[Fedora-directory-commits] ldapserver/ldap/servers/plugins/syntaxes phonetic.c, 1.5, 1.6
Noriko Hosoi
nhosoi at fedoraproject.org
Mon Jan 12 19:18:40 UTC 2009
Author: nhosoi
Update of /cvs/dirsec/ldapserver/ldap/servers/plugins/syntaxes
In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv22952
Modified Files:
phonetic.c
Log Message:
Resolves: #460613
Summary: Approximate Search '~=' Returns unexpected result
Change description: increasing the maximum length of "phonetic" string from 4
to 6. The length 4 is sometimes too short to distinguish long words. For
instance, the sample string Queensland is converted to KNSLNT if there is no
limitation; Consulting is to KNSLTNK. By cutting them at the 5th character,
the 2 strings are considered to sound like each other.
Index: phonetic.c
===================================================================
RCS file: /cvs/dirsec/ldapserver/ldap/servers/plugins/syntaxes/phonetic.c,v
retrieving revision 1.5
retrieving revision 1.6
diff -u -r1.5 -r1.6
--- phonetic.c 10 Nov 2006 23:45:31 -0000 1.5
+++ phonetic.c 12 Jan 2009 19:18:38 -0000 1.6
@@ -68,7 +68,7 @@
case 0x00A0: /* non-breaking space */
case 0x3000: /* ideographic space */
case 0xFEFF: /* zero-width non-breaking space */
- return 1;
+ return 1;
default: break;
}
return 0;
@@ -77,61 +77,61 @@
char *
first_word( char *s )
{
- if ( s == NULL ) {
- return( NULL );
- }
-
- while ( iswordbreak( s ) ) {
- if ( *s == '\0' ) {
- return( NULL );
- } else {
- LDAP_UTF8INC( s );
- }
- }
+ if ( s == NULL ) {
+ return( NULL );
+ }
+
+ while ( iswordbreak( s ) ) {
+ if ( *s == '\0' ) {
+ return( NULL );
+ } else {
+ LDAP_UTF8INC( s );
+ }
+ }
- return( s );
+ return( s );
}
char *
next_word( char *s )
{
- if ( s == NULL ) {
- return( NULL );
- }
-
- while ( ! iswordbreak( s ) ) {
- LDAP_UTF8INC( s );
- }
-
- while ( iswordbreak( s ) ) {
- if ( *s == '\0' ) {
- return( NULL );
- } else {
- LDAP_UTF8INC( s );
- }
- }
+ if ( s == NULL ) {
+ return( NULL );
+ }
+
+ while ( ! iswordbreak( s ) ) {
+ LDAP_UTF8INC( s );
+ }
+
+ while ( iswordbreak( s ) ) {
+ if ( *s == '\0' ) {
+ return( NULL );
+ } else {
+ LDAP_UTF8INC( s );
+ }
+ }
- return( s );
+ return( s );
}
char *
word_dup( char *w )
{
- char *s, *ret;
- char save;
+ char *s, *ret;
+ char save;
- for ( s = w; !iswordbreak( s ); LDAP_UTF8INC( s ))
- ; /* NULL */
- save = *s;
- *s = '\0';
- ret = slapi_ch_strdup( w );
- *s = save;
+ for ( s = w; !iswordbreak( s ); LDAP_UTF8INC( s ))
+ ; /* NULL */
+ save = *s;
+ *s = '\0';
+ ret = slapi_ch_strdup( w );
+ *s = save;
- return( ret );
+ return( ret );
}
#ifndef MAXPHONEMELEN
-#define MAXPHONEMELEN 4
+#define MAXPHONEMELEN 6
#endif
#if defined(SOUNDEX)
@@ -140,11 +140,11 @@
char *
phonetic( char *s )
{
- char code, adjacent, ch;
- char *p;
- char **c;
- int i, cmax;
- char phoneme[MAXPHONEMELEN + 1];
+ char code, adjacent, ch;
+ char *p;
+ char **c;
+ int i, cmax;
+ char phoneme[MAXPHONEMELEN + 1];
p = s;
