( path );
84
72
if( !file.exists() )
85
73
showError( R.string.error_locationnotfound );
86
74
90
78
// get files
91
79
class VCardFilter implements FilenameFilter {
92
80
public boolean accept( File dir, String name ) {
93
return name.toLowerCase( Locale.ENGLISH )
94
.endsWith( ".vcf" );
81
return name.toLowerCase().endsWith( ".vcf" );
95
82
}
96
83
}
97
84
files = file.listFiles( new VCardFilter() );
119
106
countVCardFile( files[ i ] );
120
107
setTmpProgress( i );
121
108
}
122
setProgressMax( _vcard_count ); // will also update tmp progress
109
setProgressMax( _vCardCount ); // will also update tmp progress
123
110
124
111
// import them
125
112
setProgress( 0 );
126
113
for( int i = 0; i < files.length; i++ )
127
114
importVCardFile( files[ i ] );
128
setProgress( _vcard_count );
129
115
}
130
116
131
117
private void countVCardFile( File file ) throws AbortImportException
134
120
{
135
121
// open file
136
122
BufferedReader reader = new BufferedReader(
137
new FileReader( file ) );
123
new FileReader( file ) );
138
124
139
125
// read
140
126
String line;
141
boolean in_vcard = false;
127
boolean inVCard = false;
142
128
while( ( line = reader.readLine() ) != null )
143
129
{
144
if( !in_vcard )
145
{
130
if( !inVCard ) {
146
131
// look for vcard beginning
147
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
148
in_vcard = true;
149
_vcard_count++;
150
}
151
// check for vMsg files
152
else if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VMSG.*" ) ) {
153
showError( getText( R.string.error_vcf_vmsgfile )
154
+ file.getName() );
132
if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
133
inVCard = true;
134
_vCardCount++;
155
135
}
156
136
}
157
else if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
158
in_vcard = false;
137
else if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
138
inVCard = false;
159
139
}
160
reader.close();
161
140
162
141
}
163
142
catch( FileNotFoundException e ) {
185
164
FileInputStream istream = new FileInputStream( file );
186
165
byte[] content = new byte[ (int)file.length() ];
187
166
istream.read( content );
188
istream.close();
189
167
190
168
// import
191
169
importVCardFileContent( content, file.getName() );
192
170
}
193
catch( OutOfMemoryError e ) {
194
showError( R.string.error_outofmemory );
195
}
196
171
catch( FileNotFoundException e ) {
197
172
showError( getText( R.string.error_filenotfound ) +
198
173
file.getName() );
203
178
}
204
179
205
180
private void importVCardFileContent( byte[] content, String fileName )
206
throws AbortImportException
181
throws AbortImportException
207
182
{
183
ByteBuffer buffers[] = getLinesFromContent( content );
184
208
185
// go through lines
209
Vcard vcard = null;
210
int vcard_start_line = 0;
211
ContentLineIterator cli = new ContentLineIterator( content );
212
while( cli.hasNext() )
186
VCard vCard = null;
187
for( int i = 0; i < buffers.length; i++ )
213
188
{
214
ContentLine content_line = cli.next();
215
216
// get a US-ASCII version of the string, for processing
217
String line = content_line.getUsAsciiLine();
218
219
if( vcard == null ) {
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
}
199
200
if( vCard == null ) {
220
201
// look for vcard beginning
221
if( line.matches( "(?i)BEGIN[ \t]*:[ \t]*VCARD.*" ) ) {
222
setProgress( _progress++ );
223
vcard = new Vcard();
224
vcard_start_line = cli.getLineNumber();
202
if( line.matches( "^BEGIN[ \\t]*:[ \\t]*VCARD" ) ) {
203
setProgress( ++_progress );
204
vCard = new VCard();
225
205
}
226
206
}
227
207
else {
228
208
// look for vcard content or ending
229
if( line.matches( "(?i)END[ \t]*:[ \t]*VCARD.*" ) )
209
if( line.matches( "^END[ \\t]*:[ \\t]*VCARD" ) )
230
210
{
231
// finalise the vcard/contact
211
// store vcard and do away with it
232
212
try {
233
vcard.finaliseVcard();
234
235
// pass the finalised contact to the importer
236
importContact( vcard );
237
}
238
catch( Vcard.ParseException e ) {
239
showContinueOrAbort(
240
getText( R.string.error_vcf_parse ).toString()
241
+ fileName +
242
getText( R.string.error_vcf_parse_line ).toString()
243
+ cli.getLineNumber() + ":\n" + e.getMessage() );
244
skipContact();
245
}
246
catch( ContactData.ContactNotIdentifiableException e ) {
247
showContinueOrAbort(
248
getText( R.string.error_vcf_parse ).toString()
249
+ fileName +
250
getText( R.string.error_vcf_parse_line ).toString()
251
+ vcard_start_line + ":\n" + getText(
252
R.string.error_vcf_notenoughinfo ).toString() );
253
skipContact();
254
}
255
256
// discard this vcard
257
vcard = null;
213
vCard.finaliseParsing();
214
importContact( vCard );
215
}
216
catch( VCard.ParseException e ) {
217
skipContact();
218
if( !showContinue(
219
getText( R.string.error_vcf_parse ).toString()
220
+ fileName + "\n" + e.getMessage() ) )
221
finish( ACTION_ABORT );
222
}
223
catch( VCard.SkipContactException e ) {
224
skipContact();
225
// do nothing
226
}
227
vCard = null;
258
228
}
259
229
else
260
230
{
261
231
// try giving the line to the vcard
262
232
try {
263
vcard.parseLine( content_line );
233
vCard.parseLine( buffers[ i ] );
264
234
}
265
catch( Vcard.ParseException e ) {
235
catch( VCard.ParseException e ) {
266
236
skipContact();
267
showContinueOrAbort(
268
getText( R.string.error_vcf_parse ).toString()
269
+ fileName +
270
getText( R.string.error_vcf_parse_line ).toString()
271
+ cli.getLineNumber() + "\n" + e.getMessage() );
237
if( !showContinue(
238
getText( R.string.error_vcf_parse ).toString()
239
+ fileName + "\n" + e.getMessage() ) )
240
finish( ACTION_ABORT );
272
241
273
// Although we're continuing, we still need to abort
274
// this vCard. Further lines will be ignored until we
242
// although we're continuing, we still need to abort
243
// this vCard. Further lines will be ignored until we
275
244
// get to another BEGIN:VCARD line.
