schema_init.c 105 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 = ch_strdup( in->bv_val );
237

238
239
	if( dn == NULL ) {
		return LDAP_INVALID_DN;
240

241
242
	} else if ( strlen( in->bv_val ) != in->bv_len ) {
		rc = LDAP_INVALID_DN;
243

244
245
246
247
248
249
250
251
	} else if ( dn_validate( dn ) == NULL ) {
		rc = LDAP_INVALID_DN;

	} else {
		rc = LDAP_SUCCESS;
	}

	ch_free( dn );
252
253
254
	return rc;
}

255
int
256
257
258
259
260
dnNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
261
	struct berval *out;
262

263
264
	if ( val->bv_len != 0 ) {
		char *dn;
265
#ifdef USE_DN_NORMALIZE
266
		out = ber_bvstr( UTF8normalize( val->bv_val, UTF8_CASEFOLD ) );
267
#else
268
269
		out = ber_bvdup( val );
		ldap_pvt_str2upper( out->bv_val );
270
#endif
271
272
		dn = dn_validate( out->bv_val );

273
		if( dn == NULL ) {
274
275
276
			ber_bvfree( out );
			return LDAP_INVALID_SYNTAX;
		}
277

278
279
280
281
		out->bv_val = dn;
		out->bv_len = strlen( dn );
	} else {
		out = ber_bvdup( val );
282
283
284
285
286
287
288
289
	}

	*normalized = out;
	return LDAP_SUCCESS;
}

static int
dnMatch(
290
	int *matchp,
291
	slap_mask_t flags,
292
293
294
295
296
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
297
	int match;
298
299
	struct berval *asserted = (struct berval *) assertedValue;
	
300
301
302
	match = value->bv_len - asserted->bv_len;

	if( match == 0 ) {
303
#ifdef USE_DN_NORMALIZE
304
		match = strcmp( value->bv_val, asserted->bv_val );
305
#else
306
		match = strcasecmp( value->bv_val, asserted->bv_val );
307
#endif
308
309
	}

310
#ifdef NEW_LOGGING
Gary Williams's avatar
Gary Williams committed
311
312
313
	LDAP_LOG(( "schema", LDAP_LEVEL_ENTRY,
		   "dnMatch: %d\n    %s\n    %s\n", match,
		   value->bv_val, asserted->bv_val ));
314
#else
315
316
	Debug( LDAP_DEBUG_ARGS, "dnMatch %d\n\t\"%s\"\n\t\"%s\"\n",
	    match, value->bv_val, asserted->bv_val );
317
318
#endif

319
320
321

	*matchp = match;
	return LDAP_SUCCESS;
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
410
411
412
413
414
415
416
417
418
419
420
421
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;
}

422
423
424
425
426
427
428
429
430
static int
inValidate(
	Syntax *syntax,
	struct berval *in )
{
	/* any value allowed */
	return LDAP_OTHER;
}

431
static int
432
blobValidate(
433
434
435
436
	Syntax *syntax,
	struct berval *in )
{
	/* any value allowed */
437
	return LDAP_SUCCESS;
438
439
}

440
441
442
443
444
445
446
447
448
449
450
451
452
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;
	}
453

454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
	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;
}

470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
/*
 * 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,
501
	slap_mask_t flags,
502
503
504
505
506
507
508
509
510
511
512
	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;
}

513
514
515
516
517
518
519
520
521
static int
UTF8StringValidate(
	Syntax *syntax,
	struct berval *in )
{
	ber_len_t count;
	int len;
	unsigned char *u = in->bv_val;

522
523
	if( !in->bv_len ) return LDAP_INVALID_SYNTAX;

524
	for( count = in->bv_len; count > 0; count-=len, u+=len ) {
525
526
527
528
		/* get the length indicated by the first byte */
		len = LDAP_UTF8_CHARLEN( u );

		/* should not be zero */
529
		if( len == 0 ) return LDAP_INVALID_SYNTAX;
530
531
532