if ( p == NULL || *p == '\0' ) {
@@ -152,18 +152,18 @@
}
adjacent = '0';
- phoneme[0] = TOUPPER(*p);
+ phoneme[0] = TOUPPER(*p);
- phoneme[1] = '\0';
+ phoneme[1] = '\0';
for ( i = 0; i < 99 && (! iswordbreak(p)); LDAP_UTF8INC( p )) {
- ch = TOUPPER (*p);
+ ch = TOUPPER (*p);
code = '0';
switch (ch) {
case 'B':
case 'F':
- case 'P':
+ case 'P':
case 'V':
code = (adjacent != '1') ? '1' : '0';
break;
@@ -196,18 +196,18 @@
}
if ( i == 0 ) {
- adjacent = code;
- i++;
- } else if ( code != '0' ) {
- if ( i == MAXPHONEMELEN )
- break;
+ adjacent = code;
+ i++;
+ } else if ( code != '0' ) {
+ if ( i == MAXPHONEMELEN )
+ break;
adjacent = phoneme[i] = code;
i++;
}
}
- if ( i > 0 )
- phoneme[i] = '\0';
+ if ( i > 0 )
+ phoneme[i] = '\0';
return( slapi_ch_strdup( phoneme ) );
}
@@ -224,274 +224,274 @@
/* Character coding array */
static char vsvfn[26] = {
- 1, 16, 4, 16, 9, 2, 4, 16, 9, 2, 0, 2, 2,
- /* A B C D E F G H I J K L M */
- 2, 1, 4, 0, 2, 4, 4, 1, 0, 0, 0, 8, 0};
- /* N O P Q R S T U V W X Y Z */
+ 1, 16, 4, 16, 9, 2, 4, 16, 9, 2, 0, 2, 2,
+ /* A B C D E F G H I J K L M */
+ 2, 1, 4, 0, 2, 4, 4, 1, 0, 0, 0, 8, 0};
+ /* N O P Q R S T U V W X Y Z */
/* Macros to access character coding array */
-#define vowel(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 1) /* AEIOU */
-#define same(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 2) /* FJLMNR */
-#define varson(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 4) /* CGPST */
-#define frontv(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 8) /* EIY */
-#define noghf(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 16) /* BDH */
+#define vowel(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 1) /* AEIOU */
+#define same(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 2) /* FJLMNR */
+#define varson(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 4) /* CGPST */
+#define frontv(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 8) /* EIY */
+#define noghf(x) ((x) != '\0' && vsvfn[(x) - 'A'] & 16) /* BDH */
char *
phonetic( char *Word )
{
- char *n, *n_start, *n_end; /* pointers to string */
- char *metaph_end; /* pointers to metaph */
- char ntrans[42]; /* word with uppercase letters */
- int KSflag; /* state flag for X -> KS */
- char buf[MAXPHONEMELEN + 2];
- char *Metaph;
-
- /*
- * Copy Word to internal buffer, dropping non-alphabetic characters
- * and converting to upper case
- */
- n = ntrans + 4; n_end = ntrans + 35;
- while (!iswordbreak( Word ) && n < n_end) {
- if (isascii(*Word)) {
- if (isalpha(*Word)) {
- *n++ = TOUPPER(*Word);
- }
- ++Word;
- } else {
- auto const size_t len = LDAP_UTF8COPY(n, Word);
- n += len; Word += len;
- }
- }
- Metaph = buf;
- *Metaph = '\0';
- if (n == ntrans + 4) {
- return( slapi_ch_strdup( buf ) ); /* Return if null */
- }
- n_end = n; /* Set n_end to end of string */
-
- /* ntrans[0] will always be == 0 */
- ntrans[0] = '\0';
- ntrans[1] = '\0';