276
vcard = null;
245
vCard = null;
277
246
}
278
catch( Vcard.SkipImportException e ) {
247
catch( VCard.SkipContactException e ) {
279
248
skipContact();
280
// Abort this vCard. Further lines will be ignored until
249
// abort this vCard. Further lines will be ignored until
281
250
// we get to another BEGIN:VCARD line.
282
vcard = null;
251
vCard = null;
283
252
}
284
253
}
285
254
}
286
255
}
287
256
}
288
257
289
class ContentLine
290
{
291
private ByteBuffer _buffer;
292
private boolean _folded_next;
293
private String _line;
294
295
public ContentLine( ByteBuffer buffer, boolean folded_next )
296
{
297
_buffer = buffer;
298
_folded_next = folded_next;
299
_line = null;
300
}
301
302
public ByteBuffer getBuffer()
303
{
304
return _buffer;
305
}
306
307
public boolean doesNextLineLookFolded()
308
{
309
return _folded_next;
310
}
311
312
public String getUsAsciiLine()
313
{
314
// generated line and cache it
315
if( _line == null ) {
316
try {
317
_line = new String( _buffer.array(), _buffer.position(),
318
_buffer.limit() - _buffer.position(), "US-ASCII" );
319
}
320
catch( UnsupportedEncodingException e ) {
321
// we know US-ASCII *is* supported, so appease the
322
// compiler...
323
}
324
}
325
326
// return cached line
327
return _line;
328
}
329
}
330
331
class ContentLineIterator implements Iterator< ContentLine >
332
{
333
protected byte[] _content = null;
334
protected int _pos = 0;
335
protected int _line = 0;
336
337
public ContentLineIterator( byte[] content )
338
{
339
_content = content;
340
}
341
342
@Override
343
public boolean hasNext()
344
{
345
return _pos < _content.length;
346
}
347
348
@Override
349
public ContentLine next()
350
{
351
int initial_pos = _pos;
352
353
// find newline
354
for( ; _pos < _content.length; _pos++ )
355
if( _content[ _pos ] == '\n' )
356
{
357
// adjust for a \r preceding the \n
358
int to = ( _pos > 0 && _content[ _pos - 1 ] == '\r' &&
359
_pos > initial_pos )? _pos - 1 : _pos;
360
_pos++;
361
_line++;
362
return new ContentLine(
363
ByteBuffer.wrap( _content, initial_pos,
364
to - initial_pos ),
365
doesNextLineLookFolded() );
366
}
367
368
// we didn't find one, but were there bytes left?
369
if( _pos != initial_pos ) {
370
int to = _pos;
371
_pos++;
372
_line++;
373
return new ContentLine(
374
ByteBuffer.wrap( _content, initial_pos,
375
to - initial_pos ),
376
doesNextLineLookFolded() );
377
}
378
379
// no bytes left
380
throw new NoSuchElementException();
381
}
382
383
@Override
384
public void remove()
385
{
386
throw new UnsupportedOperationException();
387
}
388
389
/**
390
* Does the next line, if there is one, look like it should be folded
391
* onto the end of this one?
