* overhaul apropos(1) interface
@ 2011-11-09 1:30 Ingo Schwarze
2011-11-09 10:19 ` Kristaps Dzonsons
2011-11-12 23:54 ` Ingo Schwarze
0 siblings, 2 replies; 6+ messages in thread
From: Ingo Schwarze @ 2011-11-09 1:30 UTC (permalink / raw)
To: tech; +Cc: jmc
Hi,
the interface of the apropos utility in the mandoc package
is very different from traditional apropos, and the -t option
hinders logical extensions, as discussed previously.
So, here is a first step to fix this, doing various things:
INTERFACE CHANGES:
* drop the -s (sort) option, it clashes with -s (section)
* introduce -N (numerical sort) instead
* drop the -c (cat) option, it clashes with -c (copy to stdout)
* rename it to the usual -s (section)
* drop the -a (arch) option, it clashes with -a (all)
* rename it to the usual -S (subsection)
* drop the -t (search type), it clashes with -t (use troff)
* use a macro= syntax instead, as discussed
* drop -e (exact) and -r (regex), it belongs to each query phrase
* use a macro== and macro=~ syntax instead
The new syntax is:
apropos [-IN] [-s section] [-S arch] query_phrase [...]
query_phrase ::= [[macro[,...]](=|==|=~)]query_value
Multiple query phrases are not yet implemented, but they are to
be or'ed in the future.
The operators are:
= substring match
== exact match
=~ regex match
If no operator is given, = is assumed.
Multiple macros can be given, joined with commas, they are or'ed.
If no macro is given, "Nm,Nd" is assumed.
STRUCTURAL CLEANUP:
* collect the common defines for mandocdb and apropos in mandocdb.h
* name the TYPE_ bitfield constants according to the new interface
* drop types and match from the global struct opts
because these are local to each query phrase
* drop the enum sort; for an alternative, a bool is sufficient
* drop the local dbf and idxf strings, use those in the global struct
This is lightly tested, but given that we are not yet in production
and lots of heavy changes are needed in after this, i consider
light testing sufficient.
OK?
Of course, when committing, i'm going to change the manual, too.
And by the way, this patch is +144 -175 :-).
Yours,
Ingo
ischwarze@isnote $ apropos An=Gray
AN(4) - Aironet Communications 4500/4800 IEEE 802.11FH/b wireless network device
APS(4) - ThinkPad Active Protection System accelerometer
BWI(4) - Broadcom AirForce IEEE 802.11b/g wireless network device
ET(4) - Agere/LSI ET1310 10/100/Gigabit Ethernet device
ETPHY(4) - Agere/LSI ET1011 TruePHY Gigabit Ethernet PHY
JME(4) - JMicron JMC250/JMC260 10/100/Gigabit Ethernet device
JMPHY(4) - JMicron JMP202/JMP211 10/100/Gigabit Ethernet PHY
MOSCOM(4) - MosChip Semiconductor MCS7703 based USB serial adapter
NFE(4) - NVIDIA nForce MCP 10/100/Gigabit Ethernet device
RTW(4) - Realtek RTL8180L IEEE 802.11b wireless network device
SPDMEM(4) - Serial Presence Detect memory
UARK(4) - Arkmicro Technologies ARK3116 based USB serial adapter
UCHCOM(4) - WinChipHead CH341/340 based USB serial adapter
UDAV(4) - Davicom DM9601 10/100 USB Ethernet device
UMSM(4) - Qualcomm MSM modem device
USLCOM(4) - Silicon Laboratories CP2101/CP2102 based USB serial adapter
ZYD(4) - ZyDAS ZD1211/ZD1211B USB IEEE 802.11b/g wireless network device
ischwarze@isnote $ apropos Xr==et.4
PCI(4) - introduction to PCI bus support
ischwarze@isnote $ apropos Nm=~^b.e$
BCE(4) - Broadcom BCM4401 10/100 Ethernet device
BGE(4) - Broadcom BCM57xx/BCM590x 10/100/Gigabit Ethernet device
--- apropos.c.orig
+++ apropos.c
@@ -1,6 +1,7 @@
/* $Id: apropos.c,v 1.2 2011/10/09 17:59:56 schwarze Exp $ */
/*
-* Copyright (c) 2011 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2011 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
@@ -31,44 +32,21 @@
#include <db.h>
#include "mandoc.h"
+#include "mandocdb.h"
#define MAXRESULTS 256
-/* Bit-fields. See mandocdb.8. */
-
-#define TYPE_NAME 0x01
-#define TYPE_FUNCTION 0x02
-#define TYPE_UTILITY 0x04
-#define TYPE_INCLUDES 0x08
-#define TYPE_VARIABLE 0x10
-#define TYPE_STANDARD 0x20
-#define TYPE_AUTHOR 0x40
-#define TYPE_CONFIG 0x80
-#define TYPE_DESC 0x100
-#define TYPE_XREF 0x200
-#define TYPE_PATH 0x400
-#define TYPE_ENV 0x800
-#define TYPE_ERR 0x1000
-
enum match {
MATCH_SUBSTR = 0,
MATCH_REGEX,
MATCH_EXACT
};
-enum sort {
- SORT_TITLE = 0,
- SORT_CAT,
- SORT__MAX
-};
-
struct opts {
- enum sort sort; /* output sorting */
+ const char *sec; /* restrict to manual section */
const char *arch; /* restrict to architecture */
- const char *cat; /* restrict to category */
- int types; /* only types in bitmask */
int insens; /* case-insensitive match */
- enum match match; /* match type */
+ int numerical; /* sort output by section */
};
struct type {
@@ -78,7 +56,7 @@ struct type {
struct rec {
char *file; /* file in file-system */
- char *cat; /* category (3p, 3, etc.) */
+ char *sec; /* section (3p, 3, etc.) */
char *title; /* title (FOO, etc.) */
char *arch; /* arch (or empty string) */