- ntrans[2] = '\0';
- ntrans[3] = '\0';
- *n++ = 0;
- *n++ = 0;
- *n++ = 0;
- *n = 0; /* Pad with nulls */
- n = ntrans + 4; /* Assign pointer to start */
-
- /* Check for PN, KN, GN, AE, WR, WH, and X at start */
- switch (*n) {
- case 'P':
- case 'K':
- case 'G':
- /* 'PN', 'KN', 'GN' becomes 'N' */
- if (*(n + 1) == 'N')
- *n++ = 0;
- break;
- case 'A':
- /* 'AE' becomes 'E' */
- if (*(n + 1) == 'E')
- *n++ = 0;
- break;
- case 'W':
- /* 'WR' becomes 'R', and 'WH' to 'H' */
- if (*(n + 1) == 'R')
- *n++ = 0;
- else if (*(n + 1) == 'H') {
- *(n + 1) = *n;
- *n++ = 0;
- }
- break;
- case 'X':
- /* 'X' becomes 'S' */
- *n = 'S';
- break;
- }
-
- /*
- * Now, loop step through string, stopping at end of string or when
- * the computed 'metaph' is MAXPHONEMELEN characters long
- */
-
- KSflag = 0; /* state flag for KS translation */
- for (metaph_end = Metaph + MAXPHONEMELEN, n_start = n;
- n <= n_end && Metaph < metaph_end; n++) {
- if (KSflag) {
- KSflag = 0;
- *Metaph++ = 'S';
- } else if (!isascii(*n)) {
- *Metaph++ = *n;
- } else {
- /* Drop duplicates except for CC */
- if (*(n - 1) == *n && *n != 'C')
- continue;
- /* Check for F J L M N R or first letter vowel */
- if (same(*n) || (n == n_start && vowel(*n))) {
- *Metaph++ = *n;
- } else {
- switch (*n) {
- case 'B':
-
- /*
- * B unless in -MB
- */
- if (n < (n_end - 1) && *(n - 1) != 'M') {
- *Metaph++ = *n;
- }
- break;
- case 'C':
-
- /*
- * X if in -CIA-, -CH- else S if in
- * -CI-, -CE-, -CY- else dropped if
- * in -SCI-, -SCE-, -SCY- else K
- */
- if (*(n - 1) != 'S' || !frontv(*(n + 1))) {
- if (*(n + 1) == 'I' && *(n + 2) == 'A') {
- *Metaph++ = 'X';
- } else if (frontv(*(n + 1))) {
- *Metaph++ = 'S';
- } else if (*(n + 1) == 'H') {
- *Metaph++ = ((n == n_start && !vowel(*(n + 2)))
- || *(n - 1) == 'S')
- ? (char) 'K' : (char) 'X';
- } else {
- *Metaph++ = 'K';
- }
- }
- break;
- case 'D':
-
- /*
- * J if in DGE or DGI or DGY else T
- */
- *Metaph++ = (*(n + 1) == 'G' && frontv(*(n + 2)))
- ? (char) 'J' : (char) 'T';
- break;
- case 'G':
-
- /*
- * F if in -GH and not B--GH, D--GH,
- * -H--GH, -H---GH else dropped if
- * -GNED, -GN, -DGE-, -DGI-, -DGY-
- * else J if in -GE-, -GI-, -GY- and
- * not GG else K
- */
- if ((*(n + 1) != 'J' || vowel(*(n + 2))) &&
- (*(n + 1) != 'N' || ((n + 1) < n_end &&
- (*(n + 2) != 'E' || *(n + 3) != 'D'))) &&
- (*(n - 1) != 'D' || !frontv(*(n + 1))))
- *Metaph++ = (frontv(*(n + 1)) &&
- *(n + 2) != 'G') ? (char) 'G' : (char) 'K';
- else if (*(n + 1) == 'H' && !noghf(*(n - 3)) &&
- *(n - 4) != 'H')
- *Metaph++ = 'F';
- break;
- case 'H':
-
- /*
- * H if before a vowel and not after
- * C, G, P, S, T else dropped
- */
- if (!varson(*(n - 1)) && (!vowel(*(n - 1)) ||
- vowel(*(n + 1))))
- *Metaph++ = 'H';
- break;
- case 'K':
-
- /*
- * dropped if after C else K
- */
- if (*(n - 1) != 'C')
- *Metaph++ = 'K';
- break;
- case 'P':
-
- /*
- * F if before H, else P
- */
- *Metaph++ = *(n + 1) == 'H' ?
