version 1.12, 2011/11/26 22:38:11 |
version 1.27, 2011/12/20 21:41:11 |
|
|
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
*/ |
*/ |
|
#ifdef HAVE_CONFIG_H |
|
#include "config.h" |
|
#endif |
|
|
#include <assert.h> |
#include <assert.h> |
#include <fcntl.h> |
#include <fcntl.h> |
#include <regex.h> |
#include <regex.h> |
|
|
#include <string.h> |
#include <string.h> |
#include <unistd.h> |
#include <unistd.h> |
|
|
#ifdef __linux__ |
#if defined(__linux__) |
|
# include <endian.h> |
# include <db_185.h> |
# include <db_185.h> |
|
#elif defined(__APPLE__) |
|
# include <libkern/OSByteOrder.h> |
|
# include <db.h> |
#else |
#else |
# include <db.h> |
# include <db.h> |
#endif |
#endif |
|
|
int regex; /* is regex? */ |
int regex; /* is regex? */ |
int index; /* index in match array */ |
int index; /* index in match array */ |
uint64_t mask; /* type-mask */ |
uint64_t mask; /* type-mask */ |
int cs; /* is case-sensitive? */ |
|
int and; /* is rhs of logical AND? */ |
int and; /* is rhs of logical AND? */ |
char *v; /* search value */ |
char *v; /* search value */ |
regex_t re; /* compiled re, if regex */ |
regex_t re; /* compiled re, if regex */ |
Line 110 static const struct type types[] = { |
|
Line 117 static const struct type types[] = { |
|
{ TYPE_Va, "Va" }, |
{ TYPE_Va, "Va" }, |
{ TYPE_Va, "Vt" }, |
{ TYPE_Va, "Vt" }, |
{ TYPE_Xr, "Xr" }, |
{ TYPE_Xr, "Xr" }, |
{ INT_MAX, "any" }, |
{ UINT64_MAX, "any" }, |
{ 0, NULL } |
{ 0, NULL } |
}; |
}; |
|
|
static DB *btree_open(void); |
static DB *btree_open(void); |
static int btree_read(const DBT *, |
static int btree_read(const DBT *, const DBT *, |
const struct mchars *, char **); |
const struct mchars *, |
|
struct db_val *, char **); |
static int expreval(const struct expr *, int *); |
static int expreval(const struct expr *, int *); |
static void exprexec(const struct expr *, |
static void exprexec(const struct expr *, |
const char *, uint64_t, struct rec *); |
const char *, uint64_t, struct rec *); |
Line 159 btree_open(void) |
|
Line 167 btree_open(void) |
|
* Return 0 if the database is insane, else 1. |
* Return 0 if the database is insane, else 1. |
*/ |
*/ |
static int |
static int |
btree_read(const DBT *v, const struct mchars *mc, char **buf) |
btree_read(const DBT *k, const DBT *v, |
|
const struct mchars *mc, |
|
struct db_val *dbv, char **buf) |
{ |
{ |
|
struct db_val raw_dbv; |
|
|
/* Sanity: are we nil-terminated? */ |
/* Are our sizes sane? */ |
|
if (k->size < 2 || sizeof(struct db_val) != v->size) |
|
return(0); |
|
|
assert(v->size > 0); |
/* Is our string nil-terminated? */ |
|
if ('\0' != ((const char *)k->data)[(int)k->size - 1]) |
if ('\0' != ((char *)v->data)[(int)v->size - 1]) |
|
return(0); |
return(0); |
|
|
norm_string((char *)v->data, mc, buf); |
norm_string((const char *)k->data, mc, buf); |
|
memcpy(&raw_dbv, v->data, v->size); |
|
dbv->rec = betoh32(raw_dbv.rec); |
|
dbv->mask = betoh64(raw_dbv.mask); |
return(1); |
return(1); |
} |
} |
|
|
Line 182 btree_read(const DBT *v, const struct mchars *mc, char |
|
Line 197 btree_read(const DBT *v, const struct mchars *mc, char |
