schema_init.c 104 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				caseIgnoreIA5Match
#define objectIdentifierMatch			caseIgnoreIA5Match
#define telephoneNumberMatch			caseIgnoreIA5Match
52
#define telephoneNumberSubstringsMatch	caseIgnoreIA5SubstringsMatch
53
54
#define generalizedTimeMatch			caseIgnoreIA5Match
#define generalizedTimeOrderingMatch	caseIgnoreIA5Match
55
#define uniqueMemberMatch				dnMatch
56

57
58
/* approx matching rules */
#define directoryStringApproxMatchOID	"1.3.6.1.4.1.4203.666.4.4"
Gary Williams's avatar
Gary Williams committed
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"
Gary Williams's avatar
Gary Williams committed
63
#define IA5StringApproxMatch			approxMatch
64
#define IA5StringApproxIndexer			approxIndexer
Gary Williams's avatar
Gary Williams committed
65
#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
#define caseIgnoreListMatch				NULL
#define caseIgnoreListSubstringsMatch	NULL
#define protocolInformationMatch		NULL
#define integerFirstComponentMatch		NULL

#define OpenLDAPaciMatch				NULL
#define authPasswordMatch				NULL
79
80

/* recycled indexing/filtering routines */
81
82
#define dnIndexer						caseIgnoreIndexer
#define dnFilter						caseIgnoreFilter
83
84
#define integerIndexer					caseIgnoreIA5Indexer
#define integerFilter					caseIgnoreIA5Filter
85

86
87
88
89
90
#define telephoneNumberIndexer			caseIgnoreIA5Indexer
#define telephoneNumberFilter			caseIgnoreIA5Filter
#define telephoneNumberSubstringsIndexer	caseIgnoreIA5SubstringsIndexer
#define telephoneNumberSubstringsFilter		caseIgnoreIA5SubstringsFilter

91
92
93
94
static char *strcasechr( const char *str, int c )
{
	char *lower = strchr( str, TOLOWER(c) );
	char *upper = strchr( str, TOUPPER(c) );
95

96
97
98
99
100
101
102
103
	if( lower && upper ) {
		return lower < upper ? lower : upper;
	} else if ( lower ) {
		return lower;
	} else {
		return upper;
	}
}
104

105
106
107
static int
octetStringMatch(
	int *matchp,
108
	slap_mask_t flags,
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
	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(
128
129
	slap_mask_t use,
	slap_mask_t flags,
130
131
132
133
134
135
136
137
138
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
	int i;
	size_t slen, mlen;
	struct berval **keys;
139
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
140
	unsigned char	HASHdigest[HASH_BYTES];
141
	struct berval digest;
142
143
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
144

145
146
147
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

148
149
150
151
152
153
154
155
156
157
	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++ ) {
158
		HASH_Init( &HASHcontext );
159
		if( prefix != NULL && prefix->bv_len > 0 ) {
160
			HASH_Update( &HASHcontext,
161
162
				prefix->bv_val, prefix->bv_len );
		}
163
		HASH_Update( &HASHcontext,
164
			syntax->ssyn_oid, slen );
165
		HASH_Update( &HASHcontext,
166
			mr->smr_oid, mlen );
167
		HASH_Update( &HASHcontext,
168
			values[i]->bv_val, values[i]->bv_len );
169
		HASH_Final( HASHdigest, &HASHcontext );
170
171
172
173
174
175
176
177
178
179
180
181
182

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

	keys[i] = NULL;

	*keysp = keys;

	return LDAP_SUCCESS;
}

/* Index generation function */
int octetStringFilter(
183
184
	slap_mask_t use,
	slap_mask_t flags,
185
186
187
188
189
190
191
192
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	void * assertValue,
	struct berval ***keysp )
{
	size_t slen, mlen;
	struct berval **keys;
193
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
194
	unsigned char	HASHdigest[HASH_BYTES];
195
196
	struct berval *value = (struct berval *) assertValue;
	struct berval digest;
197
198
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
199
200
201
202
203
204