- (char) 'F' : (char) 'P';
- break;
- case 'Q':
-
- /*
- * K
- */
- *Metaph++ = 'K';
- break;
- case 'S':
-
- /*
- * X in -SH-, -SIO- or -SIA- else S
- */
- *Metaph++ = (*(n + 1) == 'H' ||
- (*(n + 1) == 'I' && (*(n + 2) == 'O' ||
- *(n + 2) == 'A')))
- ? (char) 'X' : (char) 'S';
- break;
- case 'T':
-
- /*
- * X in -TIA- or -TIO- else 0 (zero)
- * before H else dropped if in -TCH-
- * else T
- */
- if (*(n + 1) == 'I' && (*(n + 2) == 'O' ||
- *(n + 2) == 'A'))
- *Metaph++ = 'X';
- else if (*(n + 1) == 'H')
- *Metaph++ = '0';
- else if (*(n + 1) != 'C' || *(n + 2) != 'H')
- *Metaph++ = 'T';
- break;
- case 'V':
-
- /*
- * F
- */
- *Metaph++ = 'F';
- break;
- case 'W':
-
- /*
- * W after a vowel, else dropped
- */
- case 'Y':
-
- /*
- * Y unless followed by a vowel
- */
- if (vowel(*(n + 1)))
- *Metaph++ = *n;
- break;
- case 'X':
-
- /*
- * KS
- */
- if (n == n_start)
- *Metaph++ = 'S';
- else {
- *Metaph++ = 'K'; /* Insert K, then S */
- KSflag = 1;
- }
- break;
- case 'Z':
-
- /*
- * S
- */
- *Metaph++ = 'S';
- break;
- }
- }
- }
- }
+ char *n, *n_start, *n_end; /* pointers to string */
+ char *metaph_end; /* pointers to metaph */
+ char ntrans[42]; /* word with uppercase letters */
+ int KSflag; /* state flag for X -> KS */
+ char buf[MAXPHONEMELEN + 2];
+ char *Metaph;
+
+ /*
+ * Copy Word to internal buffer, dropping non-alphabetic characters
+ * and converting to upper case
+ */
+ n = ntrans + 4; n_end = ntrans + 35;
+ while (!iswordbreak( Word ) && n < n_end) {
+ if (isascii(*Word)) {
+ if (isalpha(*Word)) {
+ *n++ = TOUPPER(*Word);
+ }
+ ++Word;
+ } else {
+ auto const size_t len = LDAP_UTF8COPY(n, Word);
+ n += len; Word += len;
+ }
+ }
+ Metaph = buf;
+ *Metaph = '\0';
+ if (n == ntrans + 4) {
+ return( slapi_ch_strdup( buf ) ); /* Return if null */
+ }
+ n_end = n; /* Set n_end to end of string */
+
+ /* ntrans[0] will always be == 0 */
+ ntrans[0] = '\0';
+ ntrans[1] = '\0';
+ ntrans[2] = '\0';
+ ntrans[3] = '\0';
+ *n++ = 0;
+ *n++ = 0;
+ *n++ = 0;
+ *n = 0; /* Pad with nulls */
+ n = ntrans + 4; /* Assign pointer to start */
+
+ /* Check for PN, KN, GN, AE, WR, WH, and X at start */
+ switch (*n) {
+ case 'P':
+ case 'K':
+ case 'G':
+ /* 'PN', 'KN', 'GN' becomes 'N' */
+ if (*(n + 1) == 'N')
+ *n++ = 0;
+ break;
+ case 'A':
+ /* 'AE' becomes 'E' */
+ if (*(n + 1) == 'E')
+ *n++ = 0;
+ break;
+ case 'W':
+ /* 'WR' becomes 'R', and 'WH' to 'H' */
+ if (*(n + 1) == 'R')
+ *n++ = 0;
+ else if (*(n + 1) == 'H') {
+ *(n + 1) = *n;
+ *n++ = 0;
+ }
+ break;
+ case 'X':
+ /* 'X' becomes 'S' */
+ *n = 'S';
+ break;
+ }
+
+ /*
+ * Now, loop step through string, stopping at end of string or when
+ * the computed 'metaph' is MAXPHONEMELEN characters long
+ */
+
+ KSflag = 0; /* state flag for KS translation */
+ for (metaph_end = Metaph + MAXPHONEMELEN, n_start = n;
+ n <= n_end && Metaph < metaph_end; n++) {
+ if (KSflag) {
+ KSflag = 0;
+ *Metaph++ = 'S';
+ } else if (!isascii(*n)) {
+ *Metaph++ = *n;
+ } else {
+ /* Drop duplicates except for CC */
+ if (*(n - 1) == *n && *n != 'C')
+ continue;
