schema_init.c 102 KB
Newer Older
1
2
3
/* schema_init.c - init builtin schema */
/* $OpenLDAP$ */
/*
Kurt Zeilenga's avatar
Kurt Zeilenga committed
4
 * Copyright 1998-2000 The OpenLDAP Foundation, All Rights Reserved.
5
6
7
8
9
10
11
12
13
14
15
16
17
 * COPYING RESTRICTIONS APPLY, see COPYRIGHT file
 */

#include "portable.h"

#include <stdio.h>

#include <ac/ctype.h>
#include <ac/string.h>
#include <ac/socket.h>

#include "slap.h"
#include "ldap_pvt.h"
18

19
20
#define UTF8MATCH 1

21
22
#ifdef USE_MD5
#include "lutil_md5.h"
23
24
25
26
27
28
/* We should replace MD5 with a faster hash */
#define HASH_BYTES				LUTIL_MD5_BYTES
#define HASH_CONTEXT			lutil_MD5_CTX
#define HASH_Init(c)			lutil_MD5Init(c)
#define HASH_Update(c,buf,len)	lutil_MD5Update(c,buf,len)
#define HASH_Final(d,c)			lutil_MD5Final(d,c)
29
30
31
32
33
34
35
36
37
#else
#include "lutil_hash.h"
/* We should replace MD5 with a faster hash */
#define HASH_BYTES				LUTIL_HASH_BYTES
#define HASH_CONTEXT			lutil_HASH_CTX
#define HASH_Init(c)			lutil_HASHInit(c)
#define HASH_Update(c,buf,len)	lutil_HASHUpdate(c,buf,len)
#define HASH_Final(d,c)			lutil_HASHFinal(d,c)
#endif
38

39
/* recycled validatation routines */
40
#define berValidate						blobValidate
41
42
43
44

/* unimplemented pretters */
#define dnPretty						NULL
#define integerPretty					NULL
45
46

/* recycled matching routines */
47
#define bitStringMatch					octetStringMatch
48
#define integerMatch					caseIgnoreIA5Match
49
50
51
#define numericStringMatch				caseIgnoreMatch
#define objectIdentifierMatch			numericStringMatch
#define telephoneNumberMatch			numericStringMatch
52
#define telephoneNumberSubstringsMatch	caseIgnoreIA5SubstringsMatch
53
54
#define generalizedTimeMatch			numericStringMatch
#define generalizedTimeOrderingMatch	numericStringMatch
55
#define uniqueMemberMatch				dnMatch
56

57
58
/* approx matching rules */
#define directoryStringApproxMatchOID	"1.3.6.1.4.1.4203.666.4.4"
59
60
61
#define directoryStringApproxMatch  	approxMatch
#define directoryStringApproxIndexer 	approxIndexer
#define directoryStringApproxFilter  	approxFilter
62
#define IA5StringApproxMatchOID			"1.3.6.1.4.1.4203.666.4.5"
63
64
65
#define IA5StringApproxMatch  			approxMatch
#define IA5StringApproxIndexer			approxIndexer
#define IA5StringApproxFilter  			approxFilter
66

67
68
69
70
/* orderring matching rules */
#define caseIgnoreOrderingMatch			caseIgnoreMatch
#define caseExactOrderingMatch			caseExactMatch

71
/* unimplemented matching routines */
72
73
74
75
76
77
78
79
#define caseIgnoreListMatch				NULL
#define caseIgnoreListSubstringsMatch	NULL
#define presentationAddressMatch		NULL
#define protocolInformationMatch		NULL
#define integerFirstComponentMatch		NULL

#define OpenLDAPaciMatch				NULL
#define authPasswordMatch				NULL
80
81

/* recycled indexing/filtering routines */
82
83
#define dnIndexer						caseIgnoreIndexer
#define dnFilter						caseIgnoreFilter
84
85
#define integerIndexer					caseIgnoreIA5Indexer
#define integerFilter					caseIgnoreIA5Filter
86

87
88
89
90
static char *strcasechr( const char *str, int c )
{
	char *lower = strchr( str, TOLOWER(c) );
	char *upper = strchr( str, TOUPPER(c) );
91

92
93
94
95
96
97
98
99
	if( lower && upper ) {
		return lower < upper ? lower : upper;
	} else if ( lower ) {
		return lower;
	} else {
		return upper;
	}
}
100

101
102
103
static int
octetStringMatch(
	int *matchp,
104
	slap_mask_t flags,
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
	int match = value->bv_len - ((struct berval *) assertedValue)->bv_len;

	if( match == 0 ) {
		match = memcmp( value->bv_val,
			((struct berval *) assertedValue)->bv_val,
			value->bv_len );
	}

	*matchp = match;
	return LDAP_SUCCESS;
}

/* Index generation function */
int octetStringIndexer(
124
125
	slap_mask_t use,
	slap_mask_t flags,
126
127
128
129
130
131
132
133
134
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
	int i;
	size_t slen, mlen;
	struct berval **keys;
135
136
	HASH_CONTEXT   HASHcontext;
	unsigned char   HASHdigest[HASH_BYTES];
137
	struct berval digest;
138
139
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
140