char *desc; /* description (from Nd) */
@@ -86,12 +64,12 @@ struct rec {
};
struct res {
+ char *sec; /* manual section */
+ char *title; /* manual title */
char *arch; /* architecture */
char *desc; /* free-form description */
char *keyword; /* matched keyword */
int types; /* bitmask of field selectors */
- char *cat; /* manual section */
- char *title; /* manual section */
char *uri; /* formatted uri of file */
recno_t rec; /* unique id of underlying manual */
/*
@@ -111,26 +89,22 @@ struct state {
const char *idxf; /* index name */
};
-static const char * const sorts[SORT__MAX] = {
- "cat", /* SORT_CAT */
- "title", /* SORT_TITLE */
-};
-
-static const struct type types[] = {
- { TYPE_NAME, "name" },
- { TYPE_FUNCTION, "func" },
- { TYPE_UTILITY, "utility" },
- { TYPE_INCLUDES, "incl" },
- { TYPE_VARIABLE, "var" },
- { TYPE_STANDARD, "stand" },
- { TYPE_AUTHOR, "auth" },
- { TYPE_CONFIG, "conf" },
- { TYPE_DESC, "desc" },
- { TYPE_XREF, "xref" },
- { TYPE_PATH, "path" },
- { TYPE_ENV, "env" },
- { TYPE_ERR, "err" },
- { INT_MAX, "all" },
+static const struct type typemap[] = {
+ { TYPE_An, "An" },
+ { TYPE_Cd, "Cd" },
+ { TYPE_Er, "Er" },
+ { TYPE_Ev, "Ev" },
+ { TYPE_Fn, "Fn" },
+ { TYPE_Fn, "Fo" },
+ { TYPE_In, "In" },
+ { TYPE_Nd, "Nd" },
+ { TYPE_Nm, "Nm" },
+ { TYPE_Pa, "Pa" },
+ { TYPE_St, "St" },
+ { TYPE_Va, "Va" },
+ { TYPE_Va, "Vt" },
+ { TYPE_Xr, "Xr" },
+ { INT_MAX, "any" },
{ 0, NULL }
};
@@ -138,7 +112,7 @@ static void buf_alloc(char **, size_t *, size_t);
static void buf_dup(struct mchars *, char **, const char *);
static void buf_redup(struct mchars *, char **,
size_t *, const char *);
-static int sort_cat(const void *, const void *);
+static int sort_sec(const void *, const void *);
static int sort_title(const void *, const void *);
static int state_getrecord(struct state *,
recno_t, struct rec *);
@@ -153,10 +127,9 @@ int
apropos(int argc, char *argv[])
{
BTREEINFO info;
- int ch, i, rc;
- const char *dbf, *idxf;
+ int ch, rc;
struct state state;
- char *q, *v;
+ char *q;
struct opts opts;
extern int optind;
extern char *optarg;
@@ -164,8 +137,8 @@ apropos(int argc, char *argv[])
memset(&opts, 0, sizeof(struct opts));
memset(&state, 0, sizeof(struct state));
- dbf = "mandoc.db";
- idxf = "mandoc.index";
+ state.dbf = MANDOC_DB;
+ state.idxf = MANDOC_IDX;
q = NULL;
rc = EXIT_FAILURE;
@@ -175,56 +148,20 @@ apropos(int argc, char *argv[])
else
++progname;
- opts.match = MATCH_SUBSTR;
-
- while (-1 != (ch = getopt(argc, argv, "a:c:eIrs:t:")))
+ while (-1 != (ch = getopt(argc, argv, "INS:s:")))
switch (ch) {
- case ('a'):
- opts.arch = optarg;
- break;
- case ('c'):
- opts.cat = optarg;
- break;
- case ('e'):
- opts.match = MATCH_EXACT;
- break;
case ('I'):
opts.insens = 1;
break;
- case ('r'):
- opts.match = MATCH_REGEX;
+ case ('N'):
+ opts.numerical = 1;
+ break;
+ case ('S'):
+ opts.arch = optarg;
break;
case ('s'):
- for (i = 0; i < SORT__MAX; i++) {
- if (strcmp(optarg, sorts[i]))
- continue;
- opts.sort = (enum sort)i;
- break;
- }
-
- if (i < SORT__MAX)
- break;
-
- fprintf(stderr, "%s: Bad sort\n", optarg);
- return(EXIT_FAILURE);
- case ('t'):
- while (NULL != (v = strsep(&optarg, ","))) {
- if ('\0' == *v)
- continue;
- for (i = 0; types[i].mask; i++) {
- if (strcmp(types[i].name, v))
- continue;
- break;
- }
- if (0 == types[i].mask)
- break;
- opts.types |= types[i].mask;
- }
- if (NULL == v)
- break;
-
- fprintf(stderr, "%s: Bad type\n", v);
- return(EXIT_FAILURE);
+ opts.sec = optarg;
+ break;
default:
usage();
return(EXIT_FAILURE);
@@ -239,9 +176,6 @@ apropos(int argc, char *argv[])
} else
q = *argv;
- if (0 == opts.types)
- opts.types = TYPE_NAME | TYPE_DESC;
-
/*
* Configure databases.
* The keyword database is a btree that allows for duplicate
@@ -252,15 +186,15 @@ apropos(int argc, char *argv[])
memset(&info, 0, sizeof(BTREEINFO));
info.flags = R_DUP;
- state.db = dbopen(dbf, O_RDONLY, 0, DB_BTREE, &info);
+ state.db = dbopen(state.dbf, O_RDONLY, 0, DB_BTREE, &info);
if (NULL == state.db) {
- perror(dbf);
+ perror(state.dbf);
goto out;
}
- state.idx = dbopen(idxf, O_RDONLY, 0, DB_RECNO, NULL);
+ state.idx = dbopen(state.idxf, O_RDONLY, 0, DB_RECNO, NULL);
if (NULL == state.idx) {
- perror(idxf);
+ perror(state.idxf);
goto out;
}
@@ -280,9 +214,9 @@ out:
static int
state_search(struct state *p, const struct opts *opts, char *q)
{
- int leaf, root, len, ch, dflag, rc;
+ int i, leaf, root, len, ch, dflag, rc, types;
struct mchars *mc;
- char *buf;
+ char *buf, *qkey, *qval;
size_t bufsz;
recno_t rec;
uint32_t fl;
@@ -292,6 +226,7 @@ state_search(struct state *p, const struct opts *opts, char *q)
regex_t *regp;
char filebuf[10];
struct rec record;
+ enum match match_method;
rc = 0;
root = leaf = -1;
@@ -302,29 +237,50 @@ state_search(struct state *p, const struct opts *opts, char *q)
regp = NULL;
/*
+ * Determine the search types.