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

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

205
	HASH_Init( &HASHcontext );
206
	if( prefix != NULL && prefix->bv_len > 0 ) {
207
		HASH_Update( &HASHcontext,
208
209
			prefix->bv_val, prefix->bv_len );
	}
210
	HASH_Update( &HASHcontext,
211
		syntax->ssyn_oid, slen );
212
	HASH_Update( &HASHcontext,
213
		mr->smr_oid, mlen );
214
	HASH_Update( &HASHcontext,
215
		value->bv_val, value->bv_len );
216
	HASH_Final( HASHdigest, &HASHcontext );
217
218
219
220
221
222
223
224

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

	*keysp = keys;

	return LDAP_SUCCESS;
}
225

226
227
228
229
230
231
232
233
234
235
static int
dnValidate(
	Syntax *syntax,
	struct berval *in )
{
	int rc;
	char *dn;

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

236
	dn = get_validated_dn( in->bv_val, 0, 0 );
237

238
239
240
241
	rc = ( dn == NULL ) ? LDAP_INVALID_SYNTAX : LDAP_SUCCESS;

	if ( dn != NULL )
		ch_free( dn );
242
243
244
245

	return rc;
}

246
int
247
248
249
250
251
dnNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
252
	struct berval *out;
253

254
255
256
257
	if ( val->bv_len == 0 ) {
		out = ber_bvdup( val );
	} else {
 		char *dn;
258
#ifdef USE_DN_NORMALIZE
259
		dn = get_validated_dn( val->bv_val, 1, 1 );
260
#else
261
		dn = get_validated_dn( val->bv_val, 0, 0 );
262
263
#endif
		if( dn == NULL ) {
264
265
 			return LDAP_INVALID_SYNTAX;
 		}
266

267
268
269
		out = (struct berval *)ch_malloc(sizeof(struct berval));
 		out->bv_val = dn;
 		out->bv_len = strlen( dn );
270
271
272
273
274
275
276
277
	}

	*normalized = out;
	return LDAP_SUCCESS;
}

static int
dnMatch(
278
	int *matchp,
279
	slap_mask_t flags,
280
281
282
283
284
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
285
	int match;
286
287
	struct berval *asserted = (struct berval *) assertedValue;
	
288
289
290
	match = value->bv_len - asserted->bv_len;

	if( match == 0 ) {
291
#ifdef USE_DN_NORMALIZE
292
		match = strcmp( value->bv_val, asserted->bv_val );
293
#else
294
		match = strcasecmp( value->bv_val, asserted->bv_val );
295
#endif
296
297
	}

298
#ifdef NEW_LOGGING
Gary Williams's avatar
Gary Williams committed
299
300
301
	LDAP_LOG(( "schema", LDAP_LEVEL_ENTRY,
		   "dnMatch: %d\n    %s\n    %s\n", match,
		   value->bv_val, asserted->bv_val ));
302
#else
303
304
	Debug( LDAP_DEBUG_ARGS, "dnMatch %d\n\t\"%s\"\n\t\"%s\"\n",
	    match, value->bv_val, asserted->bv_val );
305
306
#endif

307
308
309

	*matchp = match;
	return LDAP_SUCCESS;
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
404
405
406
407
408
409
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;
}

410
411
412
413
414
415
416
417
418
static int
inValidate(
	Syntax *syntax,
	struct berval *in )
{
	/* any value allowed */
	return LDAP_OTHER;
}

419
static int
420
blobValidate(
421
422
423
424
	Syntax *syntax,
	struct berval *in )
{
	/* any value allowed */
425
	return LDAP_SUCCESS;
426
427
}

428
429
430
431
432
433
434
435
436
437
438
439
440
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;
	}
441

442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
	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;
}

458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
/*
 * 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,
489
	slap_mask_t flags,
490
491
492
493
494
495
496
497
498
499
500
	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;
}

501
502
503
504
505
506
507
508
509
static int
UTF8StringValidate(
	Syntax *syntax,
	struct berval *in )
{
	ber_len_t count;
	int len;
	unsigned char *u = in->bv_val;

510
511
	if( !in->bv_len ) return LDAP_INVALID_SYNTAX;