141
142
143
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

144
145
146
147
148
149
150
151
152
153
	for( i=0; values[i] != NULL; i++ ) {
		/* just count them */
	}

	keys = ch_malloc( sizeof( struct berval * ) * (i+1) );

	slen = strlen( syntax->ssyn_oid );
	mlen = strlen( mr->smr_oid );

	for( i=0; values[i] != NULL; i++ ) {
154
		HASH_Init( &HASHcontext );
155
		if( prefix != NULL && prefix->bv_len > 0 ) {
156
			HASH_Update( &HASHcontext,
157
158
				prefix->bv_val, prefix->bv_len );
		}
159
		HASH_Update( &HASHcontext,
160
			syntax->ssyn_oid, slen );
161
		HASH_Update( &HASHcontext,
162
			mr->smr_oid, mlen );
163
		HASH_Update( &HASHcontext,
164
			values[i]->bv_val, values[i]->bv_len );
165
		HASH_Final( HASHdigest, &HASHcontext );
166
167
168
169
170
171
172
173
174
175
176
177
178

		keys[i] = ber_bvdup( &digest );
	}

	keys[i] = NULL;

	*keysp = keys;

	return LDAP_SUCCESS;
}

/* Index generation function */
int octetStringFilter(
179
180
	slap_mask_t use,
	slap_mask_t flags,
181
182
183
184
185
186
187
188
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	void * assertValue,
	struct berval ***keysp )
{
	size_t slen, mlen;
	struct berval **keys;
189
190
	HASH_CONTEXT   HASHcontext;
	unsigned char   HASHdigest[HASH_BYTES];
191
192
	struct berval *value = (struct berval *) assertValue;
	struct berval digest;
193
194
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
195
196
197
198
199
200

	slen = strlen( syntax->ssyn_oid );
	mlen = strlen( mr->smr_oid );

	keys = ch_malloc( sizeof( struct berval * ) * 2 );

201
	HASH_Init( &HASHcontext );
202
	if( prefix != NULL && prefix->bv_len > 0 ) {
203
		HASH_Update( &HASHcontext,
204
205
			prefix->bv_val, prefix->bv_len );
	}
206
	HASH_Update( &HASHcontext,
207
		syntax->ssyn_oid, slen );
208
	HASH_Update( &HASHcontext,
209
		mr->smr_oid, mlen );
210
	HASH_Update( &HASHcontext,
211
		value->bv_val, value->bv_len );
212
	HASH_Final( HASHdigest, &HASHcontext );
213
214
215
216
217
218
219
220

	keys[0] = ber_bvdup( &digest );
	keys[1] = NULL;

	*keysp = keys;

	return LDAP_SUCCESS;
}
221

222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
static int
dnValidate(
	Syntax *syntax,
	struct berval *in )
{
	int rc;
	char *dn;

	if( in->bv_len == 0 ) return LDAP_SUCCESS;

	dn = ch_strdup( in->bv_val );

	rc = dn_validate( dn ) == NULL
		? LDAP_INVALID_SYNTAX : LDAP_SUCCESS;

	ch_free( dn );
	return rc;
}

static int
dnNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
	struct berval *out = ber_bvdup( val );

	if( out->bv_len != 0 ) {
		char *dn;
#ifdef USE_DN_NORMALIZE
		dn = dn_normalize( out->bv_val );
#else
		dn = dn_validate( out->bv_val );
#endif

		if( dn == NULL ) {
			ber_bvfree( out );
			return LDAP_INVALID_SYNTAX;
		}

		out->bv_val = dn;
		out->bv_len = strlen( dn );
	}

	*normalized = out;
	return LDAP_SUCCESS;
}

static int
dnMatch(
272
	int *matchp,
273
	slap_mask_t flags,
274
275
276
277
278
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
279
	int match;
280
281
	struct berval *asserted = (struct berval *) assertedValue;
	
282
283
284
	match = value->bv_len - asserted->bv_len;

	if( match == 0 ) {
285
#ifdef USE_DN_NORMALIZE
286
		match = strcmp( value->bv_val, asserted->bv_val );
287
#else
288
		match = strcasecmp( value->bv_val, asserted->bv_val );
289
#endif
290
291
	}

292
293
294
295
296
#ifdef NEW_LOGGING
        LDAP_LOG(( "schema", LDAP_LEVEL_ENTRY,
                   "dnMatch: %d\n    %s\n    %s\n", match,
                   value->bv_val, asserted->bv_val ));
#else
297
298
	Debug( LDAP_DEBUG_ARGS, "dnMatch %d\n\t\"%s\"\n\t\"%s\"\n",
	    match, value->bv_val, asserted->bv_val );
299
300
#endif

301
302
303

	*matchp = match;
	return LDAP_SUCCESS;
304
}
305

306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
static int
nameUIDValidate(
	Syntax *syntax,
	struct berval *in )
{
	int rc;
	struct berval *dn;

	if( in->bv_len == 0 ) return LDAP_SUCCESS;

	dn = ber_bvdup( in );

	if( dn->bv_val[dn->bv_len-1] == '\'' ) {
		/* assume presence of optional UID */
		ber_len_t i;

		for(i=dn->bv_len-2; i>2; i--) {
			if( dn->bv_val[i] != '0' &&	dn->bv_val[i] != '1' ) {
				break;
			}
		}
		if( dn->bv_val[i] != '\'' ) {
			return LDAP_INVALID_SYNTAX;
		}
		if( dn->bv_val[i-1] != 'B' ) {
			return LDAP_INVALID_SYNTAX;
		}
		if( dn->bv_val[i-2] != '#' ) {
			return LDAP_INVALID_SYNTAX;
		}