+ */
+
+ types = 0;
+ if (NULL == (qval = strchr(q, '=')))
+ qval = q;
+ else {
+ *qval++ = '\0';
+ while (NULL != (qkey = strsep(&q, ","))) {
+ i = 0;
+ while (typemap[i].mask &&
+ strcmp(typemap[i].name, qkey))
+ i++;
+ types |= typemap[i].mask;
+ }
+ }
+ if (0 == types)
+ types = TYPE_Nm | TYPE_Nd;
+
+ /*
* Configure how we scan through results to see if we match:
* whether by regexp or exact matches.
*/
- switch (opts->match) {
- case (MATCH_REGEX):
+ switch (*qval) {
+ case ('~'):
+ match_method = MATCH_REGEX;
ch = REG_EXTENDED | REG_NOSUB |
(opts->insens ? REG_ICASE : 0);
-
- if (0 != regcomp(®, q, ch)) {
- fprintf(stderr, "%s: Bad pattern\n", q);
+ if (0 != regcomp(®, ++qval, ch)) {
+ fprintf(stderr, "%s: Bad pattern\n", qval);
return(0);
}
-
regp = ®
dflag = R_FIRST;
break;
- case (MATCH_EXACT):
- key.data = q;
- key.size = strlen(q) + 1;
+ case ('='):
+ match_method = MATCH_EXACT;
+ key.data = ++qval;
+ key.size = strlen(qval) + 1;
dflag = R_CURSOR;
break;
default:
+ match_method = MATCH_SUBSTR;
dflag = R_FIRST;
break;
}
@@ -357,24 +313,24 @@ state_search(struct state *p, const struct opts *opts, char *q)
fl = *(uint32_t *)val.data;
- if ( ! (fl & opts->types))
+ if ( ! (fl & types))
continue;
- switch (opts->match) {
+ switch (match_method) {
case (MATCH_REGEX):
if (regexec(regp, buf, 0, NULL, 0))
continue;
break;
case (MATCH_EXACT):
- if (opts->insens && strcasecmp(buf, q))
+ if (opts->insens && strcasecmp(buf, qval))
goto send;
- if ( ! opts->insens && strcmp(buf, q))
+ if ( ! opts->insens && strcmp(buf, qval))
goto send;
break;
default:
- if (opts->insens && NULL == strcasestr(buf, q))
+ if (opts->insens && NULL == strcasestr(buf, qval))
continue;
- if ( ! opts->insens && NULL == strstr(buf, q))
+ if ( ! opts->insens && NULL == strstr(buf, qval))
continue;
break;
}
@@ -391,7 +347,7 @@ state_search(struct state *p, const struct opts *opts, char *q)
/* If we're in a different section, skip... */
- if (opts->cat && strcasecmp(opts->cat, record.cat))
+ if (opts->sec && strcasecmp(opts->sec, record.sec))
continue;
if (opts->arch && strcasecmp(opts->arch, record.arch))
continue;
@@ -431,7 +387,7 @@ state_search(struct state *p, const struct opts *opts, char *q)
buf_dup(mc, &res[len].keyword, buf);
buf_dup(mc, &res[len].uri, filebuf);
- buf_dup(mc, &res[len].cat, record.cat);
+ buf_dup(mc, &res[len].sec, record.sec);
buf_dup(mc, &res[len].arch, record.arch);
buf_dup(mc, &res[len].title, record.title);
buf_dup(mc, &res[len].desc, record.desc);
@@ -454,8 +410,8 @@ state_search(struct state *p, const struct opts *opts, char *q)
send:
/* Sort our results. */
- if (SORT_CAT == opts->sort)
- qsort(res, len, sizeof(struct res), sort_cat);
+ if (opts->numerical)
+ qsort(res, len, sizeof(struct res), sort_sec);
else
qsort(res, len, sizeof(struct res), sort_title);
@@ -465,7 +421,7 @@ out:
for (len-- ; len >= 0; len--) {
free(res[len].keyword);
free(res[len].title);
- free(res[len].cat);
+ free(res[len].sec);
free(res[len].arch);
free(res[len].desc);
free(res[len].uri);
@@ -589,7 +545,7 @@ state_output(const struct res *res, int sz)
for (i = 0; i < sz; i++)
printf("%s(%s%s%s) - %s\n", res[i].title,
- res[i].cat,
+ res[i].sec,
*res[i].arch ? "/" : "",
*res[i].arch ? res[i].arch : "",
res[i].desc);
@@ -600,11 +556,9 @@ usage(void)
{
fprintf(stderr, "usage: %s "
- "[-eIr] "
- "[-a arch] "
- "[-c cat] "
- "[-s sort] "
- "[-t type[,...]] "
+ "[-IN] "
+ "[-S subsection] "
+ "[-s section] "
"key\n", progname);
}
@@ -629,8 +583,8 @@ state_getrecord(struct state *p, recno_t rec, struct rec *rp)
if ((sz = strlen(rp->file) + 1) >= val.size)
goto err;
- rp->cat = (char *)val.data + (int)sz;
- if ((sz += strlen(rp->cat) + 1) >= val.size)
+ rp->sec = (char *)val.data + (int)sz;
+ if ((sz += strlen(rp->sec) + 1) >= val.size)
goto err;
rp->title = (char *)val.data + (int)sz;
@@ -658,12 +612,12 @@ sort_title(const void *p1, const void *p2)
}
static int
-sort_cat(const void *p1, const void *p2)
+sort_sec(const void *p1, const void *p2)
{
int rc;
- rc = strcmp(((const struct res *)p1)->cat,
- ((const struct res *)p2)->cat);
+ rc = strcmp(((const struct res *)p1)->sec,
+ ((const struct res *)p2)->sec);
return(0 == rc ? sort_title(p1, p2) : rc);
}
--- mandocdb.c.orig
+++ mandocdb.c
@@ -29,28 +29,11 @@
#include "man.h"
#include "mdoc.h"
#include "mandoc.h"
+#include "mandocdb.h"
-#define MANDOC_DB "mandoc.db"
-#define MANDOC_IDX "mandoc.index"
#define MANDOC_BUFSZ BUFSIZ
#define MANDOC_SLOP 1024
-/* Bit-fields. See mandocdb.8. */
-