392
* @return
393
*/
394
private boolean doesNextLineLookFolded()
395
{
396
return _pos > 0 && _pos < _content.length &&
397
_content[ _pos - 1 ] == '\n' &&
398
( _content[ _pos ] == ' ' || _content[ _pos ] == '\t' );
399
}
400
401
public int getLineNumber()
402
{
403
return _line;
404
}
405
}
406
407
private class Vcard extends ContactData
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
282
private class VCard extends ContactData
408
283
{
409
284
private final static int NAMELEVEL_NONE = 0;
410
private final static int NAMELEVEL_N = 1;
285
private final static int NAMELEVEL_ORG = 1;
411
286
private final static int NAMELEVEL_FN = 2;
412
413
private final static int MULTILINE_NONE = 0;
414
private final static int MULTILINE_ENCODED = 1; // v2.1 quoted-printable
415
private final static int MULTILINE_ESCAPED = 2; // v2.1 \\CRLF
416
private final static int MULTILINE_FOLDED = 3; // MIME-DIR folding
287
private final static int NAMELEVEL_N = 3;
417
288
418
289
private String _version = null;
419
private Vector< ContentLine > _content_lines = null;
290
private Vector< ByteBuffer > _buffers = null;
420
291
private int _name_level = NAMELEVEL_NONE;
421
private int _parser_multiline_state = MULTILINE_NONE;
292
private boolean _parser_in_multiline = false;
422
293
private String _parser_current_name_and_params = null;
423
294
private String _parser_buffered_value_so_far = "";
424
private String _cached_organisation = null;
425
private String _cached_title = null;
426
295
427
296
protected class UnencodeResult
428
297
{
458
327
459
328
public ParseException( int res )
460
329
{
461
super( VcardImporter.this.getText( res ).toString() );
330
super( VCFImporter.this.getText( res ).toString() );
462
331
}
463
332
}
464
333
465
334
@SuppressWarnings("serial")
466
protected class SkipImportException extends Exception { }
467
468
private String extractCollonPartFromLine( ContentLine content_line,
469
boolean former )
470
{
471
// split line into name and value parts and check to make sure we
472
// only got 2 parts and that the first part is not zero in length
473
String[] parts = content_line.getUsAsciiLine().split( ":", 2 );
474
if( parts.length == 2 && parts[ 0 ].length() > 0 )
475
return parts[ former? 0 : 1 ].trim();
476
477
return null;
478
}
479
480
private String extractNameAndParamsFromLine( ContentLine content_line )
481
{
482
return extractCollonPartFromLine( content_line, true );
483
}
484
485
private String extractValueFromLine( ContentLine content_line )
486
{
487
return extractCollonPartFromLine( content_line, false );
488
}
489
490
public void parseLine( ContentLine content_line )
491
throws ParseException, SkipImportException,
492
AbortImportException
493
{
494
// do we have a version yet?
335
protected class SkipContactException extends Exception { }
336
337
public void parseLine( ByteBuffer buffer )
338
throws ParseException, SkipContactException,
339
AbortImportException
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
352
// ignore empty lines
353
if( line.trim() == "" ) return;
354
355
// split line into name and value parts (this may turn out to be
356
// unwanted if the line is a subsequent line in a multi-line
357
// value, but we have to do this now to check for and handle VCF
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[] parts = line.split( ":", 2 );
366
if( parts.length == 2 ) {
367
name_and_params = parts[ 0 ].trim();
368
string_value = parts[ 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
}
380
381
// if we haven't yet got a version, we won't be paring anything!
495
382
if( _version == null )
496
383
{
497
// tentatively get name and params from line
498
String name_and_params =
499
extractNameAndParamsFromLine( content_line );
500
501
// is it a version line?
502
if( name_and_params != null &&
503
name_and_params.equalsIgnoreCase( "VERSION" ) )
384
// is this a version?
385
if( name_and_params.equals( "VERSION" ) )
504
386
{
505
// yes, get it!
506
String value = extractValueFromLine( content_line );
507
if( value == null || (
508
!value.equals( "2.1" ) && !value.equals( "3.0" ) ) )
509
{
387
// yes, check/store it
388
if( !string_value.equals( "2.1" ) &&
389
!string_value.equals( "3.0" ) )
510
390
throw new ParseException( R.string.error_vcf_version );
511
}
512
_version = value;
391
_version = string_value;
513
392
514
// parse any buffers we've been accumulating while we waited
515
// for a version
516
if( _content_lines != null )
517
for( int i = 0; i < _content_lines.size(); i++ )
518
parseLine( _content_lines.get( i ) );
519
_content_lines = null;
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;
520
398
}
521
399
else
522
400
{
523
// no, so stash this line till we get a version
524
if( _content_lines == null )
525
_content_lines = new Vector< ContentLine >();
526
_content_lines.add( content_line );
401
// no, so stash this buffer till we have a version
402
if( _buffers == null )
403
_buffers = new Vector< ByteBuffer >();
404
_buffers.add( buffer );
527
405
}
528
406
}
529
407
else
530
408
{
531
// name and params and the position in the buffer where the