		/* trim the UID to allow use of dn_validate */
		dn->bv_val[i-2] = '\0';
	}

	rc = dn_validate( dn->bv_val ) == NULL
		? LDAP_INVALID_SYNTAX : LDAP_SUCCESS;

	ber_bvfree( dn );
	return rc;
}

static int
nameUIDNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
	struct berval *out = ber_bvdup( val );

	if( out->bv_len != 0 ) {
		char *dn;
		ber_len_t dnlen;
		char *uid = NULL;
		ber_len_t uidlen = 0;

		if( out->bv_val[out->bv_len-1] == '\'' ) {
			/* assume presence of optional UID */
			uid = strrchr( out->bv_val, '#' );

			if( uid == NULL ) {
				ber_bvfree( out );
				return LDAP_INVALID_SYNTAX;
			}

			uidlen = out->bv_len - (out->bv_val - uid);
			/* temporarily trim the UID */
			*uid = '\0';
		}

#ifdef USE_DN_NORMALIZE
		dn = dn_normalize( out->bv_val );
#else
		dn = dn_validate( out->bv_val );
#endif

		if( dn == NULL ) {
			ber_bvfree( out );
			return LDAP_INVALID_SYNTAX;
		}

		dnlen = strlen(dn);

		if( uidlen ) {
			/* restore the separator */
			*uid = '#';
			/* shift the UID */
			SAFEMEMCPY( &dn[dnlen], uid, uidlen );
		}

		out->bv_val = dn;
		out->bv_len = dnlen + uidlen;
	}

	*normalized = out;
	return LDAP_SUCCESS;
}

404
405
406
407
408
409
410
411
412
static int
inValidate(
	Syntax *syntax,
	struct berval *in )
{
	/* any value allowed */
	return LDAP_OTHER;
}

413
static int
414
blobValidate(
415
416
417
418
	Syntax *syntax,
	struct berval *in )
{
	/* any value allowed */
419
	return LDAP_SUCCESS;
420
421
}

422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
static int
bitStringValidate(
	Syntax *syntax,
	struct berval *in )
{
	ber_len_t i;

	/* very unforgiving validation, requires no normalization
	 * before simplistic matching
	 */
	if( in->bv_len < 3 ) {
		return LDAP_INVALID_SYNTAX;
	}
	if( in->bv_val[0] != 'B' ||
		in->bv_val[1] != '\'' ||
		in->bv_val[in->bv_len-1] != '\'' )
	{
		return LDAP_INVALID_SYNTAX;
	}

	for( i=in->bv_len-2; i>1; i-- ) {
		if( in->bv_val[i] != '0' && in->bv_val[i] != '1' ) {
			return LDAP_INVALID_SYNTAX;
		}
	}

	return LDAP_SUCCESS;
}

451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
/*
 * Handling boolean syntax and matching is quite rigid.
 * A more flexible approach would be to allow a variety
 * of strings to be normalized and prettied into TRUE
 * and FALSE.
 */
static int
booleanValidate(
	Syntax *syntax,
	struct berval *in )
{
	/* very unforgiving validation, requires no normalization
	 * before simplistic matching
	 */

	if( in->bv_len == 4 ) {
		if( !memcmp( in->bv_val, "TRUE", 4 ) ) {
			return LDAP_SUCCESS;
		}
	} else if( in->bv_len == 5 ) {
		if( !memcmp( in->bv_val, "FALSE", 5 ) ) {
			return LDAP_SUCCESS;
		}
	}

	return LDAP_INVALID_SYNTAX;
}

static int
booleanMatch(
	int *matchp,
482
	slap_mask_t flags,
483
484
485
486
487
488
489
490
491
492
493
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
	/* simplistic matching allowed by rigid validation */
	struct berval *asserted = (struct berval *) assertedValue;
	*matchp = value->bv_len != asserted->bv_len;
	return LDAP_SUCCESS;
}

494
#if UTF8MATCH
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
static int
UTF8casecmp(
	struct berval *right,
	struct berval *left )
{
	ber_len_t r, l;
	int rlen, llen;
	ldap_unicode_t ru, lu;
	ldap_unicode_t ruu, luu;

	for( r=0, l=0;
		r < right->bv_len && l < left->bv_len;
		r+=rlen, l+=llen )
	{
		/*
		 * XXYYZ: we convert to ucs4 even though -llunicode
		 * expects ucs2 in an unsigned long
		 */
		ru = ldap_utf8_to_ucs4( &right->bv_val[r] );
		if( ru == LDAP_UCS4_INVALID ) {
			return 1;
		}

		lu = ldap_utf8_to_ucs4( &left->bv_val[l] );
		if( lu == LDAP_UCS4_INVALID ) {
			return -1;
		}

		ruu = uctoupper( ru );
		luu = uctoupper( lu );

		if( ruu > luu ) {
			return 1;
		} else if( luu > ruu ) {
			return -1;
		}

		rlen = LDAP_UTF8_CHARLEN( &right->bv_val[r] );
		llen = LDAP_UTF8_CHARLEN( &left->bv_val[l] );
	}

	if( r < right->bv_len ) {
		/* less left */
		return -1;
	}

	if( l < left->bv_len ) {
		/* less right */
		return 1;
	}

	return 0;
}
Kurt Zeilenga's avatar
Kurt Zeilenga committed
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638

