version 1.41, 2011/05/14 17:54:42 |
version 1.61, 2014/10/26 18:07:28 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org> |
* Copyright (c) 2011, 2014 Ingo Schwarze <schwarze@openbsd.org> |
* |
* |
* Permission to use, copy, modify, and distribute this software for any |
* Permission to use, copy, modify, and distribute this software for any |
* purpose with or without fee is hereby granted, provided that the above |
* purpose with or without fee is hereby granted, provided that the above |
|
|
* 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" |
#include "config.h" |
#endif |
|
|
|
|
#include <sys/types.h> |
|
|
#include <assert.h> |
#include <assert.h> |
#include <ctype.h> |
#include <ctype.h> |
#include <stdio.h> |
|
#include <stdlib.h> |
#include <stdlib.h> |
#include <string.h> |
#include <string.h> |
|
|
#include "mandoc.h" |
#include "mandoc.h" |
|
#include "mandoc_aux.h" |
|
#include "libmandoc.h" |
|
|
#define PRINT_HI 126 |
#define PRINT_HI 126 |
#define PRINT_LO 32 |
#define PRINT_LO 32 |
|
|
const char *code; |
const char *code; |
const char *ascii; |
const char *ascii; |
int unicode; |
int unicode; |
int type; |
|
#define CHARS_CHAR (1 << 0) |
|
#define CHARS_STRING (1 << 1) |
|
#define CHARS_BOTH (CHARS_CHAR | CHARS_STRING) |
|
}; |
}; |
|
|
#define LINES_MAX 353 |
#define LINES_MAX 330 |
|
|
#define CHAR(in, ch, code) \ |
#define CHAR(in, ch, code) \ |
{ NULL, (in), (ch), (code), CHARS_CHAR }, |
{ NULL, (in), (ch), (code) }, |
#define STRING(in, ch, code) \ |
|
{ NULL, (in), (ch), (code), CHARS_STRING }, |
|
#define BOTH(in, ch, code) \ |
|
{ NULL, (in), (ch), (code), CHARS_BOTH }, |
|
|
|
#define CHAR_TBL_START static struct ln lines[LINES_MAX] = { |
#define CHAR_TBL_START static struct ln lines[LINES_MAX] = { |
#define CHAR_TBL_END }; |
#define CHAR_TBL_END }; |
|
|
struct ln **htab; |
struct ln **htab; |
}; |
}; |
|
|
static inline int match(const struct ln *, |
static const struct ln *find(const struct mchars *, |
const char *, size_t, int); |
const char *, size_t); |
static const struct ln *find(struct mchars *, const char *, size_t, int); |
|
|
|
|
|
void |
void |
mchars_free(struct mchars *arg) |
mchars_free(struct mchars *arg) |
{ |
{ |
Line 82 mchars_alloc(void) |
|
Line 75 mchars_alloc(void) |
|
/* |
/* |
* Constructs a very basic chaining hashtable. The hash routine |
* Constructs a very basic chaining hashtable. The hash routine |
* is simply the integral value of the first character. |
* is simply the integral value of the first character. |
* Subsequent entries are chained in the order they're processed |
* Subsequent entries are chained in the order they're processed. |
* (they're in-line re-ordered during lookup). |
|
*/ |
*/ |
|
|
tab = mandoc_malloc(sizeof(struct mchars)); |
tab = mandoc_malloc(sizeof(struct mchars)); |
htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **)); |
htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln *)); |
|
|
for (i = 0; i < LINES_MAX; i++) { |
for (i = 0; i < LINES_MAX; i++) { |
hash = (int)lines[i].code[0] - PRINT_LO; |
hash = (int)lines[i].code[0] - PRINT_LO; |
Line 106 mchars_alloc(void) |
|
Line 98 mchars_alloc(void) |
|
return(tab); |
return(tab); |
} |
} |
|
|
|
|
/* |
|
* Special character to Unicode codepoint. |
|
*/ |
|
int |
int |
mchars_spec2cp(struct mchars *arg, const char *p, size_t sz) |
mchars_spec2cp(const struct mchars *arg, const char *p, size_t sz) |
{ |
{ |
const struct ln *ln; |
const struct ln *ln; |
|
|
ln = find(arg, p, sz, CHARS_CHAR); |
ln = find(arg, p, sz); |
if (NULL == ln) |
return(ln != NULL ? ln->unicode : sz == 1 ? *p : -1); |
return(-1); |
|
