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				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
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
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
136
	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
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
190
	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
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;
}

241
int
242
243
244
245
246
dnNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
247
	struct berval *out;
248

249
	if ( val->bv_len != 0 ) {
250
251
		char *dn;
#ifdef USE_DN_NORMALIZE
252
		out = ber_bvstr( UTF8normalize( val->bv_val, UTF8_CASEFOLD ) );
253
#else
254
255
		out = ber_bvdup( val );
		ldap_pvt_str2upper( out->bv_val );
256
#endif
257
		dn = dn_validate( out->bv_val );
258
259
260
261
262
263
264
265

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

		out->bv_val = dn;
		out->bv_len = strlen( dn );
266
267
	} else {
		out = ber_bvdup( val );
268
269
270
271
272
273
274
275
	}

	*normalized = out;
	return LDAP_SUCCESS;
}

static int
dnMatch(
276
	int *matchp,
277
	slap_mask_t flags,
278
279
280
281
282
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
	void *assertedValue )
{
283
	int match;
284
285
	struct berval *asserted = (struct berval *) assertedValue;
	
286
287
288
	match = value->bv_len - asserted->bv_len;

	if( match == 0 ) {
289
#ifdef USE_DN_NORMALIZE
290
		match = strcmp( value->bv_val, asserted->bv_val );
291
#else
292
		match = strcasecmp( value->bv_val, asserted->bv_val );
293
#endif
294
295
	}

296
#ifdef NEW_LOGGING
Gary Williams's avatar
Gary Williams committed
297
298
299
	LDAP_LOG(( "schema", LDAP_LEVEL_ENTRY,
		   "dnMatch: %d\n    %s\n    %s\n", match,
		   value->bv_val, asserted->bv_val ));
300
#else
301
302
	Debug( LDAP_DEBUG_ARGS, "dnMatch %d\n\t\"%s\"\n\t\"%s\"\n",
	    match, value->bv_val, asserted->bv_val );
303
304
#endif

305
306
307

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

408
409
410
411
412
413
414
415
416
static int
inValidate(
	Syntax *syntax,
	struct berval *in )
{
	/* any value allowed */
	return LDAP_OTHER;
}

417
static int
418
blobValidate(
419
420
421
422
	Syntax *syntax,
	struct berval *in )
{
	/* any value allowed */
423
	return LDAP_SUCCESS;
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
451
452
453
454
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;
}

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
482
483
484
485
/*
 * 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,
486
	slap_mask_t flags,
487
488
489
490
491
492
493
494
495
496
497
	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;
}

498
#if UTF8MATCH
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
548
549
550
551
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
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
639
640
641
642

/* 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;
	}
}
643
644
#endif

645
646
647
648
649
650
651
652
653
static int
UTF8StringValidate(
	Syntax *syntax,
	struct berval *in )
{
	ber_len_t count;
	int len;
	unsigned char *u = in->bv_val;

654
655
	if( !in->bv_len ) return LDAP_INVALID_SYNTAX;

656
	for( count = in->bv_len; count > 0; count-=len, u+=len ) {
657
658
659
660
		/* get the length indicated by the first byte */
		len = LDAP_UTF8_CHARLEN( u );

		/* should not be zero */
661
		if( len == 0 ) return LDAP_INVALID_SYNTAX;
662
663
664

		/* make sure len corresponds with the offset
			to the next character */
665
		if( LDAP_UTF8_OFFSET( u ) != len ) return LDAP_INVALID_SYNTAX;
666
667
	}

668
	if( count != 0 ) return LDAP_INVALID_SYNTAX;
669

670
	return LDAP_SUCCESS;
671
672
673
674
675
676
677
678
679
}

static int
UTF8StringNormalize(
	Syntax *syntax,
	struct berval *val,
	struct berval **normalized )
{
	struct berval *newval;
680
	char *p, *q, *s;
681

682
	newval = ch_malloc( sizeof( struct berval ) );
683

684
	p = val->bv_val;
685

686
687
688
	/* Ignore initial whitespace */
	while ( ldap_utf8_isspace( p ) ) {
		LDAP_UTF8_INCR( p );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
689
	}
690