/* case insensitive UTF8 strncmp with offset for second string */
static int
UTF8oncasecmp(
	struct berval *right,
	struct berval *left,
	ber_len_t len,
	ber_len_t offset )
{
	ber_len_t r, l;
	int rlen, llen;
	int rslen, lslen;
	ldap_unicode_t ru, lu;
	ldap_unicode_t ruu, luu;

	rslen = len < right->bv_len ? len : right->bv_len;
	lslen = len + offset < left->bv_len ? len : left->bv_len;

	for( r = 0, l = offset;
		r < rslen && l < lslen;
		r+=rlen, l+=llen )
	{
		/*
		 * XXYYZ: we convert to ucs4 even though -llunicode
		 * expects ucs2 in an unsigned long
		 */
		ru = ldap_utf8_to_ucs4( &right->bv_val[r] );
		if( ru == LDAP_UCS4_INVALID ) {
			return 1;
		}

		lu = ldap_utf8_to_ucs4( &left->bv_val[l] );
		if( lu == LDAP_UCS4_INVALID ) {
			return -1;
		}

		ruu = uctoupper( ru );
		luu = uctoupper( lu );

		if( ruu > luu ) {
			return 1;
		} else if( luu > ruu ) {
			return -1;
		}

		rlen = LDAP_UTF8_CHARLEN( &right->bv_val[r] );
		llen = LDAP_UTF8_CHARLEN( &left->bv_val[l] );
	}

	if( r < rslen ) {
		/* less left */
		return -1;
	}

	if( l < lslen ) {
		/* less right */
		return 1;
	}

	return 0;
}

static char *UTF8casechr( const char *str, const char *c )
{
	char *p, *lower, *upper;
	ldap_ucs4_t tch, ch = ldap_utf8_to_ucs4(c);

	tch = uctolower ( ch );
	for( p = (char *) str; *p != '\0'; LDAP_UTF8_INCR(p) ) {
		if( ldap_utf8_to_ucs4( p ) == tch ) {
			break;
		} 
	}
	lower = *p != '\0' ? p : NULL;

	tch = uctoupper ( ch );
	for( p = (char *) str; *p != '\0'; LDAP_UTF8_INCR(p) ) {
		if( ldap_utf8_to_ucs4( p ) == tch ) {
			break;
		} 
	}
	upper = *p != '\0' ? p : NULL;
	
	if( lower && upper ) {
		return lower < upper ? lower : upper;
	} else if ( lower ) {
		return lower;
	} else {
		return upper;
	}
}
639
640
#endif

641
642
643
644
645
646
647
648
649
static int
UTF8StringValidate(
	Syntax *syntax,
	struct berval *in )
{
	ber_len_t count;
	int len;
	unsigned char *u = in->bv_val;

650
651
	if( !in->bv_len ) return LDAP_INVALID_SYNTAX;

652
	for( count = in->bv_len; count > 0; count-=len, u+=len ) {
653
654
655
656
		/* get the length indicated by the first byte */
		len = LDAP_UTF8_CHARLEN( u );

		/* should not be zero */
657
		if( len == 0 ) return LDAP_INVALID_SYNTAX;
658
659
660

		/* make sure len corresponds with the offset
			to the next character */
661
		if( LDAP_UTF8_OFFSET( u ) != len ) return LDAP_INVALID_SYNTAX;
662
663
	}

664
	if( count != 0 ) return LDAP_INVALID_SYNTAX;
665

666
	return LDAP_SUCCESS;
667
668
669
670
671
672
673
674
675
}

static int
UTF8StringNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
	struct berval *newval;
676
	char *p, *q, *s;
677

678
	newval = ch_malloc( sizeof( struct berval ) );
679

680
	p = val->bv_val;
681

682
683
684
	/* Ignore initial whitespace */
	while ( ldap_utf8_isspace( p ) ) {
		LDAP_UTF8_INCR( p );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
685
	}
686

687
688
	if( *p == '\0' ) {
		ch_free( newval );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
689
690
		return LDAP_INVALID_SYNTAX;
	}
691

692
693
694
	newval->bv_val = ch_strdup( p );
	p = q = newval->bv_val;
	s = NULL;
695

696
697
	while ( *p ) {
		int len;
698

699
700
701
702
703
		if ( ldap_utf8_isspace( p ) ) {
			len = LDAP_UTF8_COPY(q,p);
			s=q;
			p+=len;
			q+=len;
704

705
706
707
708
			/* Ignore the extra whitespace */
			while ( ldap_utf8_isspace( p ) ) {
				LDAP_UTF8_INCR( p );
			}
Kurt Zeilenga's avatar
Kurt Zeilenga committed
709
		} else {
710
711
712
713
			len = LDAP_UTF8_COPY(q,p);
			s=NULL;
			p+=len;
			q+=len;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
714
		}
715
716
717
718
	}

	assert( *newval->bv_val );
	assert( newval->bv_val < p );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
719
	assert( q <= p );
720

721
722
723
724
725
726
727
728
729
730
731
	/* cannot start with a space */
	assert( !ldap_utf8_isspace(newval->bv_val) );

	/*
	 * If the string ended in space, backup the pointer one
	 * position.  One is enough because the above loop collapsed
	 * all whitespace to a single space.
	 */

	if ( s != NULL ) {
		q = s;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
732
	}
733

734
735
736
737
738
739
740
741
	/* cannot end with a space */
	assert( !ldap_utf8_isspace( LDAP_UTF8_PREV(q) ) );