|
static size_t |
static size_t |
norm_utf8(unsigned int cp, char out[7]) |
norm_utf8(unsigned int cp, char out[7]) |
{ |
{ |
size_t rc; |
int rc; |
|
|
rc = 0; |
rc = 0; |
|
|
Line 223 norm_utf8(unsigned int cp, char out[7]) |
|
Line 238 norm_utf8(unsigned int cp, char out[7]) |
|
return(0); |
return(0); |
|
|
out[rc] = '\0'; |
out[rc] = '\0'; |
return(rc); |
return((size_t)rc); |
} |
} |
|
|
/* |
/* |
Line 350 index_read(const DBT *key, const DBT *val, int index, |
|
Line 365 index_read(const DBT *key, const DBT *val, int index, |
|
{ |
{ |
size_t left; |
size_t left; |
char *np, *cp; |
char *np, *cp; |
|
char type; |
|
|
#define INDEX_BREAD(_dst) \ |
#define INDEX_BREAD(_dst) \ |
do { \ |
do { \ |
Line 360 index_read(const DBT *key, const DBT *val, int index, |
|
Line 376 index_read(const DBT *key, const DBT *val, int index, |
|
cp = np + 1; \ |
cp = np + 1; \ |
} while (/* CONSTCOND */ 0) |
} while (/* CONSTCOND */ 0) |
|
|
left = val->size; |
if (0 == (left = val->size)) |
cp = (char *)val->data; |
return(0); |
|
|
rec->res.rec = *(recno_t *)key->data; |
cp = val->data; |
|
assert(sizeof(recno_t) == key->size); |
|
memcpy(&rec->res.rec, key->data, key->size); |
rec->res.volume = index; |
rec->res.volume = index; |
|
|
|
if ('d' == (type = *cp++)) |
|
rec->res.type = RESTYPE_MDOC; |
|
else if ('a' == type) |
|
rec->res.type = RESTYPE_MAN; |
|
else if ('c' == type) |
|
rec->res.type = RESTYPE_CAT; |
|
else |
|
return(0); |
|
|
|
left--; |
INDEX_BREAD(rec->res.file); |
INDEX_BREAD(rec->res.file); |
INDEX_BREAD(rec->res.cat); |
INDEX_BREAD(rec->res.cat); |
INDEX_BREAD(rec->res.title); |
INDEX_BREAD(rec->res.title); |
Line 442 single_search(struct rectree *tree, const struct opts |
|
Line 470 single_search(struct rectree *tree, const struct opts |
|
struct mchars *mc, int vol) |
struct mchars *mc, int vol) |
{ |
{ |
int root, leaf, ch; |
int root, leaf, ch; |
uint64_t mask; |
|
DBT key, val; |
DBT key, val; |
DB *btree, *idx; |
DB *btree, *idx; |
char *buf; |
char *buf; |
recno_t rec; |
|
struct rec *rs; |
struct rec *rs; |
struct rec r; |
struct rec r; |
struct db_val *vbuf; |
struct db_val vb; |
|
|
root = -1; |
root = -1; |
leaf = -1; |
leaf = -1; |
Line 469 single_search(struct rectree *tree, const struct opts |
|
Line 495 single_search(struct rectree *tree, const struct opts |
|
} |
} |
|
|
while (0 == (ch = (*btree->seq)(btree, &key, &val, R_NEXT))) { |
while (0 == (ch = (*btree->seq)(btree, &key, &val, R_NEXT))) { |
if (key.size < 2 || sizeof(struct db_val) != val.size) |
if ( ! btree_read(&key, &val, mc, &vb, &buf)) |
break; |
break; |
if ( ! btree_read(&key, mc, &buf)) |
|
break; |
|
|
|
vbuf = val.data; |
|
rec = vbuf->rec; |
|
mask = vbuf->mask; |
|
|
|
/* |
/* |
* See if this keyword record matches any of the |
* See if this keyword record matches any of the |
* expressions we have stored. |
* expressions we have stored. |
*/ |
*/ |
if ( ! exprmark(expr, buf, mask, NULL)) |
if ( ! exprmark(expr, buf, vb.mask, NULL)) |
continue; |
continue; |
|
|
/* |
/* |
Line 492 single_search(struct rectree *tree, const struct opts |