return(ln->unicode); |
|
} |
} |
|
|
|
char |
/* |
mchars_num2char(const char *p, size_t sz) |
* Reserved word to Unicode codepoint. |
|
*/ |
|
int |
|
mchars_res2cp(struct mchars *arg, const char *p, size_t sz) |
|
{ |
{ |
const struct ln *ln; |
int i; |
|
|
ln = find(arg, p, sz, CHARS_STRING); |
if ((i = mandoc_strntoi(p, sz, 10)) < 0) |
if (NULL == ln) |
return('\0'); |
return(-1); |
|
return(ln->unicode); |
return(i > 0 && i < 256 && isprint(i) ? i : '\0'); |
} |
} |
|
|
/* |
int |
* Numbered character to literal character. |
mchars_num2uc(const char *p, size_t sz) |
* This can only be a printable character (i.e., alnum, punct, space) so |
|
* prevent the character from ruining our state (backspace, newline, and |
|
* so on). |
|
*/ |
|
char |
|
mchars_num2char(const char *p, size_t sz) |
|
{ |
{ |
int i; |
int i; |
|
|
if (sz > 3) |
if ((i = mandoc_strntoi(p, sz, 16)) < 0) |
return('\0'); |
return(0xFFFD); |
|
|
i = atoi(p); |
/* |
/* LINTED */ |
* XXX Code is missing here to exclude bogus ranges. |
return(isprint(i) ? i : '\0'); |
*/ |
|
|
|
return(i <= 0x10FFFF ? i : 0xFFFD); |
} |
} |
|
|
/* |
|
* Special character to string array. |
|
*/ |
|
const char * |
const char * |
mchars_spec2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz) |
mchars_spec2str(const struct mchars *arg, |
|
const char *p, size_t sz, size_t *rsz) |
{ |
{ |
const struct ln *ln; |
const struct ln *ln; |
|
|
ln = find(arg, p, sz, CHARS_CHAR); |
ln = find(arg, p, sz); |
if (NULL == ln) |
if (ln == NULL) { |
return(NULL); |
*rsz = 1; |
|
return(sz == 1 ? p : NULL); |
|
} |
|
|
*rsz = strlen(ln->ascii); |
*rsz = strlen(ln->ascii); |
return(ln->ascii); |
return(ln->ascii); |
} |
} |
|
|
/* |
|
* Reserved word to string array. |
|
*/ |
|
const char * |
const char * |
mchars_res2str(struct mchars *arg, const char *p, size_t sz, size_t *rsz) |
mchars_uc2str(int uc) |
{ |
{ |
const struct ln *ln; |
int i; |
|
|
ln = find(arg, p, sz, CHARS_STRING); |
for (i = 0; i < LINES_MAX; i++) |
if (NULL == ln) |
if (uc == lines[i].unicode) |
return(NULL); |
return(lines[i].ascii); |
|
return("<?>"); |
*rsz = strlen(ln->ascii); |
|
return(ln->ascii); |
|
} |
} |
|
|
static const struct ln * |
static const struct ln * |
find(struct mchars *tab, const char *p, size_t sz, int type) |
find(const struct mchars *tab, const char *p, size_t sz) |
{ |
{ |
struct ln *pp, *prev; |
const struct ln *pp; |
struct ln **htab; |
|
int hash; |
int hash; |
|
|
assert(p); |
assert(p); |
if (0 == sz) |
|
return(NULL); |
|
|
|
if (p[0] < PRINT_LO || p[0] > PRINT_HI) |
if (0 == sz || p[0] < PRINT_LO || p[0] > PRINT_HI) |
return(NULL); |
return(NULL); |
|
|
/* |
|
* Lookup the symbol in the symbol hash. See ascii2htab for the |
|
* hashtable specs. This dynamically re-orders the hash chain |
|
* to optimise for repeat hits. |
|
*/ |
|
|
|
hash = (int)p[0] - PRINT_LO; |
hash = (int)p[0] - PRINT_LO; |
htab = tab->htab; |
|
|
|
if (NULL == (pp = htab[hash])) |
for (pp = tab->htab[hash]; pp; pp = pp->next) |
return(NULL); |
if (0 == strncmp(pp->code, p, sz) && |
|
'\0' == pp->code[(int)sz]) |
|
return(pp); |
|
|
for (prev = NULL; pp; pp = pp->next) { |
|
if ( ! match(pp, p, sz, type)) { |
|
prev = pp; |
|
continue; |
|
} |
|
|
|
if (prev) { |
|
prev->next = pp->next; |
|
pp->next = htab[hash]; |
|
htab[hash] = pp; |
|
} |
|
|
|
return(pp); |
|
} |
|
|
|
return(NULL); |
return(NULL); |
} |
|
|
|
static inline int |
|
match(const struct ln *ln, const char *p, size_t sz, int type) |
|
{ |
|
|
|
if ( ! (ln->type & type)) |
|
return(0); |
|
if (strncmp(ln->code, p, sz)) |
|
return(0); |
|
return('\0' == ln->code[(int)sz]); |
|
} |
} |