	/* null terminate */
	*q = '\0';

	newval->bv_len = q - newval->bv_val;
	*normalized = newval;
742

743
	return LDAP_SUCCESS;
744
745
}

746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
#if defined(SLAPD_APPROX_MULTISTRING)

#if defined(SLAPD_APPROX_INITIALS)
#define SLAPD_APPROX_DELIMITER "._ "
#define SLAPD_APPROX_WORDLEN 2
#else
#define SLAPD_APPROX_DELIMITER " "
#define SLAPD_APPROX_WORDLEN 1
#endif

static int
approxMatch(
	int *matchp,
	slap_mask_t flags,
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
	char *val, *assertv, **values, **words, *c;
	int i, count, len, nextchunk=0, nextavail=0;


	/* Isolate how many words there are */
	val = ch_strdup( value->bv_val );
	for( c=val,count=1; *c; c++ ) {
		c = strpbrk( c, SLAPD_APPROX_DELIMITER );
		if ( c == NULL ) break;
		*c = '\0';
		count++;
	}

	/* Get a phonetic copy of each word */
	words = (char **)ch_malloc( count * sizeof(char *) );
	values = (char **)ch_malloc( count * sizeof(char *) );
	for( c=val,i=0;  i<count;  i++,c+=strlen(c)+1 ) {
		words[i] = c;
		values[i] = phonetic(c);
	}


	/* Work through the asserted value's words, to see if  at least some
	   of the words are there, in the same order. */
	assertv = ch_strdup( ((struct berval *)assertedValue)->bv_val );
	len = 0;
	while ( nextchunk < ((struct berval *)assertedValue)->bv_len ) {
		len = strcspn( assertv + nextchunk, SLAPD_APPROX_DELIMITER);
793
794
795
796
		if( len == 0 ) {
			nextchunk++;
			continue;
		}
797
#if defined(SLAPD_APPROX_INITIALS)
798
		else if( len == 1 ) {
799
800
			/* Single letter words need to at least match one word's initial */
			for( i=nextavail; i<count; i++ )
801
802
				if( !strncasecmp( assertv+nextchunk, words[i], 1 )) {
					nextavail=i+1;
803
					break;
804
				}
805
806
		}
#endif
807
		else {
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
			/* Isolate the next word in the asserted value and phonetic it */
			assertv[nextchunk+len] = '\0';
			val = phonetic( assertv + nextchunk );

			/* See if this phonetic chunk is in the remaining words of *value */
			for( i=nextavail; i<count; i++ ){
				if( !strcmp( val, values[i] ) ){
					nextavail = i+1;
					break;
				}
			}
		}

		/* This chunk in the asserted value was NOT within the *value. */
		if( i >= count ) {
			nextavail=-1;
			break;
		}

		/* Go on to the next word in the asserted value */
		nextchunk += len+1;
	}

	/* If some of the words were seen, call it a match */
	if( nextavail > 0 ) {
		*matchp = 0;
	}
	else {
		*matchp = 1;
	}

	/* Cleanup allocs */
	ch_free( assertv );
	for( i=0; i<count; i++ ) {
		ch_free( values[i] );
	}
	ch_free( values );
	ch_free( words );
	ch_free( val );

	return LDAP_SUCCESS;
}


int 
approxIndexer(
	slap_mask_t use,
	slap_mask_t flags,
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
	char *val, *c;
	int i,j, len, wordcount, keycount=0;
	struct berval **newkeys, **keys=NULL;


	for( j=0; values[j] != NULL; j++ ) {

		/* Isolate how many words there are. There will be a key for each */
		val = ch_strdup( values[j]->bv_val );
		for( wordcount=0,c=val;  *c;  c++) {
			len = strcspn(c, SLAPD_APPROX_DELIMITER);
			if( len >= SLAPD_APPROX_WORDLEN ) wordcount++;
			c+= len;
			if (*c == '\0') break;
			*c = '\0';
		}

		/* Allocate/increase storage to account for new keys */
		newkeys = (struct berval **)ch_malloc( (keycount + wordcount + 1) 
		   * sizeof(struct berval *) );
		memcpy( newkeys, keys, keycount * sizeof(struct berval *) );
		if( keys ) ch_free( keys );
		keys = newkeys;

		/* Get a phonetic copy of each word */
		for( c=val,i=0;  i<wordcount;  c+=len+1  ) {
			len = strlen( c );
			if( len < SLAPD_APPROX_WORDLEN ) continue;
			keys[keycount] = (struct berval *)ch_malloc( sizeof(struct berval) );
			keys[keycount]->bv_val = phonetic( c );
			keys[keycount]->bv_len = strlen( keys[keycount]->bv_val );
			keycount++;
			i++;
		}

		ch_free( val );
	}
	keys[keycount] = NULL;
	*keysp = keys;

	return LDAP_SUCCESS;
}


int 
approxFilter(
	slap_mask_t use,
	slap_mask_t flags,
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	void * assertValue,
	struct berval ***keysp )
{
	char *val, *c;
	int i, count, len;
	struct berval **keys;


	/* Isolate how many words there are. There will be a key for each */
	val = ch_strdup( ((struct berval *)assertValue)->bv_val );
	for( count=0,c=val;  *c;  c++) {
		len = strcspn(c, SLAPD_APPROX_DELIMITER);
		if( len >= SLAPD_APPROX_WORDLEN ) count++;
		c+= len;
		if (*c == '\0') break;
		*c = '\0';
	}