512
	for( count = in->bv_len; count > 0; count-=len, u+=len ) {
513
514
515
516
		/* get the length indicated by the first byte */
		len = LDAP_UTF8_CHARLEN( u );

		/* should not be zero */
517
		if( len == 0 ) return LDAP_INVALID_SYNTAX;
518
519
520

		/* make sure len corresponds with the offset
			to the next character */
521
		if( LDAP_UTF8_OFFSET( u ) != len ) return LDAP_INVALID_SYNTAX;
522
523
	}

524
	if( count != 0 ) return LDAP_INVALID_SYNTAX;
525

526
	return LDAP_SUCCESS;
527
528
529
530
531
532
533
534
535
}

static int
UTF8StringNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
	struct berval *newval;
536
	char *p, *q, *s;
537

538
	newval = ch_malloc( sizeof( struct berval ) );
539

540
	p = val->bv_val;
541

542
543
544
	/* Ignore initial whitespace */
	while ( ldap_utf8_isspace( p ) ) {
		LDAP_UTF8_INCR( p );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
545
	}
546

547
548
	if( *p == '\0' ) {
		ch_free( newval );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
549
550
		return LDAP_INVALID_SYNTAX;
	}
551

552
553
554
	newval->bv_val = ch_strdup( p );
	p = q = newval->bv_val;
	s = NULL;
555

556
557
	while ( *p ) {
		int len;
558

559
560
561
562
563
		if ( ldap_utf8_isspace( p ) ) {
			len = LDAP_UTF8_COPY(q,p);
			s=q;
			p+=len;
			q+=len;
564

565
566
567
568
			/* Ignore the extra whitespace */
			while ( ldap_utf8_isspace( p ) ) {
				LDAP_UTF8_INCR( p );
			}
Kurt Zeilenga's avatar
Kurt Zeilenga committed
569
		} else {
570
571
572
573
			len = LDAP_UTF8_COPY(q,p);
			s=NULL;
			p+=len;
			q+=len;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
574
		}
575
576
577
578
	}

	assert( *newval->bv_val );
	assert( newval->bv_val < p );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
579
	assert( q <= p );
580

581
582
583
584
585
586
587
588
589
590
591
	/* 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
592
	}
593

594
595
596
597
598
599
600
601
	/* 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;
602

603
	return LDAP_SUCCESS;
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
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
#if UTF8MATCH
/* Returns Unicode cannonically normalized copy of a substring assertion
 * Skipping attribute description */
SubstringsAssertion *
UTF8SubstringsassertionNormalize(
	SubstringsAssertion *sa,
	char casefold )
{
	SubstringsAssertion *nsa;
	int i;

	nsa = (SubstringsAssertion *)ch_calloc( 1, sizeof(SubstringsAssertion) );
	if( nsa == NULL ) {
		return NULL;
	}

	if( sa->sa_initial != NULL ) {
		nsa->sa_initial = ber_bvstr( UTF8normalize( sa->sa_initial->bv_val, casefold ) );
		if( nsa->sa_initial == NULL ) {
			goto err;
		}
	}

	if( sa->sa_any != NULL ) {
		for( i=0; sa->sa_any[i] != NULL; i++ ) {
			/* empty */
		}
		nsa->sa_any = (struct berval **)ch_malloc( (i + 1) * sizeof(struct berval *) );
		for( i=0; sa->sa_any[i] != NULL; i++ ) {
			nsa->sa_any[i] = ber_bvstr( UTF8normalize( sa->sa_any[i]->bv_val, casefold ) );
			if( nsa->sa_any[i] == NULL ) {
				goto err;
			}
		}
		nsa->sa_any[i] = NULL;
	}

	if( sa->sa_final != NULL ) {
		nsa->sa_final = ber_bvstr( UTF8normalize( sa->sa_final->bv_val, casefold ) );
		if( nsa->sa_final == NULL ) {
			goto err;
		}
	}

	return nsa;

err:
	ch_free( nsa->sa_final );
	ber_bvecfree( nsa->sa_any );
	ch_free( nsa->sa_initial );
	ch_free( nsa );
	return NULL;
}
#endif

