freebsd-nq/usr.bin/tr/cset.c
Tim J. Robbins ca99cfdd14 Add support for multibyte characters. The challenge here was to use
data structures that scale better with large character sets, instead of
arrays indexed by character value:
- Sets of characters to delete/squeeze are stored in a new "cset" structure,
which is implemented as a splay tree of extents. This structure has the
ability to store character classes (ala wctype(3)), but this is not
currently fully utilized.
- Mappings between characters are stored in a new "cmap" structure, which
is also a splay tree.
- The parser no longer builds arrays containing all the characters in a
particular class; instead, next() determines them on-the-fly using
nextwctype(3).
2004-07-09 02:08:07 +00:00

304 lines
7.0 KiB
C

/*-
* Copyright (c) 2004 Tim J. Robbins.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
/*
* "Set of characters" ADT implemented as a splay tree of extents, with
* a lookup table cache to simplify looking up the first bunch of
* characters (which are presumably more common than others).
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <assert.h>
#include <stdbool.h>
#include <stdlib.h>
#include <wchar.h>
#include <wctype.h>
#include "cset.h"
static struct csnode * cset_delete(struct csnode *, wchar_t);
static __inline int cset_rangecmp(struct csnode *, wchar_t);
static struct csnode * cset_splay(struct csnode *, wchar_t);
/*
* cset_alloc --
* Allocate a set of characters.
*/
struct cset *
cset_alloc(void)
{
struct cset *cs;
if ((cs = malloc(sizeof(*cs))) == NULL)
return (NULL);
cs->cs_root = NULL;
cs->cs_classes = NULL;
cs->cs_havecache = false;
return (cs);
}
/*
* cset_add --
* Add a character to the set.
*/
bool
cset_add(struct cset *cs, wchar_t ch)
{
struct csnode *csn, *ncsn;
wchar_t oval;
cs->cs_havecache = false;
/*
* Inserting into empty tree; new item becomes the root.
*/
if (cs->cs_root == NULL) {
csn = malloc(sizeof(*cs->cs_root));
if (csn == NULL)
return (false);
csn->csn_left = csn->csn_right = NULL;
csn->csn_min = csn->csn_max = ch;
cs->cs_root = csn;
return (true);
}
/*
* Splay to check whether the item already exists, and otherwise,
* where we should put it.
*/
csn = cs->cs_root = cset_splay(cs->cs_root, ch);
/*
* Easy cases where we can avoid allocating a new node:
* (a) node already exists.
* (b) we can lower the extent's "min" to accomodate this
* character without having to coalesce.
* (c) we can raise the extent's "max" without having
* to coalesce.
*/
if (cset_rangecmp(csn, ch) == 0)
return (true);
if (ch + 1 == csn->csn_min && (csn->csn_left == NULL ||
ch > csn->csn_left->csn_max + 1)) {
csn->csn_min--;
return (true);
}
if (ch == csn->csn_max + 1 && (csn->csn_right == NULL ||
ch + 1 < csn->csn_right->csn_min)) {
csn->csn_max++;
return (true);
}
/*
* Allocate a new node and link it into the tree as a direct
* child of the root.
*/
ncsn = malloc(sizeof(*ncsn));
if (ncsn == NULL)
return (false);
ncsn->csn_min = ncsn->csn_max = ch;
if (cset_rangecmp(csn, ch) < 0) {
ncsn->csn_left = csn->csn_left;
ncsn->csn_right = csn;
csn->csn_left = NULL;
} else {
ncsn->csn_right = csn->csn_right;
ncsn->csn_left = csn;
csn->csn_right = NULL;
}
cs->cs_root = ncsn;
/*
* Splay to bring the newly inserted node to the root, then
* coalesce with left and right neighbours if possible.