532
// "value" part of the line starts
533
String name_and_params;
534
int pos;
409
// value bytes, for processing
410
ByteBuffer value;
535
411
536
if( _parser_multiline_state != MULTILINE_NONE )
412
if( _parser_in_multiline )
537
413
{
538
414
// if we're currently in a multi-line value, use the stored
539
415
// property name and parameters
540
416
name_and_params = _parser_current_name_and_params;
541
417
542
// skip some initial line characters, depending on the type
543
// of multi-line we're handling
544
pos = content_line.getBuffer().position();
545
switch( _parser_multiline_state )
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' ) )
546
424
{
547
case MULTILINE_FOLDED:
548
425
pos++;
549
break;
550
case MULTILINE_ENCODED:
551
while( pos < content_line.getBuffer().limit() && (
552
content_line.getBuffer().get( pos ) == ' ' ||
553
content_line.getBuffer().get( pos ) == '\t' ) )
554
{
555
pos++;
556
}
557
break;
558
default:
559
// do nothing
560
426
}
561
427
562
// take us out of multi-line so that we can re-detect that
563
// this line is a multi-line or not
564
_parser_multiline_state = MULTILINE_NONE;
428
// get value from buffer
429
value = ByteBuffer.wrap( buffer.array(), pos,
430
buffer.limit() - pos );
565
431
}
566
432
else
567
433
{
568
// skip empty lines
569
if( content_line.getUsAsciiLine().trim().length() == 0 )
570
return;
571
572
// get name and params from line, and since we're not
573
// parsing a subsequent line in a multi-line, this should
574
// not fail, or it's an error
575
name_and_params =
576
extractNameAndParamsFromLine( content_line );
577
if( name_and_params == null )
578
throw new ParseException(
579
R.string.error_vcf_malformed );
434
// ignore empty values
435
if( string_value.length() < 1 ) return;
580
436
581
437
// calculate how many chars to skip from beginning of line
582
438
// so we skip the property "name:" part
583
pos = content_line.getBuffer().position() +
584
name_and_params.length() + 1;
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 );
585
444
586
445
// reset the saved multi-line state
587
446
_parser_current_name_and_params = name_and_params;
588
447
_parser_buffered_value_so_far = "";
589
448
}
590
449
591
// get value from buffer, as raw bytes
592
ByteBuffer value;
593
value = ByteBuffer.wrap( content_line.getBuffer().array(), pos,
594
content_line.getBuffer().limit() - pos );
595
596
450
// get parameter parts
597
451
String[] name_param_parts = name_and_params.split( ";", -1 );
598
452
for( int i = 0; i < name_param_parts.length; i++ )
599
453
name_param_parts[ i ] = name_param_parts[ i ].trim();
600
454
601
// determine whether we care about this entry
602
final HashSet< String > interesting_fields =
603
new HashSet< String >( Arrays.asList( new String[] { "N",
604
"FN", "ORG", "TITLE", "TEL", "EMAIL", "ADR", "LABEL" }
605
) );
606
boolean is_interesting_field =
607
interesting_fields.contains(
608
name_param_parts[ 0 ].toUpperCase( Locale.ENGLISH ) );
609
610
455
// parse encoding parameter
611
456
String encoding = checkParam( name_param_parts, "ENCODING" );
612
if( encoding != null )
613
encoding = encoding.toUpperCase( Locale.ENGLISH );
614
if( is_interesting_field && encoding != null &&
615
!encoding.equalsIgnoreCase( "8BIT" ) &&
616
!encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
617
//&& !encoding.equalsIgnoreCase( "BASE64" ) )
457
if( encoding != null ) encoding = encoding.toUpperCase();
458
if( encoding != null && !encoding.equals( "8BIT" ) &&
459
!encoding.equals( "QUOTED-PRINTABLE" ) )
460
//&& !encoding.equals( "BASE64" ) )
618
461
{
619
462
throw new ParseException( R.string.error_vcf_encoding );
620
463
}
621
464
622
465
// parse charset parameter
623
466
String charset = checkParam( name_param_parts, "CHARSET" );
624
if( charset != null )
625
charset = charset.toUpperCase( Locale.ENGLISH );
626
if( charset != null &&
627
!charset.equalsIgnoreCase( "US-ASCII" ) &&
628
!charset.equalsIgnoreCase( "ASCII" ) &&
629
!charset.equalsIgnoreCase( "UTF-8" ) )
467
if( charset != null ) charset = charset.toUpperCase();
468
if( charset != null && !charset.equals( "US-ASCII" ) &&
469
!charset.equals( "ASCII" ) && !charset.equals( "UTF-8" ) )
630
470
{
631
471
throw new ParseException( R.string.error_vcf_charset );
632
472
}
634
474
// do unencoding (or default to a fake unencoding result with
635
475
// the raw string)
636
476
UnencodeResult unencoding_result = null;
637
if( encoding != null &&
638
encoding.equalsIgnoreCase( "QUOTED-PRINTABLE" ) )
639
{
477
if( encoding != null && encoding.equals( "QUOTED-PRINTABLE" ) )
640
478
unencoding_result = unencodeQuotedPrintable( value );
641
}
642
// else if( encoding != null &&
643
// encoding.equalsIgnoreCase( "BASE64" ) )
644
// {
645
// unencoding_result = unencodeBase64( props[ 1 ], charset );
646
// }
479
// else if( encoding != null && encoding.equals( "BASE64" ) )
480
// result = unencodeBase64( props[ 1 ], charset );
647
481
if( unencoding_result != null ) {
648
482
value = unencoding_result.getBuffer();
649
