{
int i, j, numvals = 0, nummods,
rc = LDAP_SUCCESS, matched;
- BerVarray nvals = NULL, nmods = NULL;
-
- /*
- * FIXME: better do the following
- *
- * - count the existing values
- * - count the new values
- *
- * - if the existing values are less than the new ones {
- * - normalize all the existing values
- * - for each new value {
- * - normalize
- * - check with existing
- * - cross-check with already normalized new vals
- * }
- * } else {
- * - for each new value {
- * - normalize
- * - cross-check with already normalized new vals
- * }
- * - for each existing value {
- * - normalize
- * - check with already normalized new values
- * }
- * }
- *
- * The first case is good when adding a lot of new values,
- * and significantly at first import of values (e.g. adding
- * a new group); the latter case seems to be quite important
- * as well, because it is likely to be the most frequently
- * used when administering the entry. The current
- * implementation will always normalize all the existing
- * values before checking. If there's no duplicate, the
- * performances should not change; they will in case of error.
- */
-
- for ( nummods = 0; mods[ nummods ].bv_val != NULL; nummods++ )
- /* count new values */ ;
-
- if ( vals ) {
- for ( numvals = 0; vals[ numvals ].bv_val != NULL; numvals++ )
- /* count existing values */ ;
-
- if ( numvals < nummods ) {
- nvals = SLAP_CALLOC( numvals + 1, sizeof( struct berval ) );
- if( nvals == NULL ) {
-#ifdef NEW_LOGGING
- LDAP_LOG( OPERATION, ERR,
- "modify_check_duplicates: SLAP_CALLOC failed", 0, 0, 0 );
-#else
- Debug( LDAP_DEBUG_ANY,
- "modify_check_duplicates: SLAP_CALLOC failed", 0, 0, 0 );
-#endif
- goto return_results;
- }
-
- /* normalize the existing values first */
- for ( j = 0; vals[ j ].bv_val != NULL; j++ ) {
- rc = value_normalize( ad, SLAP_MR_EQUALITY,
- &vals[ j ], &nvals[ j ], text );
-
- /* existing attribute values must normalize */
- assert( rc == LDAP_SUCCESS );
-
- if ( rc != LDAP_SUCCESS ) {
- nvals[ j ].bv_val = NULL;
- goto return_results;
- }
- }
- nvals[ j ].bv_val = NULL;
- }
- }
-
- /*
- * If the existing values are less than the new values,
- * it is more convenient to normalize all the existing
- * values and test each new value against them first,
- * then to other already normalized values
- */
- nmods = SLAP_CALLOC( nummods + 1, sizeof( struct berval ) );
- if ( nmods == NULL ) {
-#ifdef NEW_LOGGING
- LDAP_LOG( OPERATION, ERR,
- "modify_check_duplicates: SLAP_CALLOC failed", 0, 0, 0 );
-#else
- Debug( LDAP_DEBUG_ANY,
- "modify_check_duplicates: SLAP_CALLOC failed", 0, 0, 0 );
-#endif
- goto return_results;
- }
-
- for ( i = 0; mods[ i ].bv_val != NULL; i++ ) {
- rc = value_normalize( ad, SLAP_MR_EQUALITY,
- &mods[ i ], &nmods[ i ], text );
-
- if ( rc != LDAP_SUCCESS ) {
- nmods[ i ].bv_val = NULL;
- goto return_results;
- }
-
- if ( numvals > 0 && numvals < nummods ) {
- for ( matched = 0, j = 0; nvals[ j ].bv_val; j++ ) {
- int match;
-
- rc = (*mr->smr_match)( &match,
- SLAP_MR_ATTRIBUTE_SYNTAX_MATCH,
- ad->ad_type->sat_syntax,
- mr, &nmods[ i ], &nvals[ j ] );
- if ( rc != LDAP_SUCCESS ) {
- nmods[ i + 1 ].bv_val = NULL;
- *text = textbuf;
- snprintf( textbuf, textlen,
- "%s: matching rule failed",
- ad->ad_cname.bv_val );
- goto return_results;
- }
-
- if ( match == 0 ) {
- if ( permissive ) {
- matched++;
- continue;
- }
- *text = textbuf;
- snprintf( textbuf, textlen,
- "%s: value #%d provided more than once",
- ad->ad_cname.bv_val, i );
- rc = LDAP_TYPE_OR_VALUE_EXISTS;
- nmods[ i + 1 ].bv_val = NULL;
- goto return_results;
- }
- }
-
- if ( permissive && matched == j ) {
- nmods[ i + 1 ].bv_val = NULL;
- rc = LDAP_TYPE_OR_VALUE_EXISTS;