691
692
	if( *p == '\0' ) {
		ch_free( newval );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
693
694
		return LDAP_INVALID_SYNTAX;
	}
695

696
697
698
	newval->bv_val = ch_strdup( p );
	p = q = newval->bv_val;
	s = NULL;
699

700
701
	while ( *p ) {
		int len;
702

703
704
705
706
707
		if ( ldap_utf8_isspace( p ) ) {
			len = LDAP_UTF8_COPY(q,p);
			s=q;
			p+=len;
			q+=len;
708

709
710
711
712
			/* Ignore the extra whitespace */
			while ( ldap_utf8_isspace( p ) ) {
				LDAP_UTF8_INCR( p );
			}
Kurt Zeilenga's avatar
Kurt Zeilenga committed
713
		} else {
714
715
716
717
			len = LDAP_UTF8_COPY(q,p);
			s=NULL;
			p+=len;
			q+=len;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
718
		}
719
720
721
722
	}

	assert( *newval->bv_val );
	assert( newval->bv_val < p );
Kurt Zeilenga's avatar
Kurt Zeilenga committed
723
	assert( q <= p );
724

725
726
727
728
729
730
731
732
733
734
735
	/* 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
736
	}
737

738
739
740
741
742
743
744
745
	/* 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;
746

747
	return LDAP_SUCCESS;
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
#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
785
	for( c=val,i=0;	 i<count;  i++,c+=strlen(c)+1 ) {
786
787
788
789
790
791
792
793
794
795
796
		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);
797
798
799
800
		if( len == 0 ) {
			nextchunk++;
			continue;
		}
801
#if defined(SLAPD_APPROX_INITIALS)
802
		else if( len == 1 ) {
803
804
			/* Single letter words need to at least match one word's initial */
			for( i=nextavail; i<count; i++ )
805
806
				if( !strncasecmp( assertv+nextchunk, words[i], 1 )) {
					nextavail=i+1;
807
					break;
808
				}
809
810
		}
#endif
811
		else {
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
			/* 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
875
		for( wordcount=0,c=val;	 *c;  c++) {
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
			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
891
		for( c=val,i=0;	 i<wordcount;  c+=len+1	 ) {
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
			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
939
	for( c=val,i=0;	 i<count; c+=len+1 ) {
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
1039
1040
1041
1042
		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


1043
static int
1044
caseExactMatch(
1045
	int *matchp,
1046
	slap_mask_t flags,
1047
1048
1049
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *value,
1050
	void *assertedValue )
1051
{
1052
1053
1054
	int match = value->bv_len - ((struct berval *) assertedValue)->bv_len;

	if( match == 0 ) {
1055
1056
1057
		match = strncmp( value->bv_val,
			((struct berval *) assertedValue)->bv_val,
			value->bv_len );
1058
1059
1060
	}

	*matchp = match;
1061
	return LDAP_SUCCESS;
1062
1063
}

1064
static int
1065
caseExactSubstringsMatch(
1066
	int *matchp,
1067
	slap_mask_t flags,
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
	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;

1079
	/* Add up asserted input length */
1080
1081
1082
1083
	if( sub->sa_initial ) {
		inlen += sub->sa_initial->bv_len;
	}
	if( sub->sa_any ) {
1084
1085
		for(i=0; sub->sa_any[i] != NULL; i++) {
			inlen += sub->sa_any[i]->bv_len;
1086
1087
1088
1089
1090
1091
1092
		}
	}
	if( sub->sa_final ) {
		inlen += sub->sa_final->bv_len;
	}

	if( sub->sa_initial ) {
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1093
1094
1095
1096
1097
		if( inlen > left.bv_len ) {
			match = 1;
			goto done;
		}

1098
1099
		match = strncmp( sub->sa_initial->bv_val, left.bv_val,
			sub->sa_initial->bv_len );
1100
1101
1102
1103
1104
1105
1106

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

		left.bv_val += sub->sa_initial->bv_len;
		left.bv_len -= sub->sa_initial->bv_len;
1107
		inlen -= sub->sa_initial->bv_len;
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1108
	}
1109