if( unencoding_result.isAnotherLineRequired() )
650
_parser_multiline_state = MULTILINE_ENCODED;
483
_parser_in_multiline =
484
unencoding_result.isAnotherLineRequired();
651
485
}
652
486
653
// convert 8-bit US-ASCII charset to UTF-8 (where no charset is
654
// specified for a v2.1 vcard entry, we assume it's US-ASCII)
655
if( ( charset == null && _version.equals( "2.1" ) ) ||
656
( charset != null && (
657
charset.equalsIgnoreCase( "ASCII" ) ||
658
charset.equalsIgnoreCase( "US-ASCII" ) ) ) )
659
{
487
// convert 8-bit ASCII charset to US-ASCII
488
if( charset == null || charset == "ASCII" ) {
660
489
value = transcodeAsciiToUtf8( value );
490
charset = "UTF-8";
661
491
}
662
492
663
// process charset (value is now in UTF-8)
664
String string_value;
493
// process charset
665
494
try {
666
string_value = new String( value.array(), value.position(),
667
value.limit() - value.position(), "UTF-8" );
495
string_value =
496
new String( value.array(), value.position(),
497
value.limit() - value.position(), charset );
668
498
} catch( UnsupportedEncodingException e ) {
669
499
throw new ParseException( R.string.error_vcf_charset );
670
500
}
671
501
672
// for some entries that have semicolon-separated value parts,
673
// check to see if the value ends in an escape character, which
674
// indicates that we have a multi-line value
675
if( ( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) ||
676
name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) ||
677
name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) ) &&
678
doesStringEndInAnEscapeChar( string_value ) )
679
{
680
_parser_multiline_state = MULTILINE_ESCAPED;
681
string_value = string_value.substring( 0,
682
string_value.length() - 1 );
683
}
684
685
// if we know we're not in an encoding-based multi-line, check
686
// to see if we're in a folded multi-line
687
if( _parser_multiline_state == MULTILINE_NONE &&
688
content_line.doesNextLineLookFolded() )
689
{
690
_parser_multiline_state = MULTILINE_FOLDED;
691
}
692
693
// handle multi-lines by buffering them and parsing them when we
694
// are processing the last line in a multi-line sequence
695
if( _parser_multiline_state != MULTILINE_NONE ) {
502
// handle multi-line requests
503
if( _parser_in_multiline ) {
696
504
_parser_buffered_value_so_far += string_value;
697
505
return;
698
506
}
507
508
// add on buffered multi-line content
699
509
String complete_value =
700
( _parser_buffered_value_so_far + string_value ).trim();
701
702
// ignore empty values
703
if( complete_value.length() < 1 ) return;
510
_parser_buffered_value_so_far + string_value;
704
511
705
512
// parse some properties
706
if( name_param_parts[ 0 ].equalsIgnoreCase( "N" ) )
513
if( name_param_parts[ 0 ].equals( "N" ) )
707
514
parseN( name_param_parts, complete_value );
708
else if( name_param_parts[ 0 ].equalsIgnoreCase( "FN" ) )
515
else if( name_param_parts[ 0 ].equals( "FN" ) )
709
516
parseFN( name_param_parts, complete_value );
710
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ORG" ) )
517
else if( name_param_parts[ 0 ].equals( "ORG" ) )
711
518
parseORG( name_param_parts, complete_value );
712
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TITLE" ) )
713
parseTITLE( name_param_parts, complete_value );
714
else if( name_param_parts[ 0 ].equalsIgnoreCase( "TEL" ) )
519
else if( name_param_parts[ 0 ].equals( "TEL" ) )
715
520
parseTEL( name_param_parts, complete_value );
716
else if( name_param_parts[ 0 ].equalsIgnoreCase( "EMAIL" ) )
521
else if( name_param_parts[ 0 ].equals( "EMAIL" ) )
717
522
parseEMAIL( name_param_parts, complete_value );
718
else if( name_param_parts[ 0 ].equalsIgnoreCase( "ADR" ) )
719
parseADR( name_param_parts, complete_value );
720
else if( name_param_parts[ 0 ].equalsIgnoreCase( "LABEL" ) )
721
parseLABEL( name_param_parts, complete_value );
722
else if( name_param_parts[ 0 ].equalsIgnoreCase( "NOTE" ) )
723
parseNOTE( name_param_parts, complete_value );
724
else if( name_param_parts[ 0 ].equalsIgnoreCase( "BDAY" ) )
725
parseBDAY( name_param_parts, complete_value );
726
}
727
}
728
729
private boolean doesStringEndInAnEscapeChar( String string )
730
{
731
// count the number of backslashes at the end of the string
732
int count = 0;
733
for( int a = string.length() - 1; a >= 0; a-- )
734
if( string.charAt( a ) == '\\' )
735
count++;
736
else
737
break;
738
739
// if there are an even number of backslashes then the final one
740
// doesn't count
741
return ( count & 1 ) == 1;
742
}
743
744
private String[] splitValueByCharacter( String value, char character )
745
{
746
// split string in to parts by specified character
747
ArrayList< String > parts = new ArrayList< String >(
748
Arrays.asList( value.split( "" + character ) ) );
749
750
// go through parts
751
for( int a = 0; a < parts.size(); a++ )
752
{
753
String str = parts.get( a );
754
755
// Look for parts that end in an escape character, but ignore
756
// the final part. We've already detected escape chars at the
757
// end of the final part in parseLine() and handled multi-lines
758
// accordingly.