661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
#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 *) );
Gary Williams's avatar
Gary Williams committed
696
	for( c=val,i=0;	 i<count;  i++,c+=strlen(c)+1 ) {
697
698
699
700
701
702
703
704
705
706
707
		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);
708
709
710
711
		if( len == 0 ) {
			nextchunk++;
			continue;
		}
712
#if defined(SLAPD_APPROX_INITIALS)
713
		else if( len == 1 ) {
714
715
			/* Single letter words need to at least match one word's initial */
			for( i=nextavail; i<count; i++ )
716
717
				if( !strncasecmp( assertv+nextchunk, words[i], 1 )) {
					nextavail=i+1;
718
					break;
719
				}
720
721
		}
#endif
722
		else {
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
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
			/* 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 );
Gary Williams's avatar
Gary Williams committed
786
		for( wordcount=0,c=val;	 *c;  c++) {
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
			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 */
Gary Williams's avatar
Gary Williams committed
802
		for( c=val,i=0;	 i<wordcount;  c+=len+1	 ) {
803
804
805
806
807
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
			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 */
Gary Williams's avatar
Gary Williams committed
850
	for( c=val,i=0;	 i<count; c+=len+1 ) {
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
		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


954
static int
955
caseExactMatch(
956
	int *matchp,
957
	slap_mask_t flags,
958
959
960
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
961
	void *assertedValue )
962
{
963
964
965
	int match = value->bv_len - ((struct berval *) assertedValue)->bv_len;

	if( match == 0 ) {
966
967
968
		match = strncmp( value->bv_val,
			((struct berval *) assertedValue)->bv_val,
			value->bv_len );
969
970
971
	}

	*matchp = match;
972
	return LDAP_SUCCESS;
973
974
}

975
static int
976
caseExactSubstringsMatch(
977
	int *matchp,
978
	slap_mask_t flags,
979
980
981
982
983
984
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
	int match = 0;
985
986
	SubstringsAssertion *sub;
	struct berval left;
987
988
	int i;
	ber_len_t inlen=0;
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
#if UTF8MATCH
	char *nav;
#endif

#if UTF8MATCH
	nav = UTF8normalize( value->bv_val, UTF8_NOCASEFOLD );
	if( nav == NULL ) {
		match = 1;
		goto done;
	}
	left.bv_val = nav;
	left.bv_len = strlen( nav );
	sub = UTF8SubstringsassertionNormalize( assertedValue, UTF8_NOCASEFOLD );
#else
	left = *value;
	sub = assertedValue;
#endif
	if( sub == NULL ) {
		match = -1;
		goto done;
	}
1010

1011
	/* Add up asserted input length */
1012
1013
1014
1015
	if( sub->sa_initial ) {
		inlen += sub->sa_initial->bv_len;
	}
	if( sub->sa_any ) {
1016
1017
		for(i=0; sub->sa_any[i] != NULL; i++) {
			inlen += sub->sa_any[i]->bv_len;
1018
1019
1020
1021
1022
1023
1024
		}
	}
	if( sub->sa_final ) {
		inlen += sub->sa_final->bv_len;
	}

	if( sub->sa_initial ) {
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1025
1026
1027
1028
1029
		if( inlen > left.bv_len ) {
			match = 1;
			goto done;
		}

1030
1031
		match = strncmp( sub->sa_initial->bv_val, left.bv_val,
			sub->sa_initial->bv_len );
1032
1033
1034
1035
1036
1037
1038

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

		left.bv_val += sub->sa_initial->bv_len;
		left.bv_len -= sub->sa_initial->bv_len;
1039
		inlen -= sub->sa_initial->bv_len;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1040
	}
1041

Kurt Zeilenga's avatar
Kurt Zeilenga committed
1042
	if( sub->sa_final ) {
1043
1044
1045
1046
		if( inlen > left.bv_len ) {
			match = 1;
			goto done;
		}
1047

1048
1049
1050
		match = strncmp( sub->sa_final->bv_val,
			&left.bv_val[left.bv_len - sub->sa_final->bv_len],
			sub->sa_final->bv_len );
1051
1052
1053
1054
1055
1056