-#define TYPE_NAME 0x01
-#define TYPE_FUNCTION 0x02
-#define TYPE_UTILITY 0x04
-#define TYPE_INCLUDES 0x08
-#define TYPE_VARIABLE 0x10
-#define TYPE_STANDARD 0x20
-#define TYPE_AUTHOR 0x40
-#define TYPE_CONFIG 0x80
-#define TYPE_DESC 0x100
-#define TYPE_XREF 0x200
-#define TYPE_PATH 0x400
-#define TYPE_ENV 0x800
-#define TYPE_ERR 0x1000
-
/* Tiny list for files. No need to bring in QUEUE. */
struct of {
@@ -719,7 +702,7 @@ pmdoc_An(MDOC_ARGS)
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_AUTHOR);
+ hash_put(hash, buf, TYPE_An);
}
static void
@@ -780,7 +763,7 @@ pmdoc_Fd(MDOC_ARGS)
buf_appendb(buf, start, (size_t)(end - start + 1));
buf_appendb(buf, "", 1);
- hash_put(hash, buf, TYPE_INCLUDES);
+ hash_put(hash, buf, TYPE_In);
}
/* ARGSUSED */
@@ -792,7 +775,7 @@ pmdoc_Cd(MDOC_ARGS)
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_CONFIG);
+ hash_put(hash, buf, TYPE_Cd);
}
/* ARGSUSED */
@@ -806,7 +789,7 @@ pmdoc_In(MDOC_ARGS)
return;
buf_append(buf, n->child->string);
- hash_put(hash, buf, TYPE_INCLUDES);
+ hash_put(hash, buf, TYPE_In);
}
/* ARGSUSED */
@@ -832,7 +815,7 @@ pmdoc_Fn(MDOC_ARGS)
cp++;
buf_append(buf, cp);
- hash_put(hash, buf, TYPE_FUNCTION);
+ hash_put(hash, buf, TYPE_Fn);
}
/* ARGSUSED */
@@ -846,7 +829,7 @@ pmdoc_St(MDOC_ARGS)
return;
buf_append(buf, n->child->string);
- hash_put(hash, buf, TYPE_STANDARD);
+ hash_put(hash, buf, TYPE_St);
}
/* ARGSUSED */
@@ -865,7 +848,7 @@ pmdoc_Xr(MDOC_ARGS)
} else
buf_appendb(buf, ".", 2);
- hash_put(hash, buf, TYPE_XREF);
+ hash_put(hash, buf, TYPE_Xr);
}
/* ARGSUSED */
@@ -902,7 +885,7 @@ pmdoc_Vt(MDOC_ARGS)
buf_appendb(buf, start, sz);
buf_appendb(buf, "", 1);
- hash_put(hash, buf, TYPE_VARIABLE);
+ hash_put(hash, buf, TYPE_Va);
}
/* ARGSUSED */
@@ -916,7 +899,7 @@ pmdoc_Fo(MDOC_ARGS)
return;
buf_append(buf, n->child->string);
- hash_put(hash, buf, TYPE_FUNCTION);
+ hash_put(hash, buf, TYPE_Fn);
}
@@ -931,7 +914,7 @@ pmdoc_Nd(MDOC_ARGS)
buf_appendmdoc(dbuf, n->child, 1);
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_DESC);
+ hash_put(hash, buf, TYPE_Nd);
}
/* ARGSUSED */
@@ -943,7 +926,7 @@ pmdoc_Er(MDOC_ARGS)
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_ERR);
+ hash_put(hash, buf, TYPE_Er);
}
/* ARGSUSED */
@@ -955,7 +938,7 @@ pmdoc_Ev(MDOC_ARGS)
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_ENV);
+ hash_put(hash, buf, TYPE_Ev);
}
/* ARGSUSED */
@@ -967,7 +950,7 @@ pmdoc_Pa(MDOC_ARGS)
return;
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_PATH);
+ hash_put(hash, buf, TYPE_Pa);
}
/* ARGSUSED */
@@ -977,7 +960,7 @@ pmdoc_Nm(MDOC_ARGS)
if (SEC_NAME == n->sec) {
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_NAME);
+ hash_put(hash, buf, TYPE_Nm);
return;
} else if (SEC_SYNOPSIS != n->sec || MDOC_HEAD != n->type)
return;
@@ -986,7 +969,7 @@ pmdoc_Nm(MDOC_ARGS)
buf_append(buf, m->name);
buf_appendmdoc(buf, n->child, 0);
- hash_put(hash, buf, TYPE_UTILITY);
+ hash_put(hash, buf, TYPE_Nm);
}
static void
@@ -1116,7 +1099,7 @@ pman_node(MAN_ARGS)
buf_appendb(buf, start, sz);
buf_appendb(buf, "", 1);
- hash_put(hash, buf, TYPE_NAME);
+ hash_put(hash, buf, TYPE_Nm);
if (' ' == start[(int)sz]) {
start += (int)sz + 1;
@@ -1155,7 +1138,7 @@ pman_node(MAN_ARGS)
buf_appendb(dbuf, start, sz);
buf_appendb(buf, start, sz);
- hash_put(hash, buf, TYPE_DESC);
+ hash_put(hash, buf, TYPE_Nd);
}
}
--- /dev/null
+++ mandocdb.h
@@ -0,0 +1,32 @@
+/* $Id$ */
+/*
+ * Copyright (c) 2011 Kristaps Dzonsons <kristaps@bsd.lv>
+ *
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+
+#define MANDOC_DB "mandoc.db"
+#define MANDOC_IDX "mandoc.index"
+
+#define TYPE_An 0x01
+#define TYPE_Cd 0x02
+#define TYPE_Er 0x04
+#define TYPE_Ev 0x08
+#define TYPE_Fn 0x10
+#define TYPE_In 0x20
+#define TYPE_Nd 0x40
+#define TYPE_Nm 0x100
+#define TYPE_Pa 0x200
+#define TYPE_St 0x400
+#define TYPE_Va 0x1000
+#define TYPE_Xr 0x2000
--
To unsubscribe send an email to tech+unsubscribe@mdocml.bsd.lv
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: overhaul apropos(1) interface
2011-11-09 1:30 overhaul apropos(1) interface Ingo Schwarze
@ 2011-11-09 10:19 ` Kristaps Dzonsons
2011-11-12 23:54 ` Ingo Schwarze
1 sibling, 0 replies; 6+ messages in thread
From: Kristaps Dzonsons @ 2011-11-09 10:19 UTC (permalink / raw)
To: tech
On 09/11/2011 02:30, Ingo Schwarze wrote:
> Hi,
>
> the interface of the apropos utility in the mandoc package
> is very different from traditional apropos, and the -t option
> hinders logical extensions, as discussed previously.