	/* Allocate storage for new keys */
	keys = (struct berval **)ch_malloc( (count + 1) * sizeof(struct berval *) );

	/* Get a phonetic copy of each word */
	for( c=val,i=0;  i<count; c+=len+1 ) {
		len = strlen(c);
		if( len < SLAPD_APPROX_WORDLEN ) continue;
		keys[i] = (struct berval *)ch_malloc( sizeof(struct berval) );
		keys[i]->bv_val = phonetic( c );
		keys[i]->bv_len = strlen( keys[i]->bv_val );
		i++;
	}

	ch_free( val );

	keys[count] = NULL;
	*keysp = keys;

	return LDAP_SUCCESS;
}


#else
/* No other form of Approximate Matching is defined */

static int
approxMatch(
	int *matchp,
	slap_mask_t flags,
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
	char *vapprox, *avapprox;

	vapprox = phonetic( value->bv_val );
	avapprox = phonetic( ((struct berval *)assertedValue)->bv_val);

	*matchp = strcmp( vapprox, avapprox );

	ch_free( vapprox );
	ch_free( avapprox );

	return LDAP_SUCCESS;
}

int 
approxIndexer(
	slap_mask_t use,
	slap_mask_t flags,
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
	int i;
	struct berval **keys;


	for( i=0; values[i] != NULL; i++ ) {
		/* just count them */
	}
	assert( i > 0 );

	keys = (struct berval **)ch_malloc( sizeof( struct berval * ) * (i+1) );

	/* Copy each value and run it through phonetic() */
	for( i=0; values[i] != NULL; i++ ) {
		keys[i] = ch_malloc( sizeof( struct berval * ) );
		keys[i]->bv_val = phonetic( values[i]->bv_val );
		keys[i]->bv_len = strlen( keys[i]->bv_val );
	}
	keys[i] = NULL;

	*keysp = keys;
	return LDAP_SUCCESS;
}


int 
approxFilter(
	slap_mask_t use,
	slap_mask_t flags,
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	void * assertValue,
	struct berval ***keysp )
{
	struct berval **keys;


	keys = (struct berval **)ch_malloc( sizeof( struct berval * ) * 2 );

	/* Copy the value and run it through phonetic() */
	keys[0] = ch_malloc( sizeof( struct berval * ) );
	keys[0]->bv_val = phonetic( ((struct berval *)assertValue)->bv_val );
	keys[0]->bv_len = strlen( keys[0]->bv_val );
	keys[1] = NULL;

	*keysp = keys;
	return LDAP_SUCCESS;
}
#endif


1039
static int
1040
caseExactMatch(
1041
	int *matchp,
1042
	slap_mask_t flags,
1043
1044
1045
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
1046
	void *assertedValue )
1047
{
1048
1049
1050
	int match = value->bv_len - ((struct berval *) assertedValue)->bv_len;

	if( match == 0 ) {
1051
1052
1053
		match = strncmp( value->bv_val,
			((struct berval *) assertedValue)->bv_val,
			value->bv_len );
1054
1055
1056
	}

	*matchp = match;
1057
	return LDAP_SUCCESS;
1058
1059
}

1060
static int
1061
caseExactSubstringsMatch(
1062
	int *matchp,
1063
	slap_mask_t flags,
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
	int match = 0;
	SubstringsAssertion *sub = assertedValue;
	struct berval left = *value;
	int i;
	ber_len_t inlen=0;

1075
	/* Add up asserted input length */
1076
1077
1078
1079
	if( sub->sa_initial ) {
		inlen += sub->sa_initial->bv_len;
	}
	if( sub->sa_any ) {
1080
1081
		for(i=0; sub->sa_any[i] != NULL; i++) {
			inlen += sub->sa_any[i]->bv_len;
1082
1083
1084
1085
1086
1087
1088
		}
	}
	if( sub->sa_final ) {
		inlen += sub->sa_final->bv_len;
	}

	if( sub->sa_initial ) {
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1089
1090
1091
1092
1093
		if( inlen > left.bv_len ) {
			match = 1;
			goto done;
		}

1094
1095
		match = strncmp( sub->sa_initial->bv_val, left.bv_val,
			sub->sa_initial->bv_len );
1096
1097
1098
1099
1100
1101
1102

		if( match != 0 ) {
			goto done;
		}

		left.bv_val += sub->sa_initial->bv_len;
		left.bv_len -= sub->sa_initial->bv_len;
1103
		inlen -= sub->sa_initial->bv_len;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1104
	}
1105

Kurt Zeilenga's avatar
Kurt Zeilenga committed
1106
	if( sub->sa_final ) {
1107
1108
1109
1110
		if( inlen > left.bv_len ) {
			match = 1;
			goto done;
		}
1111

1112
1113
1114
		match = strncmp( sub->sa_final->bv_val,
			&left.bv_val[left.bv_len - sub->sa_final->bv_len],
			sub->sa_final->bv_len );
1115
1116
1117
1118
1119
1120

		if( match != 0 ) {
			goto done;
		}

		left.bv_len -= sub->sa_final->bv_len;
1121
		inlen -= sub->sa_final->bv_len;
1122
1123
1124
	}

	if( sub->sa_any ) {
1125
1126
1127
1128
1129
		for(i=0; sub->sa_any[i]; i++) {
			ber_len_t idx;
			char *p;

retry:
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1130
1131
1132
1133
1134
1135
			if( inlen > left.bv_len ) {
				/* not enough length */
				match = 1;
				goto done;
			}