- goto return_results;
- }
- }
-
- for ( matched = 0, j = 0; j < i; j++ ) {
- int match;
-
- rc = (*mr->smr_match)( &match,
- SLAP_MR_ATTRIBUTE_SYNTAX_MATCH,
- ad->ad_type->sat_syntax,
- mr, &nmods[ i ], &nmods[ j ] );
- if ( rc != LDAP_SUCCESS ) {
- nmods[ i + 1 ].bv_val = NULL;
- *text = textbuf;
- snprintf( textbuf, textlen,
- "%s: matching rule failed",
- ad->ad_cname.bv_val );
- goto return_results;
- }
-
- if ( match == 0 ) {
- if ( permissive ) {
- matched++;
- continue;
- }
- *text = textbuf;
- snprintf( textbuf, textlen,
- "%s: value #%d provided more than once",
- ad->ad_cname.bv_val, j );
- rc = LDAP_TYPE_OR_VALUE_EXISTS;
- nmods[ i + 1 ].bv_val = NULL;
- goto return_results;
- }
- }
-
- if ( permissive && matched == j ) {
- nmods[ i + 1 ].bv_val = NULL;
- rc = LDAP_TYPE_OR_VALUE_EXISTS;
- goto return_results;
- }
- }
- nmods[ i ].bv_val = NULL;
-
- /*
- * if new values are more than existing values, it is more
- * convenient to normalize and check all new values first,
- * then check each new value against existing values, which
- * can be normalized in place
- */
-
- if ( numvals >= nummods ) {
- for ( j = 0; vals[ j ].bv_val; j++ ) {
- struct berval asserted;
-
- rc = value_normalize( ad, SLAP_MR_EQUALITY,
- &vals[ j ], &asserted, text );
-
- if ( rc != LDAP_SUCCESS ) {
- goto return_results;
- }
-
- for ( matched = 0, i = 0; nmods[ i ].bv_val; i++ ) {
- int match;
-
- rc = (*mr->smr_match)( &match,
- SLAP_MR_ATTRIBUTE_SYNTAX_MATCH,
- ad->ad_type->sat_syntax,
- mr, &nmods[ i ], &asserted );
- if ( rc != LDAP_SUCCESS ) {
- *text = textbuf;
- snprintf( textbuf, textlen,
- "%s: matching rule failed",
- ad->ad_cname.bv_val );
- goto return_results;
- }
-
- if ( match == 0 ) {
- if ( permissive ) {
- matched++;
- continue;
- }
- *text = textbuf;
- snprintf( textbuf, textlen,
- "%s: value #%d provided more than once",
- ad->ad_cname.bv_val, j );
- rc = LDAP_TYPE_OR_VALUE_EXISTS;
- goto return_results;
- }
- }
-
- if ( permissive && matched == i ) {
- rc = LDAP_TYPE_OR_VALUE_EXISTS;
- goto return_results;
- }
- }
- }
-
-return_results:;
- if ( nvals ) {
- ber_bvarray_free( nvals );
- }
- if ( nmods ) {
- ber_bvarray_free( nmods );
- }
-
+ /* this function is no longer used */
return rc;
}
/* test asserted values against existing values */
if( a ) {
for( matched = 0, j = 0; a->a_vals[j].bv_val != NULL; j++ ) {
- if ( bvmatch( &mod->sm_bvalues[i],
- &a->a_vals[j] ) ) {
+ if ( bvmatch( &mod->sm_bvalues[i], &a->a_vals[j] ) ) {
if ( permissive ) {
matched++;
continue;
/* test asserted values against themselves */
for( j = 0; j < i; j++ ) {
- if ( bvmatch( &mod->sm_bvalues[i],
- &mod->sm_bvalues[j] ) ) {
-
+ if ( bvmatch( &mod->sm_bvalues[i], &mod->sm_bvalues[j] ) ) {
/* value exists already */
*text = textbuf;
snprintf( textbuf, textlen,
}
} else {
- /*
- * The original code performs ( n ) normalizations
- * and ( n * ( n - 1 ) / 2 ) matches, which hide
- * the same number of normalizations. The new code
- * performs the same number of normalizations ( n )
- * and ( n * ( n - 1 ) / 2 ) mem compares, far less
- * expensive than an entire match, if a match is
- * equivalent to a normalization and a mem compare ...
- *
- * This is far more memory expensive than the previous,
- * but it can heavily improve performances when big
- * chunks of data are added (typical example is a group
- * with thousands of DN-syntax members; on my system:
- * for members of 5-RDN DNs,
-
- members orig bvmatch (dirty) new
- 1000 0m38.456s 0m0.553s 0m0.608s
- 2000 2m33.341s 0m0.851s 0m1.003s
-
- * Moreover, 100 groups with 10000 members each were
- * added in 37m27.933s (an analogous LDIF file was
- * loaded into Active Directory in 38m28.682s, BTW).