> So, here is a first step to fix this, doing various things:
>
> INTERFACE CHANGES:
> * drop the -s (sort) option, it clashes with -s (section)
> * introduce -N (numerical sort) instead
> * drop the -c (cat) option, it clashes with -c (copy to stdout)
> * rename it to the usual -s (section)
> * drop the -a (arch) option, it clashes with -a (all)
> * rename it to the usual -S (subsection)
> * drop the -t (search type), it clashes with -t (use troff)
> * use a macro= syntax instead, as discussed
> * drop -e (exact) and -r (regex), it belongs to each query phrase
> * use a macro== and macro=~ syntax instead
>
> The new syntax is:
>
> apropos [-IN] [-s section] [-S arch] query_phrase [...]
>
> query_phrase ::= [[macro[,...]](=|==|=~)]query_value
>
> Multiple query phrases are not yet implemented, but they are to
> be or'ed in the future.
>
> The operators are:
>
> = substring match
> == exact match
> =~ regex match
>
> If no operator is given, = is assumed.
> Multiple macros can be given, joined with commas, they are or'ed.
> If no macro is given, "Nm,Nd" is assumed.
>
> STRUCTURAL CLEANUP:
> * collect the common defines for mandocdb and apropos in mandocdb.h
> * name the TYPE_ bitfield constants according to the new interface
> * drop types and match from the global struct opts
> because these are local to each query phrase
> * drop the enum sort; for an alternative, a bool is sufficient
> * drop the local dbf and idxf strings, use those in the global struct
>
> This is lightly tested, but given that we are not yet in production
> and lots of heavy changes are needed in after this, i consider
> light testing sufficient.
>
> OK?
>
> Of course, when committing, i'm going to change the manual, too.
>
> And by the way, this patch is +144 -175 :-).
Unfortunately, this comes in after an overhaul I checked in last night.
However, that overhaul overwhelmingly concerns the search function
itself, not the interface, so there are few conflicts -- just name
changes. I'll piece this into the new version soon.
K.
--
To unsubscribe send an email to tech+unsubscribe@mdocml.bsd.lv
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: overhaul apropos(1) interface
2011-11-09 1:30 overhaul apropos(1) interface Ingo Schwarze
2011-11-09 10:19 ` Kristaps Dzonsons
@ 2011-11-12 23:54 ` Ingo Schwarze
2011-11-13 0:07 ` Kristaps Dzonsons
1 sibling, 1 reply; 6+ messages in thread
From: Ingo Schwarze @ 2011-11-12 23:54 UTC (permalink / raw)
To: tech
Hi,
here is a new version of my overhaul after Kristaps' changes.
The patch given below is not likely to apply cleanly to anybody's
repo, it requires the apropos_db.* rename first. It is meant
for quick review; if you agree with the direction, i'll figure
out how to get this in cleanly.
So, here is what the overhaul does:
* Remove -I from the main program.
Logically, that's not a global option,
but a per-expression thingy.
* Sort the arguments of exprcomp.
All the world uses (argc, argv),
why should we suddenly go for (argv, argc)?
* I agree with dropping MATCH_EXACT:
That's rarely needed and can easily be constructed
as MATCH_REGEX with ^...$.
* For the same reason, let's drop case-sensitive MATCH_STR:
It's rarely needed and can easily be constructed with MATCH_REGEX.
* Keeping MATCH_STR seems OK: It is needed as the default
when the type is unspecified.
* The MATCH_REGEXCASE enum item is unused already now,
since the iflag is compiled into the regex object.
So drop that one as well.
* Only two enum items remain; that's better and more easily
expressed by a single boolean integer (expr.regex).
* The exprexec() function requires a mask argument,
or all search keys will act as "any" (important bug fix
along the way!).
The most massive changes are in exprcomp().
I strongly dislike the proposed interface.
It is cumbersome and requires too much typing.
The -eq and -re arguments are exceedingly ugly, and the
implementation is hard to get right - if i remember
correctly, not all cases work properly right now.
Thus, i have simplified my interface proposal to just this:
apropos [-s section] [-S arch] query_phrase [...]
query_phrase ::= [[macro[,...]](=|~)]query_value
So, the value to be searched for can optionally
be preceded by '=' (for string search) or '~' (for regex search),
and that can optionally be preceded by one or more macro names,
joined by commas. Including "i" among the macros switches
regex searches to case-insensitive and has no effect on
string searches.
OK?