1136
1137
1138
1139
			if( sub->sa_any[i]->bv_len == 0 ) {
				continue;
			}

1140
			p = strchr( left.bv_val, *sub->sa_any[i]->bv_val );
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163

			if( p == NULL ) {
				match = 1;
				goto done;
			}

			idx = p - left.bv_val;
			assert( idx < left.bv_len );

			if( idx >= left.bv_len ) {
				/* this shouldn't happen */
				return LDAP_OTHER;
			}

			left.bv_val = p;
			left.bv_len -= idx;

			if( sub->sa_any[i]->bv_len > left.bv_len ) {
				/* not enough left */
				match = 1;
				goto done;
			}

1164
1165
1166
			match = strncmp( left.bv_val,
				sub->sa_any[i]->bv_val,
				sub->sa_any[i]->bv_len );
1167
1168

			if( match != 0 ) {
1169
1170
				left.bv_val++;
				left.bv_len--;
1171
1172
1173
1174
1175
				goto retry;
			}

			left.bv_val += sub->sa_any[i]->bv_len;
			left.bv_len -= sub->sa_any[i]->bv_len;
1176
			inlen -= sub->sa_any[i]->bv_len;
1177
		}
1178
1179
1180
1181
1182
1183
1184
	}

done:
	*matchp = match;
	return LDAP_SUCCESS;
}

1185
/* Index generation function */
1186
int caseExactIndexer(
1187
1188
	slap_mask_t use,
	slap_mask_t flags,
1189
1190
1191
1192
1193
1194
1195
1196
1197
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
	int i;
	size_t slen, mlen;
	struct berval **keys;
1198
1199
	HASH_CONTEXT   HASHcontext;
	unsigned char   HASHdigest[HASH_BYTES];
1200
	struct berval digest;
1201
1202
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1203

1204
1205
1206
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
	for( i=0; values[i] != NULL; i++ ) {
		/* just count them */
	}

	keys = ch_malloc( sizeof( struct berval * ) * (i+1) );

	slen = strlen( syntax->ssyn_oid );
	mlen = strlen( mr->smr_oid );

	for( i=0; values[i] != NULL; i++ ) {
		struct berval *value = values[i];

1219
		HASH_Init( &HASHcontext );
1220
		if( prefix != NULL && prefix->bv_len > 0 ) {
1221
			HASH_Update( &HASHcontext,
1222
1223
				prefix->bv_val, prefix->bv_len );
		}
1224
		HASH_Update( &HASHcontext,
1225
			syntax->ssyn_oid, slen );
1226
		HASH_Update( &HASHcontext,
1227
			mr->smr_oid, mlen );
1228
		HASH_Update( &HASHcontext,
1229
			value->bv_val, value->bv_len );
1230
		HASH_Final( HASHdigest, &HASHcontext );
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240

		keys[i] = ber_bvdup( &digest );
	}

	keys[i] = NULL;
	*keysp = keys;
	return LDAP_SUCCESS;
}

/* Index generation function */
1241
int caseExactFilter(
1242
1243
	slap_mask_t use,
	slap_mask_t flags,
1244
1245
1246
1247
1248
1249
1250
1251
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	void * assertValue,
	struct berval ***keysp )
{
	size_t slen, mlen;
	struct berval **keys;
1252
1253
	HASH_CONTEXT   HASHcontext;
	unsigned char   HASHdigest[HASH_BYTES];
1254
1255
	struct berval *value;
	struct berval digest;
1256
1257
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1258
1259
1260
1261
1262
1263
1264
1265

	slen = strlen( syntax->ssyn_oid );
	mlen = strlen( mr->smr_oid );

	value = (struct berval *) assertValue;

	keys = ch_malloc( sizeof( struct berval * ) * 2 );

1266
	HASH_Init( &HASHcontext );
1267
	if( prefix != NULL && prefix->bv_len > 0 ) {
1268
		HASH_Update( &HASHcontext,
1269
1270
			prefix->bv_val, prefix->bv_len );
	}
1271
	HASH_Update( &HASHcontext,
1272
		syntax->ssyn_oid, slen );
1273
	HASH_Update( &HASHcontext,
1274
		mr->smr_oid, mlen );
1275
	HASH_Update( &HASHcontext,
1276
		value->bv_val, value->bv_len );
1277
	HASH_Final( HASHdigest, &HASHcontext );
1278
1279
1280
1281
1282
1283
1284

	keys[0] = ber_bvdup( &digest );
	keys[1] = NULL;

	*keysp = keys;
	return LDAP_SUCCESS;
}
1285

1286
/* Substrings Index generation function */
1287
int caseExactSubstringsIndexer(
1288
1289
	slap_mask_t use,
	slap_mask_t flags,
1290
1291
1292
1293
1294
1295
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
1296
	ber_len_t i, nkeys;
1297
1298
	size_t slen, mlen;
	struct berval **keys;
1299
1300
	HASH_CONTEXT   HASHcontext;
	unsigned char   HASHdigest[HASH_BYTES];
1301
	struct berval digest;
1302
1303
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1304

1305
1306
1307
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

1308
1309
1310
	nkeys=0;
	for( i=0; values[i] != NULL; i++ ) {
		/* count number of indices to generate */
1311
		if( values[i]->bv_len < SLAP_INDEX_SUBSTR_MINLEN ) {
1312
1313
1314
			continue;
		}