+ /* Check for F J L M N R or first letter vowel */
+ if (same(*n) || (n == n_start && vowel(*n))) {
+ *Metaph++ = *n;
+ } else {
+ switch (*n) {
+ case 'B':
+
+ /*
+ * B unless in -MB
+ */
+ if (n < (n_end - 1) && *(n - 1) != 'M') {
+ *Metaph++ = *n;
+ }
+ break;
+ case 'C':
+
+ /*
+ * X if in -CIA-, -CH- else S if in
+ * -CI-, -CE-, -CY- else dropped if
+ * in -SCI-, -SCE-, -SCY- else K
+ */
+ if (*(n - 1) != 'S' || !frontv(*(n + 1))) {
+ if (*(n + 1) == 'I' && *(n + 2) == 'A') {
+ *Metaph++ = 'X';
+ } else if (frontv(*(n + 1))) {
+ *Metaph++ = 'S';
+ } else if (*(n + 1) == 'H') {
+ *Metaph++ = ((n == n_start && !vowel(*(n + 2)))
+ || *(n - 1) == 'S')
+ ? (char) 'K' : (char) 'X';
+ } else {
+ *Metaph++ = 'K';
+ }
+ }
+ break;
+ case 'D':
+
+ /*
+ * J if in DGE or DGI or DGY else T
+ */
+ *Metaph++ = (*(n + 1) == 'G' && frontv(*(n + 2)))
+ ? (char) 'J' : (char) 'T';
+ break;
+ case 'G':
+
+ /*
+ * F if in -GH and not B--GH, D--GH,
+ * -H--GH, -H---GH else dropped if
+ * -GNED, -GN, -DGE-, -DGI-, -DGY-
+ * else J if in -GE-, -GI-, -GY- and
+ * not GG else K
+ */
+ if ((*(n + 1) != 'J' || vowel(*(n + 2))) &&
+ (*(n + 1) != 'N' || ((n + 1) < n_end &&
+ (*(n + 2) != 'E' || *(n + 3) != 'D'))) &&
+ (*(n - 1) != 'D' || !frontv(*(n + 1))))
+ *Metaph++ = (frontv(*(n + 1)) &&
+ *(n + 2) != 'G') ? (char) 'G' : (char) 'K';
+ else if (*(n + 1) == 'H' && !noghf(*(n - 3)) &&
+ *(n - 4) != 'H')
+ *Metaph++ = 'F';
+ break;
+ case 'H':
+
+ /*
+ * H if before a vowel and not after
+ * C, G, P, S, T else dropped
+ */
+ if (!varson(*(n - 1)) && (!vowel(*(n - 1)) ||
+ vowel(*(n + 1))))
+ *Metaph++ = 'H';
+ break;
+ case 'K':
+
+ /*
+ * dropped if after C else K
+ */
+ if (*(n - 1) != 'C')
+ *Metaph++ = 'K';
+ break;
+ case 'P':
+
+ /*
+ * F if before H, else P
+ */
+ *Metaph++ = *(n + 1) == 'H' ?
+ (char) 'F' : (char) 'P';
+ break;
+ case 'Q':
+
+ /*
+ * K
+ */
+ *Metaph++ = 'K';
+ break;
+ case 'S':
+
+ /*
+ * X in -SH-, -SIO- or -SIA- else S
+ */
+ *Metaph++ = (*(n + 1) == 'H' ||
+ (*(n + 1) == 'I' && (*(n + 2) == 'O' ||
+ *(n + 2) == 'A')))
+ ? (char) 'X' : (char) 'S';
+ break;
+ case 'T':
+
+ /*
+ * X in -TIA- or -TIO- else 0 (zero)
+ * before H else dropped if in -TCH-
+ * else T
+ */
+ if (*(n + 1) == 'I' && (*(n + 2) == 'O' ||
+ *(n + 2) == 'A'))
+ *Metaph++ = 'X';
+ else if (*(n + 1) == 'H')
+ *Metaph++ = '0';
+ else if (*(n + 1) != 'C' || *(n + 2) != 'H')
+ *Metaph++ = 'T';
+ break;
+ case 'V':
+
+ /*
+ * F
+ */
+ *Metaph++ = 'F';
+ break;
+ case 'W':
+
+ /*
+ * W after a vowel, else dropped
+ */
+ case 'Y':
+
+ /*
+ * Y unless followed by a vowel
+ */
+ if (vowel(*(n + 1)))
+ *Metaph++ = *n;
+ break;
+ case 'X':
+
+ /*
+ * KS
+ */
+ if (n == n_start)
+ *Metaph++ = 'S';
+ else {
+ *Metaph++ = 'K'; /* Insert K, then S */
+ KSflag = 1;
+ }
+ break;
+ case 'Z':
+
+ /*
+ * S
+ */
+ *Metaph++ = 'S';
+ break;
+ }
+ }
+ }
+ }
- *Metaph = 0; /* Null terminate */
- return( slapi_ch_strdup( buf ) );
+ *Metaph = 0; /* Null terminate */
+ return( slapi_ch_strdup( buf ) );
}
#endif /* METAPHONE */
More information about the Fedora-directory-commits
mailing list