759
if( a < parts.size() - 1 &&
760
doesStringEndInAnEscapeChar( str ) )
761
{
762
// append the escaped character, join the next part to this
763
// part and remove the next part
764
parts.set( a, str.substring( 0, str.length() - 1 ) +
765
character + parts.get( a + 1 ) );
766
parts.remove( a + 1 );
767
768
// re-visit this part
769
a--;
770
continue;
771
}
772
773
// trim and replace string
774
str = str.trim();
775
parts.set( a, str );
776
}
777
778
String[] ret = new String[ parts.size() ];
779
return parts.toArray( ret );
780
}
781
782
private String unescapeValue( String value )
783
{
784
StringBuilder ret = new StringBuilder( value.length() );
785
boolean in_escape = false;
786
for( int a = 0; a < value.length(); a++ )
787
{
788
int c = value.codePointAt( a );
789
790
// process a normal character
791
if( !in_escape ) {
792
if( c == '\\' )
793
in_escape = true;
794
else
795
ret.append( Character.toChars( c ) );
796
continue;
797
}
798
799
// process an escape sequence
800
in_escape = false;
801
switch( c )
802
{
803
case 'T':
804
case 't':
805
// add tab (invalid/non-standard, but accepted)
806
ret.append( '\t' );
807
break;
808
case 'N':
809
case 'n':
810
// add newline
811
ret.append( '\n' );
812
break;
813
case '\\':
814
case ',':
815
case ';':
816
// add escaped character
817
ret.append( Character.toChars( c ) );
818
break;
819
default:
820
// unknown escape sequence, so add it unescaped
821
// (invalid/non-standard, but accepted)
822
ret.append( "\\" );
823
ret.append( Character.toChars( c ) );
824
break;
825
}
826
}
827
828
return ret.toString();
523
}
829
524
}
830
525
831
526
private void parseN( String[] params, String value )
527
throws ParseException, SkipContactException,
528
AbortImportException
832
529
{
833
530
// already got a better name?
834
531
if( _name_level >= NAMELEVEL_N ) return;
835
532
836
533
// get name parts
837
String[] name_parts = splitValueByCharacter( value, ';' );
534
String[] name_parts = value.split( ";" );
535
for( int i = 0; i < name_parts.length; i++ )
536
name_parts[ i ] = name_parts[ i ].trim();
838
537
839
538
// build name
840
539
value = "";
841
final int[] part_order = { 3, 1, 2, 0, 4 };
842
for( int a = 0; a < part_order.length; a++ )
843
if( name_parts.length > part_order[ a ] &&
844
name_parts[ part_order[ a ] ].length() > 0 )
845
{
846
// split this part in to it's comma-separated bits
847
String[] name_part_parts = splitValueByCharacter(
848
name_parts[ part_order[ a ] ], ',' );
849
for( int b = 0; b < name_part_parts.length; b++ )
850
if( name_part_parts[ b ].length() > 0 )
851
{
852
if( value.length() > 0 ) value += " ";
853
value += name_part_parts[ b ];
854
}
855
}
540
if( name_parts.length > 1 && name_parts[ 1 ].length() > 0 )
541
value += name_parts[ 1 ];
542
if( name_parts.length > 0 && name_parts[ 0 ].length() > 0 )
543
value += ( value.length() == 0? "" : " " ) + name_parts[ 0 ];
856
544
857
545
// set name
858
setName( unescapeValue( value ) );
546
setName( value );
859
547
_name_level = NAMELEVEL_N;
548
549
// check now to see if we need to import this contact (to avoid
550
// parsing the rest of the vCard unnecessarily)
551
if( !isImportRequired( getName() ) )
552
throw new SkipContactException();
860
553
}
861
554
862
555
private void parseFN( String[] params, String value )
556
throws ParseException, SkipContactException
863
557
{
864
558
// already got a better name?
865
559
if( _name_level >= NAMELEVEL_FN ) return;
866
560
867
561
// set name
868
setName( unescapeValue( value ) );
562
setName( value );
869
563
_name_level = NAMELEVEL_FN;
870
564
}
871
565
872
566
private void parseORG( String[] params, String value )
567
throws ParseException, SkipContactException
873
568
{
569
// already got a better name?
570
if( _name_level >= NAMELEVEL_ORG ) return;
571
874
572
// get org parts
875
String[] org_parts = splitValueByCharacter( value, ';' );
876
if( org_parts == null || org_parts.length < 1 ) return;
877
878
// build organisation name
879
StringBuilder builder = new StringBuilder(
880
String.valueOf( org_parts[ 0 ] ) );
881
for( int a = 1; a < org_parts.length; a++ )
882
builder.append( ", " ).append( org_parts[ a ] );
883
String organisation = unescapeValue( builder.toString() );
884
885
// set organisation name (using a title we've previously found)
886
addOrganisation( organisation, _cached_title, true );
887
888
// if we've not previously found a title, store this organisation
889
// name (we'll need it when we find a title to update the
890
// organisation, by name), else if we *have* previously found a
891
// title, clear it (since we just used it)
892
if( _cached_title == null )
893
_cached_organisation = organisation;
894
else
895
_cached_title = null;
896
}
897
898
private void parseTITLE( String[] params, String value )
899
{
900
value = unescapeValue( value );
901
902
// if we previously had an organisation, look it up and append this
903
// title to it
904
if( _cached_organisation != null && hasOrganisations() ) {
905
HashMap< String, ExtraDetail > datas = getOrganisations();
906
ExtraDetail detail = datas.get( _cached_organisation );
907
if( detail != null )
908
detail.setExtra( value );
909
}
910
911
// same as when handling organisation, if we've not previously found
912
// an organisation we store this title, else we clear it (since we
913
// just appended this title to it)
914
if( _cached_organisation == null )
915
_cached_title = value;
916
else
917
_cached_organisation = null;
573
String[] org_parts = value.split( ";" );
574
for( int i = 0; i < org_parts.length; i++ )
575
org_parts[ i ] = org_parts[ i ].trim();
576
577
// build name
578
if( org_parts.length > 1 && org_parts[ 0 ].length() == 0 )
579
value = org_parts[ 1 ];
580
else
581
value = org_parts[ 0 ];
582
583
// set name
584
setName( value );
585
_name_level = NAMELEVEL_ORG;
918
586
}
919
587
920
588
private void parseTEL( String[] params, String value )
589
throws ParseException
921
590
{
922
591
if( value.length() == 0 ) return;
923
592
924
593
Set< String > types = extractTypes( params, Arrays.asList(
925
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
926
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
594
"PREF", "HOME", "WORK", "VOICE", "FAX", "MSG", "CELL",
595
"PAGER", "BBS", "MODEM", "CAR", "ISDN", "VIDEO" ) );
927
596
928
597
// here's the logic...