1315
		if( flags & SLAP_INDEX_SUBSTR_INITIAL ) {
1316
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
1317
				nkeys += SLAP_INDEX_SUBSTR_MAXLEN -
1318
					( SLAP_INDEX_SUBSTR_MINLEN - 1);
1319
			} else {
1320
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MINLEN - 1 );
1321
1322
1323
			}
		}

1324
		if( flags & SLAP_INDEX_SUBSTR_ANY ) {
1325
1326
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MAXLEN - 1 );
1327
1328
1329
			}
		}

1330
		if( flags & SLAP_INDEX_SUBSTR_FINAL ) {
1331
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
1332
				nkeys += SLAP_INDEX_SUBSTR_MAXLEN -
1333
					( SLAP_INDEX_SUBSTR_MINLEN - 1);
1334
			} else {
1335
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MINLEN - 1 );
1336
			}
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
		}
	}

	if( nkeys == 0 ) {
		/* no keys to generate */
		*keysp = NULL;
		return LDAP_SUCCESS;
	}

	keys = ch_malloc( sizeof( struct berval * ) * (nkeys+1) );

	slen = strlen( syntax->ssyn_oid );
	mlen = strlen( mr->smr_oid );

	nkeys=0;
	for( i=0; values[i] != NULL; i++ ) {
1353
		ber_len_t j,max;
1354
1355
		struct berval *value;

1356
		value = values[i];
1357
		if( value->bv_len < SLAP_INDEX_SUBSTR_MINLEN ) continue;
1358

1359
		if( ( flags & SLAP_INDEX_SUBSTR_ANY ) &&
1360
			( value->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) )
1361
1362
		{
			char pre = SLAP_INDEX_SUBSTR_PREFIX;
1363
			max = value->bv_len - ( SLAP_INDEX_SUBSTR_MAXLEN - 1);
1364

1365
			for( j=0; j<max; j++ ) {
1366
				HASH_Init( &HASHcontext );
1367
				if( prefix != NULL && prefix->bv_len > 0 ) {
1368
					HASH_Update( &HASHcontext,
1369
1370
						prefix->bv_val, prefix->bv_len );
				}
1371

1372
				HASH_Update( &HASHcontext,
1373
					&pre, sizeof( pre ) );
1374
				HASH_Update( &HASHcontext,
1375
					syntax->ssyn_oid, slen );
1376
				HASH_Update( &HASHcontext,
1377
					mr->smr_oid, mlen );
1378
				HASH_Update( &HASHcontext,
1379
1380
					&value->bv_val[j],
					SLAP_INDEX_SUBSTR_MAXLEN );
1381
				HASH_Final( HASHdigest, &HASHcontext );
1382
1383
1384
1385
1386

				keys[nkeys++] = ber_bvdup( &digest );
			}
		}

1387
1388
		max = SLAP_INDEX_SUBSTR_MAXLEN < value->bv_len
			? SLAP_INDEX_SUBSTR_MAXLEN : value->bv_len;
1389

1390
		for( j=SLAP_INDEX_SUBSTR_MINLEN; j<=max; j++ ) {
1391
1392
			char pre;

1393
			if( flags & SLAP_INDEX_SUBSTR_INITIAL ) {
1394
				pre = SLAP_INDEX_SUBSTR_INITIAL_PREFIX;
1395
				HASH_Init( &HASHcontext );
1396
				if( prefix != NULL && prefix->bv_len > 0 ) {
1397
					HASH_Update( &HASHcontext,
1398
1399
						prefix->bv_val, prefix->bv_len );
				}
1400
				HASH_Update( &HASHcontext,
1401
					&pre, sizeof( pre ) );
1402
				HASH_Update( &HASHcontext,
1403
					syntax->ssyn_oid, slen );
1404
				HASH_Update( &HASHcontext,
1405
					mr->smr_oid, mlen );
1406
				HASH_Update( &HASHcontext,
1407
					value->bv_val, j );
1408
				HASH_Final( HASHdigest, &HASHcontext );
1409
1410
1411
1412

				keys[nkeys++] = ber_bvdup( &digest );
			}

1413
			if( flags & SLAP_INDEX_SUBSTR_FINAL ) {
1414
				pre = SLAP_INDEX_SUBSTR_FINAL_PREFIX;
1415
				HASH_Init( &HASHcontext );
1416
				if( prefix != NULL && prefix->bv_len > 0 ) {
1417
					HASH_Update( &HASHcontext,
1418
1419
						prefix->bv_val, prefix->bv_len );
				}
1420
				HASH_Update( &HASHcontext,
1421
					&pre, sizeof( pre ) );
1422
				HASH_Update( &HASHcontext,
1423
					syntax->ssyn_oid, slen );
1424
				HASH_Update( &HASHcontext,
1425
					mr->smr_oid, mlen );
1426
				HASH_Update( &HASHcontext,
1427
					&value->bv_val[value->bv_len-j], j );
1428
				HASH_Final( HASHdigest, &HASHcontext );
1429
1430
1431
1432
1433
1434
1435

				keys[nkeys++] = ber_bvdup( &digest );
			}

		}
	}

1436
1437
1438
1439
1440
1441
1442
1443
	if( nkeys > 0 ) {
		keys[nkeys] = NULL;
		*keysp = keys;
	} else {
		ch_free( keys );
		*keysp = NULL;
	}