- *
- * Maybe we could switch to the new algorithm when
- * the number of values overcomes a given threshold?
+ /* no normalization is done in this routine nor
+ * in the matching routines called by this routine.
+ * values are now normalized once on input to the
+ * server (whether from LDAP or from the underlying
+ * database).
+ * This should outperform the old code. No numbers
+ * are available yet.
*/
int rc;
- if ( mod->sm_bvalues[ 1 ].bv_val == 0 ) {
+ if ( mod->sm_bvalues[1].bv_val == 0 ) {
if ( a != NULL ) {
- struct berval asserted;
int i;
- rc = value_normalize( mod->sm_desc, SLAP_MR_EQUALITY,
- &mod->sm_bvalues[ 0 ], &asserted, text );
-
- if ( rc != LDAP_SUCCESS ) {
- return rc;
- }
-
for ( matched = 0, i = 0; a->a_vals[ i ].bv_val; i++ ) {
int match;
- rc = value_match( &match, mod->sm_desc, mr,
- SLAP_MR_ATTRIBUTE_SYNTAX_MATCH,
- &a->a_vals[ i ], &asserted, text );
+ if( mod->sm_nvalues ) {
+ rc = value_match( &match, mod->sm_desc, mr,
+ SLAP_MR_EQUALITY
+ | SLAP_MR_VALUE_OF_ASSERTION_SYNTAX
+ | SLAP_MR_ASSERTED_VALUE_NORMALIZED_MATCH
+ | SLAP_MR_ATTRIBUTE_VALUE_NORMALIZED_MATCH,
+ &a->a_nvals[i],
+ &mod->sm_nvalues[0],
+ text );
+
+ } else {
+ rc = value_match( &match, mod->sm_desc, mr,
+ SLAP_MR_EQUALITY
+ | SLAP_MR_VALUE_OF_ATTRIBUTE_SYNTAX,
+ &a->a_vals[i],
+ &mod->sm_values[0],
+ text );
+ }
+
if( rc == LDAP_SUCCESS && match == 0 ) {
if ( permissive ) {
matched++;
continue;
}
- free( asserted.bv_val );
*text = textbuf;
snprintf( textbuf, textlen,
"modify/%s: %s: value #0 already exists",
} else {
rc = modify_check_duplicates( mod->sm_desc, mr,
a ? a->a_vals : NULL, mod->sm_bvalues,
- permissive,
- text, textbuf, textlen );
+ permissive, text, textbuf, textlen );
if ( permissive && rc == LDAP_TYPE_OR_VALUE_EXISTS ) {
return LDAP_SUCCESS;
}
/* no - add them */
-#ifdef SLAP_NVALUES
if( attr_merge( e, mod->sm_desc, mod->sm_values, mod->sm_nvalues ) != 0 )
-#else
- if( attr_merge( e, mod->sm_desc, mod->sm_bvalues ) != 0 )
-#endif
{
/* this should return result of attr_merge */
*text = textbuf;
int i, j, k, rc = LDAP_SUCCESS;
Attribute *a;
MatchingRule *mr = mod->sm_desc->ad_type->sat_equality;
- BerVarray nvals = NULL;
char dummy = '\0';
/*
return LDAP_NO_SUCH_ATTRIBUTE;
}
- /* find each value to delete */
- for ( j = 0; a->a_vals[ j ].bv_val != NULL; j++ )
- /* count existing values */ ;
-
- nvals = (BerVarray)SLAP_CALLOC( j + 1, sizeof ( struct berval ) );
- if( nvals == NULL ) {
-#ifdef NEW_LOGGING
- LDAP_LOG( OPERATION, ERR,
- "modify_delete_values: SLAP_CALLOC failed", 0, 0, 0 );
-#else
- Debug( LDAP_DEBUG_ANY,
- "modify_delete_values: SLAP_CALLOC failed", 0, 0, 0 );
-#endif
- goto return_results;
- }
-
- /* normalize existing values */
- for ( j = 0; a->a_vals[ j ].bv_val != NULL; j++ ) {
- rc = value_normalize( a->a_desc, SLAP_MR_EQUALITY,
- &a->a_vals[ j ], &nvals[ j ], text );
-
- if ( rc != LDAP_SUCCESS ) {
- nvals[ j ].bv_val = NULL;
- goto return_results;
- }
- }
- for ( i = 0; mod->sm_bvalues[ i ].bv_val != NULL; i++ ) {
- struct berval asserted;
+ for ( i = 0; mod->sm_values[i].bv_val != NULL; i++ ) {