929
boolean is_preferred = types.contains( "PREF" );
930
int type;
598
boolean preferred = types.contains( "PREF" );
599
int type = PhonesColumns.TYPE_MOBILE;
600
if( types.contains( "VOICE" ) )
601
if( types.contains( "WORK" ) )
602
type = PhonesColumns.TYPE_WORK;
603
else
604
type = PhonesColumns.TYPE_HOME;
605
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
606
type = PhonesColumns.TYPE_MOBILE;
931
607
if( types.contains( "FAX" ) )
932
608
if( types.contains( "HOME" ) )
933
type = TYPE_FAX_HOME;
609
type = PhonesColumns.TYPE_FAX_HOME;
934
610
else
935
type = TYPE_FAX_WORK;
936
else if( types.contains( "CELL" ) || types.contains( "VIDEO" ) )
937
type = TYPE_MOBILE;
938
else if( types.contains( "PAGER" ) )
939
type = TYPE_PAGER;
940
else if( types.contains( "WORK" ) )
941
type = TYPE_WORK;
942
else
943
type = TYPE_HOME;
611
type = PhonesColumns.TYPE_FAX_WORK;
612
if( types.contains( "PAGER" ) )
613
type = PhonesColumns.TYPE_PAGER;
944
614
945
615
// add phone number
946
addNumber( value, type, is_preferred );
616
addPhone( value, type, preferred );
947
617
}
948
618
949
619
public void parseEMAIL( String[] params, String value )
620
throws ParseException
950
621
{
951
622
if( value.length() == 0 ) return;
952
623
953
624
Set< String > types = extractTypes( params, Arrays.asList(
954
"PREF", "WORK", "HOME", "INTERNET" ) );
955
956
// add email address
957
boolean is_preferred = types.contains( "PREF" );
958
int type;
959
if( types.contains( "WORK" ) )
960
type = TYPE_WORK;
961
else
962
type = TYPE_HOME;
963
964
addEmail( unescapeValue( value ), type, is_preferred );
965
}
966
967
private void parseADR( String[] params, String value )
968
{
969
// get address parts
970
String[] adr_parts = splitValueByCharacter( value, ';' );
971
972
// build address
973
value = "";
974
for( int a = 0; a < adr_parts.length; a++ )
975
if( adr_parts[ a ].length() > 0 )
976
{
977
// version 3.0 vCards allow further splitting by comma
978
if( _version.equals( "3.0" ) )
979
{
980
// split this part in to it's comma-separated bits and
981
// add them on individual lines
982
String[] adr_part_parts =
983
splitValueByCharacter( adr_parts[ a ], ',' );
984
for( int b = 0; b < adr_part_parts.length; b++ )
985
if( adr_part_parts[ b ].length() > 0 )
986
{
987
if( value.length() > 0 ) value += "\n";
988
value += adr_part_parts[ b ];
989
}
990
}
991
else
992
{
993
// add this part on an individual line
994
if( value.length() > 0 ) value += "\n";
995
value += adr_parts[ a ];
996
}
997
}
998
999
Set< String > types = extractTypes( params, Arrays.asList(
1000
"PREF", "WORK", "HOME" ) );
1001
1002
// add address
1003
int type;
1004
if( types.contains( "WORK" ) )
1005
type = TYPE_WORK;
1006
else
1007
type = TYPE_HOME;
1008
1009
addAddress( unescapeValue( value ), type );
1010
}
1011
1012
private void parseLABEL( String[] params, String value )
1013
{
1014
Set< String > types = extractTypes( params, Arrays.asList(
1015
"PREF", "WORK", "HOME" ) );
1016
1017
// add address
1018
int type;
1019
if( types.contains( "WORK" ) )
1020
type = TYPE_WORK;
1021
else
1022
type = TYPE_HOME;
1023
1024
addAddress( unescapeValue( value ), type );
1025
}
1026
1027
private void parseNOTE( String[] params, String value )
1028
{
1029
addNote( unescapeValue( value ) );
1030
}
1031
1032
private void parseBDAY( String[] params, String value )
1033
{
1034
setBirthday( value );
1035
}
1036
1037
public void finaliseVcard()
1038
throws ParseException, ContactNotIdentifiableException
625
"PREF", "WORK", "HOME", "INTERNET" ) );
626
627
// here's the logic...
628
boolean preferred = types.contains( "PREF" );
629
if( types.contains( "WORK" ) )
630
addEmail( value, Contacts.ContactMethods.TYPE_WORK, preferred );
631
else
632
addEmail( value, Contacts.ContactMethods.TYPE_HOME, preferred );
633
}
634
635
public void finaliseParsing()
636
throws ParseException, SkipContactException,
637
AbortImportException
1039
638
{
1040
639
// missing version (and data is present)
1041
if( _version == null && _content_lines != null )
640
if( _version == null && _buffers != null )
1042
641
throw new ParseException( R.string.error_vcf_malformed );
1043
642
1044
// finalise the parent class
1045
finalise();
643
// missing name properties?