Kurt Zeilenga's avatar
Kurt Zeilenga committed
1110
	if( sub->sa_final ) {
1111
1112
1113
1114
		if( inlen > left.bv_len ) {
			match = 1;
			goto done;
		}
1115

1116
1117
1118
		match = strncmp( sub->sa_final->bv_val,
			&left.bv_val[left.bv_len - sub->sa_final->bv_len],
			sub->sa_final->bv_len );
1119
1120
1121
1122
1123
1124

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

		left.bv_len -= sub->sa_final->bv_len;
1125
		inlen -= sub->sa_final->bv_len;
1126
1127
1128
	}

	if( sub->sa_any ) {
1129
1130
1131
1132
1133
		for(i=0; sub->sa_any[i]; i++) {
			ber_len_t idx;
			char *p;

retry:
Kurt Zeilenga's avatar
Kurt Zeilenga committed
1134
1135
1136
1137
1138
1139
			if( inlen > left.bv_len ) {
				/* not enough length */
				match = 1;
				goto done;
			}

1140
1141
1142
1143
			if( sub->sa_any[i]->bv_len == 0 ) {
				continue;
			}

1144
			p = strchr( left.bv_val, *sub->sa_any[i]->bv_val );
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167

			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;
			}

1168
1169
1170
			match = strncmp( left.bv_val,
				sub->sa_any[i]->bv_val,
				sub->sa_any[i]->bv_len );
1171
1172

			if( match != 0 ) {
1173
1174
				left.bv_val++;
				left.bv_len--;
1175
1176
1177
1178
1179
				goto retry;
			}

			left.bv_val += sub->sa_any[i]->bv_len;
			left.bv_len -= sub->sa_any[i]->bv_len;
1180
			inlen -= sub->sa_any[i]->bv_len;
1181
		}
1182
1183
1184
1185
1186
1187
1188
	}

done:
	*matchp = match;
	return LDAP_SUCCESS;
}

1189
/* Index generation function */
1190
int caseExactIndexer(
1191
1192
	slap_mask_t use,
	slap_mask_t flags,
1193
1194
1195
1196
1197
1198
1199
1200
1201
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
	int i;
	size_t slen, mlen;
	struct berval **keys;
1202
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1203
	unsigned char	HASHdigest[HASH_BYTES];
1204
	struct berval digest;
1205
1206
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1207

1208
1209
1210
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
	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];

1223
		HASH_Init( &HASHcontext );
1224
		if( prefix != NULL && prefix->bv_len > 0 ) {
1225
			HASH_Update( &HASHcontext,
1226
1227
				prefix->bv_val, prefix->bv_len );
		}
1228
		HASH_Update( &HASHcontext,
1229
			syntax->ssyn_oid, slen );
1230
		HASH_Update( &HASHcontext,
1231
			mr->smr_oid, mlen );
1232
		HASH_Update( &HASHcontext,
1233
			value->bv_val, value->bv_len );
1234
		HASH_Final( HASHdigest, &HASHcontext );
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244

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

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

/* Index generation function */
1245
int caseExactFilter(
1246
1247
	slap_mask_t use,
	slap_mask_t flags,
1248
1249
1250
1251
1252
1253
1254
1255
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	void * assertValue,
	struct berval ***keysp )
{
	size_t slen, mlen;
	struct berval **keys;
1256
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1257
	unsigned char	HASHdigest[HASH_BYTES];
1258
1259
	struct berval *value;
	struct berval digest;
1260
1261
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1262
1263
1264
1265
1266
1267
1268
1269

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

	value = (struct berval *) assertValue;

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

1270
	HASH_Init( &HASHcontext );
1271
	if( prefix != NULL && prefix->bv_len > 0 ) {
1272
		HASH_Update( &HASHcontext,
1273
1274
			prefix->bv_val, prefix->bv_len );
	}
1275
	HASH_Update( &HASHcontext,
1276
		syntax->ssyn_oid, slen );
1277
	HASH_Update( &HASHcontext,
1278
		mr->smr_oid, mlen );
1279
	HASH_Update( &HASHcontext,
1280
		value->bv_val, value->bv_len );
1281
	HASH_Final( HASHdigest, &HASHcontext );
1282
1283
1284
1285
1286
1287
1288