		/* make sure len corresponds with the offset
			to the next character */
533
		if( LDAP_UTF8_OFFSET( u ) != len ) return LDAP_INVALID_SYNTAX;
534
535
	}

536
	if( count != 0 ) return LDAP_INVALID_SYNTAX;
537

538
	return LDAP_SUCCESS;
539
540
541
542
543
544
545
546
547
}

static int
UTF8StringNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
	struct berval *newval;
548
	char *p, *q, *s;
549

550
	newval = ch_malloc( sizeof( struct berval ) );
551

552
	p = val->bv_val;
553

554
555
556
	/* Ignore initial whitespace */
	while ( ldap_utf8_isspace( p ) ) {
		LDAP_UTF8_INCR( p );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
557
	}
558

559
560
	if( *p == '\0' ) {
		ch_free( newval );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
561
562
		return LDAP_INVALID_SYNTAX;
	}
563

564
565
566
	newval->bv_val = ch_strdup( p );
	p = q = newval->bv_val;
	s = NULL;
567

568
569
	while ( *p ) {
		int len;
570

571
572
573
574
575
		if ( ldap_utf8_isspace( p ) ) {
			len = LDAP_UTF8_COPY(q,p);
			s=q;
			p+=len;
			q+=len;
576

577
578
579
580
			/* Ignore the extra whitespace */
			while ( ldap_utf8_isspace( p ) ) {
				LDAP_UTF8_INCR( p );
			}
Kurt Zeilenga's avatar
Kurt Zeilenga committed
581
		} else {
582
583
584
585
			len = LDAP_UTF8_COPY(q,p);
			s=NULL;
			p+=len;
			q+=len;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
586
		}
587
588
589
590
	}

	assert( *newval->bv_val );
	assert( newval->bv_val < p );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
591
	assert( q <= p );
592

593
594
595
596
597
598
599
600
601
602
603
	/* 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
604
	}
605

606
607
608
609
610
611
612
613
	/* 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;
614

615
	return LDAP_SUCCESS;
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
661
662
663
664
665
666
667
668
669
670
671
672
#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

673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
#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
708
	for( c=val,i=0;	 i<count;  i++,c+=strlen(c)+1 ) {
709
710
711
712
713
714
715
716
717
718
719
		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);
720
721
722
723
		if( len == 0 ) {
			nextchunk++;
			continue;
		}
724
#if defined(SLAPD_APPROX_INITIALS)
725
		else if( len == 1 ) {
726
727
			/* Single letter words need to at least match one word's initial */
			for( i=nextavail; i<count; i++ )
728
729
				if( !strncasecmp( assertv+nextchunk, words[i], 1 )) {
					nextavail=i+1;
730
					break;
731
				}
732
733
		}
#endif
734
		else {
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
786
787
788
789
790
791
792
793
794
795
796
797
			/* 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
798
		for( wordcount=0,c=val;	 *c;  c++) {
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
			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
814
		for( c=val,i=0;	 i<wordcount;  c+=len+1	 ) {
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
			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
862
	for( c=val,i=0;	 i<count; c+=len+1 ) {
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
		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


966
static int
967
caseExactMatch(
968
	int *matchp,
969
	slap_mask_t flags,
970
971
972
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
973
	void *assertedValue )
974
{
975
976
977
978
979
#if UTF8MATCH
	*matchp = UTF8normcmp( value->bv_val,
		((struct berval *) assertedValue)->bv_val,
		UTF8_NOCASEFOLD );
#else
980
981
982
	int match = value->bv_len - ((struct berval *) assertedValue)->bv_len;

	if( match == 0 ) {
983
984
985
		match = strncmp( value->bv_val,
			((struct berval *) assertedValue)->bv_val,
			value->bv_len );
986
987
988
	}

	*matchp = match;
989
#endif
990
	return LDAP_SUCCESS;
991
992
}

993
static int
994
caseExactSubstringsMatch(
995
	int *matchp,
996
	slap_mask_t flags,
997
998
999
1000
1001
1002
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
	int match = 0;
1003
1004
	SubstringsAssertion *sub;
	struct berval left;
1005
1006
	int i;
	ber_len_t inlen=0;
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
#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;
	}
1028