|
Line 512 single_search(struct rectree *tree, const struct opts |
|
*/ |
*/ |
|
|
for (leaf = root; leaf >= 0; ) |
for (leaf = root; leaf >= 0; ) |
if (rec > rs[leaf].res.rec && |
if (vb.rec > rs[leaf].res.rec && |
rs[leaf].rhs >= 0) |
rs[leaf].rhs >= 0) |
leaf = rs[leaf].rhs; |
leaf = rs[leaf].rhs; |
else if (rec < rs[leaf].res.rec && |
else if (vb.rec < rs[leaf].res.rec && |
rs[leaf].lhs >= 0) |
rs[leaf].lhs >= 0) |
leaf = rs[leaf].lhs; |
leaf = rs[leaf].lhs; |
else |
else |
Line 507 single_search(struct rectree *tree, const struct opts |
|
Line 527 single_search(struct rectree *tree, const struct opts |
|
* try to evaluate it now and continue anyway. |
* try to evaluate it now and continue anyway. |
*/ |
*/ |
|
|
if (leaf >= 0 && rs[leaf].res.rec == rec) { |
if (leaf >= 0 && rs[leaf].res.rec == vb.rec) { |
if (0 == rs[leaf].matched) |
if (0 == rs[leaf].matched) |
exprexec(expr, buf, mask, &rs[leaf]); |
exprexec(expr, buf, vb.mask, &rs[leaf]); |
continue; |
continue; |
} |
} |
|
|
Line 519 single_search(struct rectree *tree, const struct opts |
|
Line 539 single_search(struct rectree *tree, const struct opts |
|
* database, then begin partial evaluation. |
* database, then begin partial evaluation. |
*/ |
*/ |
|
|
key.data = &rec; |
key.data = &vb.rec; |
key.size = sizeof(recno_t); |
key.size = sizeof(recno_t); |
|
|
if (0 != (*idx->get)(idx, &key, &val, 0)) |
if (0 != (*idx->get)(idx, &key, &val, 0)) |
Line 533 single_search(struct rectree *tree, const struct opts |
|
Line 553 single_search(struct rectree *tree, const struct opts |
|
|
|
if (opts->cat && strcasecmp(opts->cat, r.res.cat)) |
if (opts->cat && strcasecmp(opts->cat, r.res.cat)) |
continue; |
continue; |
if (opts->arch && strcasecmp(opts->arch, r.res.arch)) |
|
continue; |
|
|
|
|
if (opts->arch && *r.res.arch) |
|
if (strcasecmp(opts->arch, r.res.arch)) |
|
continue; |
|
|
tree->node = rs = mandoc_realloc |
tree->node = rs = mandoc_realloc |
(rs, (tree->len + 1) * sizeof(struct rec)); |
(rs, (tree->len + 1) * sizeof(struct rec)); |
|
|
memcpy(&rs[tree->len], &r, sizeof(struct rec)); |
memcpy(&rs[tree->len], &r, sizeof(struct rec)); |
|
memset(&r, 0, sizeof(struct rec)); |
rs[tree->len].matches = |
rs[tree->len].matches = |
mandoc_calloc(terms, sizeof(int)); |
mandoc_calloc(terms, sizeof(int)); |
|
|
exprexec(expr, buf, mask, &rs[tree->len]); |
exprexec(expr, buf, vb.mask, &rs[tree->len]); |
|
|
/* Append to our tree. */ |
/* Append to our tree. */ |
|
|
if (leaf >= 0) { |
if (leaf >= 0) { |
if (rec > rs[leaf].res.rec) |
if (vb.rec > rs[leaf].res.rec) |
rs[leaf].rhs = tree->len; |
rs[leaf].rhs = tree->len; |
else |
else |
rs[leaf].lhs = tree->len; |
rs[leaf].lhs = tree->len; |
} else |
} else |
root = tree->len; |
root = tree->len; |
|
|
memset(&r, 0, sizeof(struct rec)); |
|
tree->len++; |
tree->len++; |
} |
} |
|
|
Line 563 single_search(struct rectree *tree, const struct opts |
|
Line 585 single_search(struct rectree *tree, const struct opts |
|
(*idx->close)(idx); |
(*idx->close)(idx); |
|
|
free(buf); |
free(buf); |
|
recfree(&r); |