644
if( _name_level == NAMELEVEL_NONE )
645
throw new ParseException( R.string.error_vcf_noname );
646
647
// check if we should import this one? If we've already got an 'N'-
648
// type name, this will already have been done by parseN() so we
649
// mustn't do this here (or it could prompt twice!)
650
if( _name_level < NAMELEVEL_N && !isImportRequired( getName() ) )
651
throw new SkipContactException();
1046
652
}
1047
653
1048
/**
1049
* Amongst the params, find the value of the first, only, of any with
1050
* the specified name.
1051
*
1052
* @param params
1053
* @param name
1054
* @return a value, or null
1055
*/
1056
654
private String checkParam( String[] params, String name )
1057
655
{
1058
String[] res = checkParams( params, name );
1059
return res.length > 0? res[ 0 ] : null;
1060
}
1061
1062
/**
1063
* Amongst the params, find the values of any with the specified name.
1064
*
1065
* @param params
1066
* @param name
1067
* @return an array of values, or null
1068
*/
1069
private String[] checkParams( String[] params, String name )
1070
{
1071
HashSet< String > ret = new HashSet< String >();
1072
1073
Pattern p = Pattern.compile(
1074
"^" + name + "[ \\t]*=[ \\t]*(\"?)(.*)\\1$",
1075
Pattern.CASE_INSENSITIVE );
656
Pattern p = Pattern.compile( "^" + name + "[ \\t]*=[ \\t]*(.*)$" );
1076
657
for( int i = 0; i < params.length; i++ ) {
1077
658
Matcher m = p.matcher( params[ i ] );
1078
659
if( m.matches() )
1079
ret.add( m.group( 2 ) );
660
return m.group( 1 );
1080
661
}
1081
1082
return (String[]) ret.toArray( new String[ ret.size() ] );
662
return null;
1083
663
}
1084
664
1085
/**
1086
* Amongst the params, return any type values present. For v2.1 vCards,
1087
* those types are just parameters. For v3.0, they are prefixed with
1088
* "TYPE=". There may also be multiple type parameters.
1089
*
1090
* @param params an array of params to look for types in
1091
* @param valid_types an list of upper-case type values to look for
1092
* @return a set of present type values
1093
*/
1094
665
private Set< String > extractTypes( String[] params,
1095
List< String > valid_types )
666
List< String > valid_types )
1096
667
{
1097
668
HashSet< String > types = new HashSet< String >();
1098
669
1099
670
// get 3.0-style TYPE= param
1100
String type_params[] = checkParams( params, "TYPE" );
1101
for( int a = 0; a < type_params.length; a++ )
1102
{
1103
// check for a comma-separated list of types (why? I don't think
1104
// this is in the specs!)
1105
String[] parts = type_params[ a ].split( "," );
1106
for( int i = 0; i < parts.length; i++ ) {
1107
String ucpart = parts[ i ].toUpperCase( Locale.ENGLISH );
1108
if( valid_types.contains( ucpart ) )
1109
types.add( ucpart );
1110
}
671
String type_param;
672
if( ( type_param = checkParam( params, "TYPE" ) ) != null ) {
673
String[] parts = type_param.split( "," );
674
for( int i = 0; i < parts.length; i++ )
675
if( valid_types.contains( parts[ i ] ) )
676
types.add( parts[ i ] );
1111
677
}
1112
678
1113
679
// get 2.1-style type param
1114
680
if( _version.equals( "2.1" ) ) {
1115
for( int i = 1; i < params.length; i++ ) {
1116
String ucparam = params[ i ].toUpperCase( Locale.ENGLISH );
1117
if( valid_types.contains( ucparam ) )
1118
types.add( ucparam );
1119
}
681
for( int i = 1; i < params.length; i++ )
682
if( valid_types.contains( params[ i ] ) )
683
types.add( params[ i ] );
1120
684
}
1121
685
1122
686
return types;
1126
690
{
1127
691
boolean another = false;
1128
692
1129
// unencode quoted-printable encoding, as per RFC1521 section 5.1
693
// unencode quoted-pritable encoding, as per RFC1521 section 5.1
1130
694
byte[] out = new byte[ in.limit() - in.position() ];
1131
695
int j = 0;
1132
696
for( int i = in.position(); i < in.limit(); i++ )
1137
701
{
1138
702
// we found a =XX format byte, add it
1139
703
out[ j ] = (byte)(
1140
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
1141
Character.digit( in.array()[ i + 2 ], 16 ) );
704
Character.digit( in.array()[ i + 1 ], 16 ) * 16 +
705
Character.digit( in.array()[ i + 2 ], 16 ) );
1142
706
i += 2;
1143
707
}
1144
708
else if( ch == '=' && i == in.limit() - 1 )
1145
709
{
1146
710
// we found a '=' at the end of a line signifying a multi-
1147
// line string, so we don't add it
711
// line string, so we don't add it.
1148
712
another = true;
1149
713
continue;
1150
714
}
Loggerhead is a web-based interface for Bazaar branches
Version: 1.18.2, Revision: 480