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

		left.bv_len -= sub->sa_final->bv_len;
1057
		inlen -= sub->sa_final->bv_len;
1058
1059
1060
	}

	if( sub->sa_any ) {
1061
1062
1063
1064
1065
		for(i=0; sub->sa_any[i]; i++) {
			ber_len_t idx;
			char *p;

retry:
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1066
1067
1068
1069
1070
1071
			if( inlen > left.bv_len ) {
				/* not enough length */
				match = 1;
				goto done;
			}

1072
1073
1074
1075
			if( sub->sa_any[i]->bv_len == 0 ) {
				continue;
			}

1076
			p = strchr( left.bv_val, *sub->sa_any[i]->bv_val );
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099

			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;
			}

1100
1101
1102
			match = strncmp( left.bv_val,
				sub->sa_any[i]->bv_val,
				sub->sa_any[i]->bv_len );
1103
1104

			if( match != 0 ) {
1105
1106
				left.bv_val++;
				left.bv_len--;
1107
1108
1109
1110
1111
				goto retry;
			}

			left.bv_val += sub->sa_any[i]->bv_len;
			left.bv_len -= sub->sa_any[i]->bv_len;
1112
			inlen -= sub->sa_any[i]->bv_len;
1113
		}
1114
1115
1116
	}

done:
1117
1118
1119
1120
1121
1122
1123
1124
1125
#if UTF8MATCH
	free( nav );
	if( sub != NULL ) {
		ch_free( sub->sa_final );
		ber_bvecfree( sub->sa_any );
		ch_free( sub->sa_initial );
		ch_free( sub );
	}
#endif
1126
1127
1128
1129
	*matchp = match;
	return LDAP_SUCCESS;
}

1130
/* Index generation function */
1131
int caseExactIndexer(
1132
1133
	slap_mask_t use,
	slap_mask_t flags,
1134
1135
1136
1137
1138
1139
1140
1141
1142
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
	int i;
	size_t slen, mlen;
	struct berval **keys;
1143
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1144
	unsigned char	HASHdigest[HASH_BYTES];
1145
	struct berval digest;
1146
1147
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1148

1149
1150
1151
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
	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++ ) {
1162
1163
1164
1165
1166
1167
		struct berval *value;
#if UTF8MATCH
		value = ber_bvstr( UTF8normalize( values[i]->bv_val, UTF8_NOCASEFOLD ) );
#else
		value = values[i];
#endif
1168

1169
		HASH_Init( &HASHcontext );
1170
		if( prefix != NULL && prefix->bv_len > 0 ) {
1171
			HASH_Update( &HASHcontext,
1172
1173
				prefix->bv_val, prefix->bv_len );
		}
1174
		HASH_Update( &HASHcontext,
1175
			syntax->ssyn_oid, slen );
1176
		HASH_Update( &HASHcontext,
1177
			mr->smr_oid, mlen );
1178
		HASH_Update( &HASHcontext,
1179
			value->bv_val, value->bv_len );
1180
		HASH_Final( HASHdigest, &HASHcontext );
1181

1182
1183
1184
1185
#if UTF8MATCH
		ber_bvfree( value );
#endif

1186
1187
1188
1189
1190
1191
1192
1193
1194
		keys[i] = ber_bvdup( &digest );
	}

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

/* Index generation function */
1195
int caseExactFilter(
1196
1197
	slap_mask_t use,
	slap_mask_t flags,
1198
1199
1200
1201
1202
1203
1204
1205
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	void * assertValue,
	struct berval ***keysp )
{
	size_t slen, mlen;
	struct berval **keys;
1206
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1207
	unsigned char	HASHdigest[HASH_BYTES];
1208
1209
	struct berval *value;
	struct berval digest;
1210
1211
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1212
1213
1214
1215

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

1216
#if UTF8MATCH
1217
1218
1219
1220
1221
1222
1223
1224
	value = ber_bvstr( UTF8normalize( ((struct berval *) assertValue)->bv_val,
		UTF8_NOCASEFOLD ) );
	/* This usually happens if filter contains bad UTF8 */
	if( value == NULL ) {
		keys = ch_malloc( sizeof( struct berval * ) );
		keys[0] = NULL;
		return LDAP_SUCCESS;
	}
1225
#else
1226
	value = (struct berval *) assertValue;
1227
#endif
1228
1229
1230