1029
	/* Add up asserted input length */
1030
1031
1032
1033
	if( sub->sa_initial ) {
		inlen += sub->sa_initial->bv_len;
	}
	if( sub->sa_any ) {
1034
1035
		for(i=0; sub->sa_any[i] != NULL; i++) {
			inlen += sub->sa_any[i]->bv_len;
1036
1037
1038
1039
1040
1041
1042
		}
	}
	if( sub->sa_final ) {
		inlen += sub->sa_final->bv_len;
	}

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

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

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

		left.bv_val += sub->sa_initial->bv_len;
		left.bv_len -= sub->sa_initial->bv_len;
1057
		inlen -= sub->sa_initial->bv_len;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1058
	}
1059

Kurt Zeilenga's avatar
Kurt Zeilenga committed
1060
	if( sub->sa_final ) {
1061
1062
1063
1064
		if( inlen > left.bv_len ) {
			match = 1;
			goto done;
		}
1065

1066
1067
1068
		match = strncmp( sub->sa_final->bv_val,
			&left.bv_val[left.bv_len - sub->sa_final->bv_len],
			sub->sa_final->bv_len );
1069
1070
1071
1072
1073
1074

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

		left.bv_len -= sub->sa_final->bv_len;
1075
		inlen -= sub->sa_final->bv_len;
1076
1077
1078
	}

	if( sub->sa_any ) {
1079
1080
1081
1082
1083
		for(i=0; sub->sa_any[i]; i++) {
			ber_len_t idx;
			char *p;

retry:
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1084
1085
1086
1087
1088
1089
			if( inlen > left.bv_len ) {
				/* not enough length */
				match = 1;
				goto done;
			}

1090
1091
1092
1093
			if( sub->sa_any[i]->bv_len == 0 ) {
				continue;
			}

1094
			p = strchr( left.bv_val, *sub->sa_any[i]->bv_val );
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117

			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;
			}

1118
1119
1120
			match = strncmp( left.bv_val,
				sub->sa_any[i]->bv_val,
				sub->sa_any[i]->bv_len );
1121
1122

			if( match != 0 ) {
1123
1124
				left.bv_val++;
				left.bv_len--;
1125
1126
1127
1128
1129
				goto retry;
			}

			left.bv_val += sub->sa_any[i]->bv_len;
			left.bv_len -= sub->sa_any[i]->bv_len;
1130
			inlen -= sub->sa_any[i]->bv_len;
1131
		}
1132
1133
1134
	}

done:
1135
1136
1137
1138
1139
1140
1141
1142
1143
#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
1144
1145
1146
1147
	*matchp = match;
	return LDAP_SUCCESS;
}

1148
/* Index generation function */
1149
int caseExactIndexer(
1150
1151
	slap_mask_t use,
	slap_mask_t flags,
1152
1153
1154
1155
1156
1157
1158
1159
1160
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
	int i;
	size_t slen, mlen;
	struct berval **keys;
1161
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1162
	unsigned char	HASHdigest[HASH_BYTES];
1163
	struct berval digest;
1164
1165
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1166

1167
1168
1169
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
	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++ ) {
1180
1181
1182
1183
1184
1185
		struct berval *value;
#if UTF8MATCH
		value = ber_bvstr( UTF8normalize( values[i]->bv_val, UTF8_NOCASEFOLD ) );
#else
		value = values[i];
#endif
1186

1187
		HASH_Init( &HASHcontext );
1188
		if( prefix != NULL && prefix->bv_len > 0 ) {
1189
			HASH_Update( &HASHcontext,
1190
1191
				prefix->bv_val, prefix->bv_len );
		}
1192
		HASH_Update( &HASHcontext,
1193
			syntax->ssyn_oid, slen );
1194
		HASH_Update( &HASHcontext,
1195
			mr->smr_oid, mlen );
1196
		HASH_Update( &HASHcontext,
1197
			value->bv_val, value->bv_len );
1198
		HASH_Final( HASHdigest, &HASHcontext );
1199