*/
csn = cs->cs_root = cset_splay(cs->cs_root, ch);
if (csn->csn_left != NULL &&
csn->csn_left->csn_max + 1 == csn->csn_min) {
oval = csn->csn_left->csn_min;
cs->cs_root = cset_delete(cs->cs_root,
csn->csn_left->csn_min);
ncsn->csn_min = oval;
}
csn = cs->cs_root = cset_splay(cs->cs_root, ch);
if (csn->csn_right != NULL &&
csn->csn_right->csn_min - 1 == csn->csn_max) {
oval = csn->csn_right->csn_max;
cs->cs_root = cset_delete(cs->cs_root,
csn->csn_right->csn_min);
ncsn->csn_max = oval;
}
return (true);
}
/*
* cset_in_hard --
* Determine whether a character is in the set without using
* the cache.
*/
bool
cset_in_hard(struct cset *cs, wchar_t ch)
{
struct csclass *csc;
for (csc = cs->cs_classes; csc != NULL; csc = csc->csc_next)
if (csc->csc_invert ^ iswctype(ch, csc->csc_type) != 0)
return (cs->cs_invert ^ true);
if (cs->cs_root != NULL) {
cs->cs_root = cset_splay(cs->cs_root, ch);
return (cs->cs_invert ^ cset_rangecmp(cs->cs_root, ch) == 0);
}
return (cs->cs_invert ^ false);
}
/*
* cset_cache --
* Update the cache.
*/
void
cset_cache(struct cset *cs)
{
wchar_t i;
for (i = 0; i < CS_CACHE_SIZE; i++)
cs->cs_cache[i] = cset_in_hard(cs, i);
cs->cs_havecache = true;
}
/*
* cset_invert --
* Invert the character set.
*/
void
cset_invert(struct cset *cs)
{
cs->cs_invert ^= true;
cs->cs_havecache = false;
}
/*
* cset_addclass --
* Add a wctype()-style character class to the set, optionally
* inverting it.
*/
bool
cset_addclass(struct cset *cs, wctype_t type, bool invert)
{
struct csclass *csc;
csc = malloc(sizeof(*csc));
if (csc == NULL)
return (false);
csc->csc_type = type;
csc->csc_invert = invert;
csc->csc_next = cs->cs_classes;
cs->cs_classes = csc;
cs->cs_havecache = false;
return (true);
}
static __inline int
cset_rangecmp(struct csnode *t, wchar_t ch)
{
if (ch < t->csn_min)
return (-1);
if (ch > t->csn_max)
return (1);
return (0);
}
static struct csnode *
cset_splay(struct csnode *t, wchar_t ch)
{
struct csnode N, *l, *r, *y;
/*
* Based on public domain code from Sleator.
*/
assert(t != NULL);
N.csn_left = N.csn_right = NULL;
l = r = &N;
for (;;) {
if (cset_rangecmp(t, ch) < 0) {
if (t->csn_left != NULL &&
cset_rangecmp(t->csn_left, ch) < 0) {
y = t->csn_left;
t->csn_left = y->csn_right;
y->csn_right = t;
t = y;
}
if (t->csn_left == NULL)
break;
r->csn_left = t;
r = t;
t = t->csn_left;
} else if (cset_rangecmp(t, ch) > 0) {
if (t->csn_right != NULL &&
cset_rangecmp(t->csn_right, ch) > 0) {
y = t->csn_right;
t->csn_right = y->csn_left;
y->csn_left = t;
t = y;
}
if (t->csn_right == NULL)
break;
l->csn_right = t;
l = t;
t = t->csn_right;
} else
break;
}
l->csn_right = t->csn_left;
r->csn_left = t->csn_right;
t->csn_left = N.csn_right;
t->csn_right = N.csn_left;
return (t);
}
static struct csnode *
cset_delete(struct csnode *t, wchar_t ch)
{
struct csnode *x;
assert(t != NULL);
t = cset_splay(t, ch);
assert(cset_rangecmp(t, ch) == 0);
if (t->csn_left == NULL)
x = t->csn_right;
else {
x = cset_splay(t->csn_left, ch);
x->csn_right = t->csn_right;
}
free(t);
return x;
}