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

1231
	HASH_Init( &HASHcontext );
1232
	if( prefix != NULL && prefix->bv_len > 0 ) {
1233
		HASH_Update( &HASHcontext,
1234
1235
			prefix->bv_val, prefix->bv_len );
	}
1236
	HASH_Update( &HASHcontext,
1237
		syntax->ssyn_oid, slen );
1238
	HASH_Update( &HASHcontext,
1239
		mr->smr_oid, mlen );
1240
	HASH_Update( &HASHcontext,
1241
		value->bv_val, value->bv_len );
1242
	HASH_Final( HASHdigest, &HASHcontext );
1243
1244
1245
1246

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

1247
1248
1249
1250
#if UTF8MATCH
	ber_bvfree( value );
#endif

1251
1252
1253
	*keysp = keys;
	return LDAP_SUCCESS;
}
1254

1255
/* Substrings Index generation function */
1256
int caseExactSubstringsIndexer(
1257
1258
	slap_mask_t use,
	slap_mask_t flags,
1259
1260
1261
1262
1263
1264
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
1265
	ber_len_t i, nkeys;
1266
1267
	size_t slen, mlen;
	struct berval **keys;
1268
1269
1270
1271
#if UTF8MATCH
        struct berval **nvalues;
#endif

1272
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1273
	unsigned char	HASHdigest[HASH_BYTES];
1274
	struct berval digest;
1275
1276
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1277

1278
1279
1280
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

1281
	nkeys=0;
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295

#if UTF8MATCH
        /* create normalized copy of values */
        for( i=0; values[i] != NULL; i++ ) {
                /* empty */
        }
        nvalues = ch_malloc( sizeof( struct berval * ) * (i+1) );
        for( i=0; values[i] != NULL; i++ ) {
                nvalues[i] = ber_bvstr( UTF8normalize( values[i]->bv_val, UTF8_NOCASEFOLD ) );
        }
        nvalues[i] = NULL;
        values = nvalues;
#endif

1296
1297
	for( i=0; values[i] != NULL; i++ ) {
		/* count number of indices to generate */
1298
		if( values[i]->bv_len < SLAP_INDEX_SUBSTR_MINLEN ) {
1299
1300
1301
			continue;
		}

1302
		if( flags & SLAP_INDEX_SUBSTR_INITIAL ) {
1303
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
1304
				nkeys += SLAP_INDEX_SUBSTR_MAXLEN -
1305
					( SLAP_INDEX_SUBSTR_MINLEN - 1);
1306
			} else {
1307
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MINLEN - 1 );
1308
1309
1310
			}
		}

1311
		if( flags & SLAP_INDEX_SUBSTR_ANY ) {
1312
1313
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MAXLEN - 1 );
1314
1315
1316
			}
		}

1317
		if( flags & SLAP_INDEX_SUBSTR_FINAL ) {
1318
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
1319
				nkeys += SLAP_INDEX_SUBSTR_MAXLEN -
1320
					( SLAP_INDEX_SUBSTR_MINLEN - 1);
1321
			} else {
1322
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MINLEN - 1 );
1323
			}
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
		}
	}

	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++ ) {
1340
		ber_len_t j,max;
1341
1342
		struct berval *value;

1343
1344
		if( values[i]->bv_len < SLAP_INDEX_SUBSTR_MINLEN ) continue;

1345
		value = values[i];
1346

1347
		if( ( flags & SLAP_INDEX_SUBSTR_ANY ) &&
1348
			( value->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) )
1349
1350
		{
			char pre = SLAP_INDEX_SUBSTR_PREFIX;
1351
			max = value->bv_len - ( SLAP_INDEX_SUBSTR_MAXLEN - 1);
1352

1353
			for( j=0; j<max; j++ ) {
1354
				HASH_Init( &HASHcontext );
1355
				if( prefix != NULL && prefix->bv_len > 0 ) {
1356
					HASH_Update( &HASHcontext,
1357
1358
						prefix->bv_val, prefix->bv_len );
				}
1359