Ingo
--- apropos.c.orig
+++ apropos.c
@@ -33,7 +33,7 @@ static char *progname;
int
apropos(int argc, char *argv[])
{
- int ch, cs;
+ int ch;
struct opts opts;
struct expr *e;
extern int optind;
@@ -47,9 +47,7 @@ apropos(int argc, char *argv[])
else
++progname;
- cs = 0;
-
- while (-1 != (ch = getopt(argc, argv, "S:s:I")))
+ while (-1 != (ch = getopt(argc, argv, "S:s:")))
switch (ch) {
case ('S'):
opts.arch = optarg;
@@ -57,9 +55,6 @@ apropos(int argc, char *argv[])
case ('s'):
opts.cat = optarg;
break;
- case ('I'):
- cs = 1;
- break;
default:
usage();
return(EXIT_FAILURE);
@@ -71,7 +66,7 @@ apropos(int argc, char *argv[])
if (0 == argc)
return(EXIT_SUCCESS);
- if (NULL == (e = exprcomp(cs, argv, argc))) {
+ if (NULL == (e = exprcomp(argc, argv))) {
fprintf(stderr, "Bad expression\n");
return(EXIT_FAILURE);
}
--- apropos_db.c.orig
+++ apropos_db.c
@@ -1,6 +1,7 @@
/* $Id: db.c,v 1.1 2011/11/09 01:24:23 kristaps Exp $ */
/*
* Copyright (c) 2011 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
@@ -31,15 +32,8 @@
#include "apropos_db.h"
#include "mandoc.h"
-enum match {
- MATCH_REGEX,
- MATCH_REGEXCASE,
- MATCH_STR,
- MATCH_STRCASE
-};
-
struct expr {
- enum match match;
+ int regex;
int mask;
char *v;
regex_t re;
@@ -71,7 +65,7 @@ static const struct type types[] = {
static DB *btree_open(void);
static int btree_read(const DBT *, const struct mchars *, char **);
-static int exprexec(const struct expr *, char *);
+static int exprexec(const struct expr *, char *, int);
static DB *index_open(void);
static int index_read(const DBT *, const DBT *,
const struct mchars *, struct rec *);
@@ -368,7 +362,7 @@ apropos_search(const struct opts *opts, const struct expr *expr,
if ( ! btree_read(&key, mc, &buf))
break;
- if ( ! exprexec(expr, buf))
+ if ( ! exprexec(expr, buf, *(int *)val.data))
continue;
memcpy(&rec, val.data + 4, sizeof(recno_t));
@@ -460,55 +454,55 @@ out:
}
struct expr *
-exprcomp(int cs, char *argv[], int argc)
+exprcomp(int argc, char *argv[])
{
struct expr *p;
struct expr e;
- int i, pos, ch;
+ char *key;
+ int i, icase;
- pos = 0;
-
- if (pos > argc)
+ if (0 >= argc)
return(NULL);
- for (i = 0; 0 != types[i].mask; i++)
- if (0 == strcmp(types[i].name, argv[pos]))
- break;
-
- if (0 == (e.mask = types[i].mask))
- return(NULL);
-
- if (++pos > argc--)
- return(NULL);
+ /*
+ * Choose regex or substring match.
+ */
- if ('-' != *argv[pos])
- e.match = cs ? MATCH_STRCASE : MATCH_STR;
- else if (0 == strcmp("-eq", argv[pos]))
- e.match = cs ? MATCH_STRCASE : MATCH_STR;
- else if (0 == strcmp("-ieq", argv[pos]))
- e.match = MATCH_STRCASE;
- else if (0 == strcmp("-re", argv[pos]))
- e.match = cs ? MATCH_REGEXCASE : MATCH_REGEX;
- else if (0 == strcmp("-ire", argv[pos]))
- e.match = MATCH_REGEXCASE;
- else
- return(NULL);
+ if (NULL == (e.v = strpbrk(*argv, "=~"))) {
+ e.regex = 0;
+ e.v = *argv;
+ } else {
+ e.regex = '~' == *e.v;
+ *e.v++ = '\0';
+ }
- if ('-' == *argv[pos])
- pos++;
+ /*
+ * Determine the record types to search for.
+ */
+
+ icase = 0;
+ e.mask = 0;
+ if (*argv < e.v) {
+ while (NULL != (key = strsep(argv, ","))) {
+ if ('i' == key[0] && '\0' == key[1]) {
+ icase = REG_ICASE;
+ continue;
+ }
+ i = 0;
+ while (types[i].mask &&
+ strcmp(types[i].name, key))
+ i++;
+ e.mask |= types[i].mask;
+ }
+ }
+ if (0 == e.mask)
+ e.mask = TYPE_Nm | TYPE_Nd;
- if (pos > argc--)
+ if (e.regex &&
+ regcomp(&e.re, e.v, REG_EXTENDED | REG_NOSUB | icase))
return(NULL);
- e.v = mandoc_strdup(argv[pos]);
-
- if (MATCH_REGEX == e.match || MATCH_REGEXCASE == e.match) {
- ch = REG_EXTENDED | REG_NOSUB;
- if (MATCH_REGEXCASE == e.match)
- ch |= REG_ICASE;
- if (regcomp(&e.re, e.v, ch))
- return(NULL);
- }
+ e.v = mandoc_strdup(e.v);
p = mandoc_calloc(1, sizeof(struct expr));
memcpy(p, &e, sizeof(struct expr));
@@ -522,7 +516,7 @@ exprfree(struct expr *p)
if (NULL == p)
return;
- if (MATCH_REGEX == p->match)
+ if (p->regex)
regfree(&p->re);
free(p->v);
@@ -530,14 +524,14 @@ exprfree(struct expr *p)
}
static int
-exprexec(const struct expr *p, char *cp)
+exprexec(const struct expr *p, char *cp, int mask)
{
- if (MATCH_STR == p->match)
- return(0 == strcmp(p->v, cp));
- else if (MATCH_STRCASE == p->match)
- return(0 == strcasecmp(p->v, cp));
+ if ( ! (mask & p->mask))
+ return(0);
- assert(MATCH_REGEX == p->match);
- return(0 == regexec(&p->re, cp, 0, NULL, 0));
+ if (p->regex)
+ return(0 == regexec(&p->re, cp, 0, NULL, 0));
+ else
+ return(NULL != strcasestr(cp, p->v));
}
--- apropos_db.h.orig
+++ apropos_db.h
@@ -49,7 +49,7 @@ void apropos_search(const struct opts *,
const struct expr *, void *,
void (*)(struct rec *, size_t, void *));
-struct expr *exprcomp(int, char *[], int);
+struct expr *exprcomp(int, char *[]);
void exprfree(struct expr *);
__END_DECLS
--
To unsubscribe send an email to tech+unsubscribe@mdocml.bsd.lv
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: overhaul apropos(1) interface
2011-11-12 23:54 ` Ingo Schwarze
@ 2011-11-13 0:07 ` Kristaps Dzonsons
2011-11-13 0:39 ` Ingo Schwarze
0 siblings, 1 reply; 6+ messages in thread
From: Kristaps Dzonsons @ 2011-11-13 0:07 UTC (permalink / raw)
To: tech; +Cc: Ingo Schwarze
On 13/11/2011 00:54, Ingo Schwarze wrote:
> Hi,
>
> here is a new version of my overhaul after Kristaps' changes.