return(1 == ch); |
return(1 == ch); |
} |
} |
|
|
Line 579 recfree(struct rec *rec) |
|
Line 602 recfree(struct rec *rec) |
|
free(rec->matches); |
free(rec->matches); |
} |
} |
|
|
|
/* |
|
* Compile a list of straight-up terms. |
|
* The arguments are re-written into ~[[:<:]]term[[:>:]], or "term" |
|
* surrounded by word boundaries, then pumped through exprterm(). |
|
* Terms are case-insensitive. |
|
* This emulates whatis(1) behaviour. |
|
*/ |
struct expr * |
struct expr * |
|
termcomp(int argc, char *argv[], size_t *tt) |
|
{ |
|
char *buf; |
|
int pos; |
|
struct expr *e, *next; |
|
size_t sz; |
|
|
|
buf = NULL; |
|
e = NULL; |
|
*tt = 0; |
|
|
|
for (pos = argc - 1; pos >= 0; pos--) { |
|
sz = strlen(argv[pos]) + 18; |
|
buf = mandoc_realloc(buf, sz); |
|
strlcpy(buf, "Nm~[[:<:]]", sz); |
|
strlcat(buf, argv[pos], sz); |
|
strlcat(buf, "[[:>:]]", sz); |
|
if (NULL == (next = exprterm(buf, 0))) { |
|
free(buf); |
|
exprfree(e); |
|
return(NULL); |
|
} |
|
next->next = e; |
|
e = next; |
|
(*tt)++; |
|
} |
|
|
|
free(buf); |
|
return(e); |
|
} |
|
|
|
/* |
|
* Compile a sequence of logical expressions. |
|
* See apropos.1 for a grammar of this sequence. |
|
*/ |
|
struct expr * |
exprcomp(int argc, char *argv[], size_t *tt) |
exprcomp(int argc, char *argv[], size_t *tt) |
{ |
{ |
int pos, lvl; |
int pos, lvl; |
Line 650 exprexpr(int argc, char *argv[], int *pos, int *lvl, s |
|
Line 716 exprexpr(int argc, char *argv[], int *pos, int *lvl, s |
|
++(*pos); |
++(*pos); |
++(*lvl); |
++(*lvl); |
next = mandoc_calloc(1, sizeof(struct expr)); |
next = mandoc_calloc(1, sizeof(struct expr)); |
next->cs = 1; |
|
next->subexpr = exprexpr(argc, argv, pos, lvl, tt); |
next->subexpr = exprexpr(argc, argv, pos, lvl, tt); |
if (NULL == next->subexpr) { |
if (NULL == next->subexpr) { |
free(next); |
free(next); |
Line 701 exprterm(char *buf, int cs) |
|
Line 766 exprterm(char *buf, int cs) |
|
|
|
memset(&e, 0, sizeof(struct expr)); |
memset(&e, 0, sizeof(struct expr)); |
|
|
e.cs = cs; |
|
|
|
/* Choose regex or substring match. */ |
/* Choose regex or substring match. */ |
|
|
if (NULL == (e.v = strpbrk(buf, "=~"))) { |
if (NULL == (e.v = strpbrk(buf, "=~"))) { |
Line 729 exprterm(char *buf, int cs) |
|
Line 792 exprterm(char *buf, int cs) |
|
e.mask = TYPE_Nm | TYPE_Nd; |
e.mask = TYPE_Nm | TYPE_Nd; |
|
|
if (e.regex) { |
if (e.regex) { |
i = REG_EXTENDED | REG_NOSUB | cs ? 0 : REG_ICASE; |
i = REG_EXTENDED | REG_NOSUB | (cs ? 0 : REG_ICASE); |
if (regcomp(&e.re, e.v, i)) |
if (regcomp(&e.re, e.v, i)) |
return(NULL); |
return(NULL); |
} |
} |
Line 774 exprmark(const struct expr *p, const char *cp, |
|
Line 837 exprmark(const struct expr *p, const char *cp, |
|
if (p->regex) { |
if (p->regex) { |
if (regexec(&p->re, cp, 0, NULL, 0)) |
if (regexec(&p->re, cp, 0, NULL, 0)) |
continue; |
continue; |
} else if (p->cs) { |
} else if (NULL == strcasestr(cp, p->v)) |
if (NULL == strstr(cp, p->v)) |
continue; |
continue; |
|
} else { |
|
if (NULL == strcasestr(cp, p->v)) |
|
continue; |
|
} |
|
|
|
if (NULL == ms) |
if (NULL == ms) |
return(1); |
return(1); |