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

	*keysp = keys;
	return LDAP_SUCCESS;
}
1289

1290
/* Substrings Index generation function */
1291
int caseExactSubstringsIndexer(
1292
1293
	slap_mask_t use,
	slap_mask_t flags,
1294
1295
1296
1297
1298
1299
	Syntax *syntax,
	MatchingRule *mr,
	struct berval *prefix,
	struct berval **values,
	struct berval ***keysp )
{
1300
	ber_len_t i, nkeys;
1301
1302
	size_t slen, mlen;
	struct berval **keys;
1303
	HASH_CONTEXT   HASHcontext;
Gary Williams's avatar
Gary Williams committed
1304
	unsigned char	HASHdigest[HASH_BYTES];
1305
	struct berval digest;
1306
1307
	digest.bv_val = HASHdigest;
	digest.bv_len = sizeof(HASHdigest);
1308

1309
1310
1311
	/* we should have at least one value at this point */
	assert( values != NULL && values[0] != NULL );

1312
1313
1314
	nkeys=0;
	for( i=0; values[i] != NULL; i++ ) {
		/* count number of indices to generate */
1315
		if( values[i]->bv_len < SLAP_INDEX_SUBSTR_MINLEN ) {
1316
1317
1318
			continue;
		}

1319
		if( flags & SLAP_INDEX_SUBSTR_INITIAL ) {
1320
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
1321
				nkeys += SLAP_INDEX_SUBSTR_MAXLEN -
1322
					( SLAP_INDEX_SUBSTR_MINLEN - 1);
1323
			} else {
1324
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MINLEN - 1 );
1325
1326
1327
			}
		}

1328
		if( flags & SLAP_INDEX_SUBSTR_ANY ) {
1329
1330
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MAXLEN - 1 );
1331
1332
1333
			}
		}

1334
		if( flags & SLAP_INDEX_SUBSTR_FINAL ) {
1335
			if( values[i]->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) {
1336
				nkeys += SLAP_INDEX_SUBSTR_MAXLEN -
1337
					( SLAP_INDEX_SUBSTR_MINLEN - 1);
1338
			} else {
1339
				nkeys += values[i]->bv_len - ( SLAP_INDEX_SUBSTR_MINLEN - 1 );
1340
			}
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
		}
	}

	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++ ) {
1357
		ber_len_t j,max;
1358
1359
		struct berval *value;

1360
		value = values[i];
1361
		if( value->bv_len < SLAP_INDEX_SUBSTR_MINLEN ) continue;
1362

1363
		if( ( flags & SLAP_INDEX_SUBSTR_ANY ) &&
1364
			( value->bv_len >= SLAP_INDEX_SUBSTR_MAXLEN ) )
1365
1366
		{
			char pre = SLAP_INDEX_SUBSTR_PREFIX;
1367
			max = value->bv_len - ( SLAP_INDEX_SUBSTR_MAXLEN - 1);
1368

1369
			for( j=0; j<max; j++ ) {
1370
				HASH_Init( &HASHcontext );
1371
				if( prefix != NULL && prefix->bv_len > 0 ) {
1372
					HASH_Update( &HASHcontext,
1373
1374
						prefix->bv_val, prefix->bv_len );
				}
1375

1376
				HASH_Update( &HASHcontext,
1377
					&pre, sizeof( pre ) );
1378
				HASH_Update( &HASHcontext,
1379
					syntax->ssyn_oid, slen );
1380
				HASH_Update( &HASHcontext,
1381
					mr->smr_oid, mlen );
1382
				HASH_Update( &HASHcontext,
1383
1384
					&value->bv_val[j],
					SLAP_INDEX_SUBSTR_MAXLEN );
1385
				HASH_Final( HASHdigest, &HASHcontext );
1386
1387
1388
1389
1390

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

1391
1392
		max = SLAP_INDEX_SUBSTR_MAXLEN < value->bv_len
			? SLAP_INDEX_SUBSTR_MAXLEN : value->bv_len;
1393

1394
		for( j=SLAP_INDEX_SUBSTR_MINLEN; j<=max; j++ ) {