> The patch given below is not likely to apply cleanly to anybody's
> repo, it requires the apropos_db.* rename first. It is meant
> for quick review; if you agree with the direction, i'll figure
> out how to get this in cleanly.
>
> So, here is what the overhaul does:
>
> * Remove -I from the main program.
> Logically, that's not a global option,
> but a per-expression thingy.
> * Sort the arguments of exprcomp.
> All the world uses (argc, argv),
> why should we suddenly go for (argv, argc)?
> * I agree with dropping MATCH_EXACT:
> That's rarely needed and can easily be constructed
> as MATCH_REGEX with ^...$.
> * For the same reason, let's drop case-sensitive MATCH_STR:
> It's rarely needed and can easily be constructed with MATCH_REGEX.
> * Keeping MATCH_STR seems OK: It is needed as the default
> when the type is unspecified.
> * The MATCH_REGEXCASE enum item is unused already now,
> since the iflag is compiled into the regex object.
> So drop that one as well.
> * Only two enum items remain; that's better and more easily
> expressed by a single boolean integer (expr.regex).
> * The exprexec() function requires a mask argument,
> or all search keys will act as "any" (important bug fix
> along the way!).
>
> The most massive changes are in exprcomp().
> I strongly dislike the proposed interface.
> It is cumbersome and requires too much typing.
> The -eq and -re arguments are exceedingly ugly, and the
> implementation is hard to get right - if i remember
> correctly, not all cases work properly right now.
>
> Thus, i have simplified my interface proposal to just this:
>
> apropos [-s section] [-S arch] query_phrase [...]
>
> query_phrase ::= [[macro[,...]](=|~)]query_value
>
> So, the value to be searched for can optionally
> be preceded by '=' (for string search) or '~' (for regex search),
> and that can optionally be preceded by one or more macro names,
> joined by commas. Including "i" among the macros switches
> regex searches to case-insensitive and has no effect on
> string searches.
Ingo,
I'm working on the final parts of this check-in, so please hold off on
this file! It's by no means finished; as mentioned in the source
checkin, I'll post to tech@ when the implementation is feature complete.
Consider, to gauge the complexity:
apropos Ar == foo -a Ar =~ baz
(I don't care at all about the connecting syntax, ==, etc., so long as
it's regular. Your notation is not extensible to case-insensitive
matching: can one extend to include these?)
Anyway, the "AND" is tricky: each file's evaluation state must be
retained for all keyword entries (which have no guarantee on ordering)
then post-operated. Thus, I'm maintaining evaluation trees during the
parse. The goal is to make the trivial case -- "Ar foo", say -- as fast
as in the simple implementation before. (I guarantee the trivial case
by the partial evaluation.)
Anyway, I anticipate a few days til I get the final checkins. The code
is not tricky in implementation (not much, anyway) and guarantees
arbitrary expressions with well-defined compute time.
I'm also not at all married to the filenames, but let's hold off for a
bit more as I get these chunks into place. apropos_db.c is fine by me,
for the record.
Thanks again,
Kristaps
--
To unsubscribe send an email to tech+unsubscribe@mdocml.bsd.lv
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: overhaul apropos(1) interface
2011-11-13 0:07 ` Kristaps Dzonsons
@ 2011-11-13 0:39 ` Ingo Schwarze
2011-11-13 9:23 ` Kristaps Dzonsons
0 siblings, 1 reply; 6+ messages in thread
From: Ingo Schwarze @ 2011-11-13 0:39 UTC (permalink / raw)
To: tech
Hi Kristaps,
Kristaps Dzonsons wrote on Sun, Nov 13, 2011 at 01:07:20AM +0100:
> I'm working on the final parts of this check-in, so please hold off
> on this file!
I cannot; i need to get this working in the OpenBSD tree ASAP,
i.e. tomorrow. The logical and and or are not critical for me
right now, i can do without those for some time, but i need a
working mandocdb-apropos toolchain in-tree to build upon.
Ports hackathons are short, and i want to get on as fast as
possible with the following critical path:
1) get mandocdb working on single directories
to produce databases in the macro-format,
i.e. with types like TYPE_An, TYPE_Cd
2) get apropos to work with that, such that the
databases can be used
3) get rid of the most glaring bugs
and complete backward compatibility
4) integrate the man.conf parser into mandocdb
such that mandocdb can walk the MANPATH
just like makewhatis(8) does
5) integrate the man.conf parser into apropos such that
mandoc-apropos gets useable as a real apropos replacement
6) rudimentary formatted page parsing in mandocdb
7) integrate mandocdb into pkg_add such that it gets
useable as a real makewhatis replacement
I realized in Ljubljana that this is more work than i thought
before, and i can't hold off, especially not this week, or i will
surely miss the 5.1 release.
Maybe you can hold off and bring in the and/or stuff
afterwards, i.e. after the patches i have posted so far?
You need not wait long, i would *gladly* push all my stuff
tomorrow in the morning, and then you have clean earth to
till and to adjust your work to it.
> It's by no means finished; as mentioned in the source checkin,
> I'll post to tech@ when the implementation is feature complete.
Sure, no doubt, but blocking system integration at a critical
time to implement optional, fancy features is a bad idea!
> Consider, to gauge the complexity:
>
> apropos Ar == foo -a Ar =~ baz
>
> (I don't care at all about the connecting syntax, ==, etc., so long
> as it's regular. Your notation is not extensible to
> case-insensitive matching: can one extend to include these?)
It is.
ischwarze@isnote $ cd /usr/share/man
ischwarze@isnote $ apropos.m Nm~^b.e$
BCE(4) - Broadcom BCM4401 10/100 Ethernet device
BGE(4) - Broadcom BCM57xx/BCM590x 10/100/Gigabit Ethernet device
ischwarze@isnote $ apropos.m Nm~^B.e$
ischwarze@isnote $ apropos.m Nm,i~^B.e$
BCE(4) - Broadcom BCM4401 10/100 Ethernet device
BGE(4) - Broadcom BCM57xx/BCM590x 10/100/Gigabit Ethernet device
That's not a mockup, that's what i'm running right now.
However, i propose that substring match always be case-insensitive.