1200
1201
1202
1203
#if UTF8MATCH
		ber_bvfree( value );
#endif

1204
1205
1206
1207
1208
1209
1210
1211
1212
		keys[i] = ber_bvdup( &digest );
	}

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

/* Index generation function */
1213
int caseExactFilter(
1214
1215
	slap_mask_t use,
	slap_mask_t flags,
1216
1217
1218
1219
1220
1221
1222
1223
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	void * assertValue,
	struct berval ***keysp )
{
	size_t slen, mlen;
	struct berval **keys;
1224
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1225
	unsigned char	HASHdigest[HASH_BYTES];
1226
1227
	struct berval *value;
	struct berval digest;
1228
1229
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1230
1231
1232
1233

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

1234
#if UTF8MATCH
1235
1236
1237
1238
1239
1240
1241
1242
	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;
	}
1243
#else
1244
	value = (struct berval *) assertValue;
1245
#endif
1246
1247
1248

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

1249
	HASH_Init( &HASHcontext );
1250
	if( prefix != NULL && prefix->bv_len > 0 ) {
1251
		HASH_Update( &HASHcontext,
1252
1253
			prefix->bv_val, prefix->bv_len );
	}
1254
	HASH_Update( &HASHcontext,
1255
		syntax->ssyn_oid, slen );
1256
	HASH_Update( &HASHcontext,
1257
		mr->smr_oid, mlen );
1258
	HASH_Update( &HASHcontext,
1259
		value->bv_val, value->bv_len );
1260
	HASH_Final( HASHdigest, &HASHcontext );
1261
1262
1263
1264

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

1265
1266
1267
1268
#if UTF8MATCH
	ber_bvfree( value );
#endif

1269
1270
1271
	*keysp = keys;
	return LDAP_SUCCESS;
}
1272

1273
/* Substrings Index generation function */
1274
int caseExactSubstringsIndexer(
1275
1276
	slap_mask_t use,
	slap_mask_t flags,
1277
1278
1279
1280
1281
1282
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
1283
	ber_len_t i, nkeys;
1284
1285
	size_t slen, mlen;
	struct berval **keys;
1286
1287
1288
1289
#if UTF8MATCH
        struct berval **nvalues;
#endif

1290
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1291
	unsigned char	HASHdigest[HASH_BYTES];
1292
	struct berval digest;
1293
1294
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1295

1296
1297
1298
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

1299
	nkeys=0;
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313

#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

1314
1315
	for( i=0; values[i] != NULL; i++ ) {
		/* count number of indices to generate */
1316
		if( values[i]->bv_len < SLAP_INDEX_SUBSTR_MINLEN ) {
1317
1318
1319
			continue;
		}

1320
		if( flags & SLAP_INDEX_SUBSTR_INITIAL ) {
1321
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
1322
				nkeys += SLAP_INDEX_SUBSTR_MAXLEN -
1323
					( SLAP_INDEX_SUBSTR_MINLEN - 1);
1324
			} else {
1325
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MINLEN - 1 );
1326
1327
1328
			}
		}

1329
		if( flags & SLAP_INDEX_SUBSTR_ANY ) {
1330
1331
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MAXLEN - 1 );
1332
1333
1334
			}
		}

1335
		if( flags & SLAP_INDEX_SUBSTR_FINAL ) {
1336
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
1337
				nkeys += SLAP_INDEX_SUBSTR_MAXLEN -
1338
					( SLAP_INDEX_SUBSTR_MINLEN - 1);
1339
			} else {
1340
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MINLEN - 1 );
1341
			}
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
		}
	}

	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++ ) {
1358
		ber_len_t j,max;
1359
1360
		struct berval *value;

1361
1362
		if( values[i]->bv_len < SLAP_INDEX_SUBSTR_MINLEN ) continue;

1363
		value = values[i];
1364

1365
		if( ( flags & SLAP_INDEX_SUBSTR_ANY ) &&
1366
			( value->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) )
1367
1368
		{
			char pre = SLAP_INDEX_SUBSTR_PREFIX;