int found = 0;
-
- /* normalize the value to be deleted */
- rc = value_normalize( mod->sm_desc, SLAP_MR_EQUALITY,
- &mod->sm_bvalues[ i ], &asserted, text );
-
- if( rc != LDAP_SUCCESS ) {
- goto return_results;
- }
-
- /* search it */
- for ( j = 0; nvals[ j ].bv_val != NULL; j++ ) {
+ for ( j = 0; a->a_vals[j].bv_val != NULL; j++ )
+ {
int match;
- if ( nvals[ j ].bv_val == &dummy ) {
- continue;
+ if( mod->sm_nvalues ) {
+ assert( a->a_nvals );
+ rc = (*mr->smr_match)( &match,
+ SLAP_MR_VALUE_OF_ASSERTION_SYNTAX
+ | SLAP_MR_ASSERTED_VALUE_NORMALIZED_MATCH
+ | SLAP_MR_ATTRIBUTE_VALUE_NORMALIZED_MATCH,
+ a->a_desc->ad_type->sat_syntax,
+ mr, &a->a_nvals[j],
+ &mod->sm_nvalues[i] );
+ } else {
+#if 0
+ assert( a->a_nvals == NULL );
+#endif
+ rc = (*mr->smr_match)( &match,
+ SLAP_MR_VALUE_OF_ATTRIBUTE_SYNTAX,
+ a->a_desc->ad_type->sat_syntax,
+ mr, &a->a_vals[j],
+ &mod->sm_values[i] );
}
- rc = (*mr->smr_match)( &match,
- SLAP_MR_ATTRIBUTE_SYNTAX_MATCH,
- a->a_desc->ad_type->sat_syntax,
- mr, &nvals[ j ], &asserted );
-
if ( rc != LDAP_SUCCESS ) {
- free( asserted.bv_val );
*text = textbuf;
snprintf( textbuf, textlen,
"%s: matching rule failed",
found = 1;
/* delete value and mark it as dummy */
- free( nvals[ j ].bv_val );
- nvals[ j ].bv_val = &dummy;
+ free( a->a_vals[j].bv_val );
+ a->a_vals[j].bv_val = &dummy;
+ if( a->a_nvals != a->a_vals ) {
+ free( a->a_nvals[j].bv_val );
+ a->a_nvals[j].bv_val = &dummy;
+ }
break;
}
- free( asserted.bv_val );
if ( found == 0 ) {
*text = textbuf;
}
/* compact array skipping dummies */
- for ( k = 0, j = 0; nvals[ k ].bv_val != NULL; j++, k++ ) {
-
- /* delete and skip dummies */ ;
- for ( ; nvals[ k ].bv_val == &dummy; k++ ) {
- free( a->a_vals[ k ].bv_val );
+ for ( k = 0, j = 0; a->a_vals[k].bv_val != NULL; k++ )
+ {
+ /* skip dummies */
+ if( a->a_vals[k].bv_val == &dummy ) {
+ assert( a->a_nvals == NULL || a->a_nvals[k].bv_val == &dummy );
+ continue;
}
-
if ( j != k ) {
a->a_vals[ j ] = a->a_vals[ k ];
-#ifdef SLAP_NVALUES
- if (a->a_nvals) {
- free( a->a_nvals[j].bv_val );
+ if (a->a_nvals != a->a_vals) {
a->a_nvals[ j ] = a->a_nvals[ k ];
}
-#endif
}
- if ( a->a_vals[ k ].bv_val == NULL ) {
- break;
- }
+ j++;
}
- a->a_vals[ j ].bv_val = NULL;
-#ifdef SLAP_NVALUES
- if (a->a_nvals) a->a_nvals[ j ].bv_val = NULL;
-#endif
- assert( i == k - j );
+ a->a_vals[j].bv_val = NULL;
+ if (a->a_nvals != a->a_vals) a->a_nvals[j].bv_val = NULL;
/* if no values remain, delete the entire attribute */
if ( a->a_vals[0].bv_val == NULL ) {
}
return_results:;
- if ( nvals ) {
- /* delete the remaining normalized values */
- for ( j = 0; nvals[ j ].bv_val != NULL; j++ ) {
- if ( nvals[ j ].bv_val != &dummy ) {
- ber_memfree( nvals[ j ].bv_val );
- }
- }
- ber_memfree( nvals );
- }
return rc;
}
if ( mod->sm_values != NULL ) ber_bvarray_free( mod->sm_values );
mod->sm_values = NULL;
-#ifdef SLAP_NVALUES
if ( mod->sm_nvalues != NULL ) ber_bvarray_free( mod->sm_nvalues );
mod->sm_nvalues = NULL;
-#endif
if( freeit ) free( mod );
}
projects / openldap / blobdiff