Substring match is a simplification for daily wear and tear
and doesn't need such complexity. If case-insensitive substring
match drowns you in noise, just use regex matching - done.
> Anyway, the "AND" is tricky: each file's evaluation state must be
> retained for all keyword entries (which have no guarantee on
> ordering) then post-operated. Thus, I'm maintaining evaluation
> trees during the parse. The goal is to make the trivial case -- "Ar
> foo", say -- as fast as in the simple implementation before. (I
> guarantee the trivial case by the partial evaluation.)
>
> Anyway, I anticipate a few days til I get the final checkins. The
> code is not tricky in implementation (not much, anyway) and
> guarantees arbitrary expressions with well-defined compute time.
>
> I'm also not at all married to the filenames, but let's hold off for
> a bit more as I get these chunks into place. apropos_db.c is fine
> by me, for the record.
Sounds good, except that a hackathon is a bad time to make me wait.
Thanks for your understanding (and your work, above all!),
Ingo
--
To unsubscribe send an email to tech+unsubscribe@mdocml.bsd.lv
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: overhaul apropos(1) interface
2011-11-13 0:39 ` Ingo Schwarze
@ 2011-11-13 9:23 ` Kristaps Dzonsons
0 siblings, 0 replies; 6+ messages in thread
From: Kristaps Dzonsons @ 2011-11-13 9:23 UTC (permalink / raw)
To: tech
> I cannot; i need to get this working in the OpenBSD tree ASAP,
> i.e. tomorrow. The logical and and or are not critical for me
> right now, i can do without those for some time, but i need a
> working mandocdb-apropos toolchain in-tree to build upon.
Ingo,
Oi oi, a hackathon! In this case, I'm fine with your patch -- I wasn't
aware of the conditions. If exprexec() simply does the type-check and
the strcmp/regexec, then it works for the single-case expression: Ar ==
foo, Ar =~ foo, etc. This can then be further built up, later.
Further, if everything's kept within exprcomp() and exprexec(), it's
nice and neat to upgrade later.
Better yet, as we've agreed upon the tests (==, =~, etc.), the upgrade
to allow logical statements (and, or, etc.) is backwards-compatible.
So please go ahead with these changes; I'll merge and check my
improvements (they'll more or less re-write exprcomp() and exprexec(),
and add structs etc.) afteward.
Is that alright with you?
Thanks,
Kristaps
> Ports hackathons are short, and i want to get on as fast as
> possible with the following critical path:
>
> 1) get mandocdb working on single directories
> to produce databases in the macro-format,
> i.e. with types like TYPE_An, TYPE_Cd
> 2) get apropos to work with that, such that the
> databases can be used
> 3) get rid of the most glaring bugs
> and complete backward compatibility
> 4) integrate the man.conf parser into mandocdb
> such that mandocdb can walk the MANPATH
> just like makewhatis(8) does
> 5) integrate the man.conf parser into apropos such that
> mandoc-apropos gets useable as a real apropos replacement
> 6) rudimentary formatted page parsing in mandocdb
> 7) integrate mandocdb into pkg_add such that it gets
> useable as a real makewhatis replacement
>
> I realized in Ljubljana that this is more work than i thought
> before, and i can't hold off, especially not this week, or i will
> surely miss the 5.1 release.
>
> Maybe you can hold off and bring in the and/or stuff
> afterwards, i.e. after the patches i have posted so far?
> You need not wait long, i would *gladly* push all my stuff
> tomorrow in the morning, and then you have clean earth to
> till and to adjust your work to it.
>
>> It's by no means finished; as mentioned in the source checkin,
>> I'll post to tech@ when the implementation is feature complete.
>
> Sure, no doubt, but blocking system integration at a critical
> time to implement optional, fancy features is a bad idea!
>
>> Consider, to gauge the complexity:
>>
>> apropos Ar == foo -a Ar =~ baz
>>
>> (I don't care at all about the connecting syntax, ==, etc., so long
>> as it's regular. Your notation is not extensible to
>> case-insensitive matching: can one extend to include these?)
>
> It is.
>
> ischwarze@isnote $ cd /usr/share/man
> ischwarze@isnote $ apropos.m Nm~^b.e$
> BCE(4) - Broadcom BCM4401 10/100 Ethernet device
> BGE(4) - Broadcom BCM57xx/BCM590x 10/100/Gigabit Ethernet device
> ischwarze@isnote $ apropos.m Nm~^B.e$
> ischwarze@isnote $ apropos.m Nm,i~^B.e$
> BCE(4) - Broadcom BCM4401 10/100 Ethernet device
> BGE(4) - Broadcom BCM57xx/BCM590x 10/100/Gigabit Ethernet device
>
> That's not a mockup, that's what i'm running right now.
>
> However, i propose that substring match always be case-insensitive.
> Substring match is a simplification for daily wear and tear
> and doesn't need such complexity. If case-insensitive substring
> match drowns you in noise, just use regex matching - done.
>
>> Anyway, the "AND" is tricky: each file's evaluation state must be
>> retained for all keyword entries (which have no guarantee on
>> ordering) then post-operated. Thus, I'm maintaining evaluation
>> trees during the parse. The goal is to make the trivial case -- "Ar
>> foo", say -- as fast as in the simple implementation before. (I
>> guarantee the trivial case by the partial evaluation.)
>>
>> Anyway, I anticipate a few days til I get the final checkins. The
>> code is not tricky in implementation (not much, anyway) and
>> guarantees arbitrary expressions with well-defined compute time.
>>
>> I'm also not at all married to the filenames, but let's hold off for
>> a bit more as I get these chunks into place. apropos_db.c is fine
>> by me, for the record.
>
> Sounds good, except that a hackathon is a bad time to make me wait.
--
To unsubscribe send an email to tech+unsubscribe@mdocml.bsd.lv
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2011-11-13 9:23 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-11-09 1:30 overhaul apropos(1) interface Ingo Schwarze
2011-11-09 10:19 ` Kristaps Dzonsons
2011-11-12 23:54 ` Ingo Schwarze
2011-11-13 0:07 ` Kristaps Dzonsons
2011-11-13 0:39 ` Ingo Schwarze
2011-11-13 9:23 ` Kristaps Dzonsons
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).