=================================================================== RCS file: /cvs/mandoc/chars.c,v retrieving revision 1.1 retrieving revision 1.28 diff -u -p -r1.1 -r1.28 --- mandoc/chars.c 2009/09/17 07:41:28 1.1 +++ mandoc/chars.c 2010/08/29 11:36:49 1.28 @@ -1,6 +1,6 @@ -/* $Id: chars.c,v 1.1 2009/09/17 07:41:28 kristaps Exp $ */ +/* $Id: chars.c,v 1.28 2010/08/29 11:36:49 kristaps Exp $ */ /* - * Copyright (c) 2009 Kristaps Dzonsons + * Copyright (c) 2009, 2010 Kristaps Dzonsons * * Permission to use, copy, modify, and distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -14,49 +14,54 @@ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */ +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + #include -#include +#include #include #include +#include "mandoc.h" #include "chars.h" -#define ASCII_PRINT_HI 126 -#define ASCII_PRINT_LO 32 +#define PRINT_HI 126 +#define PRINT_LO 32 struct ln { struct ln *next; const char *code; - const char *out; - size_t codesz; - size_t outsz; + const char *ascii; + int unicode; int type; #define CHARS_CHAR (1 << 0) #define CHARS_STRING (1 << 1) -#define CHARS_BOTH (0x03) +#define CHARS_BOTH (CHARS_CHAR | CHARS_STRING) }; -#define LINES_MAX 266 +#define LINES_MAX 362 -#define CHAR(w, x, y, z) \ - { NULL, (w), (y), (x), (z), CHARS_CHAR }, -#define STRING(w, x, y, z) \ - { NULL, (w), (y), (x), (z), CHARS_STRING }, -#define BOTH(w, x, y, z) \ - { NULL, (w), (y), (x), (z), CHARS_BOTH }, +#define CHAR(in, ch, code) \ + { NULL, (in), (ch), (code), CHARS_CHAR }, +#define STRING(in, ch, code) \ + { NULL, (in), (ch), (code), CHARS_STRING }, +#define BOTH(in, ch, code) \ + { NULL, (in), (ch), (code), CHARS_BOTH }, -static struct ln lines[LINES_MAX] = { +#define CHAR_TBL_START static struct ln lines[LINES_MAX] = { +#define CHAR_TBL_END }; + #include "chars.in" -}; struct tbl { + enum chars type; struct ln **htab; }; static inline int match(const struct ln *, const char *, size_t, int); -static const char *find(struct tbl *, const char *, - size_t, size_t *, int); +static const struct ln *find(struct tbl *, const char *, size_t, int); void @@ -71,7 +76,6 @@ chars_free(void *arg) } -/* ARGSUSED */ void * chars_init(enum chars type) { @@ -87,22 +91,21 @@ chars_init(enum chars type) * (they're in-line re-ordered during lookup). */ - if (NULL == (tab = malloc(sizeof(struct tbl)))) - err(1, "malloc"); + tab = malloc(sizeof(struct tbl)); + if (NULL == tab) { + perror(NULL); + exit(MANDOCLEVEL_SYSERR); + } - htab = calloc(ASCII_PRINT_HI - ASCII_PRINT_LO + 1, - sizeof(struct ln **)); + htab = calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **)); + if (NULL == htab) { + perror(NULL); + exit(MANDOCLEVEL_SYSERR); + } - if (NULL == htab) - err(1, "malloc"); - for (i = 0; i < LINES_MAX; i++) { - assert(lines[i].codesz > 0); - assert(lines[i].code); - assert(lines[i].out); + hash = (int)lines[i].code[0] - PRINT_LO; - hash = (int)lines[i].code[0] - ASCII_PRINT_LO; - if (NULL == (pp = htab[hash])) { htab[hash] = &lines[i]; continue; @@ -110,42 +113,91 @@ chars_init(enum chars type) for ( ; pp->next; pp = pp->next) /* Scan ahead. */ ; - pp->next = &lines[i]; } tab->htab = htab; + tab->type = type; return(tab); } +/* + * Special character to Unicode codepoint. + */ +int +chars_spec2cp(void *arg, const char *p, size_t sz) +{ + const struct ln *ln; + + ln = find((struct tbl *)arg, p, sz, CHARS_CHAR); + if (NULL == ln) + return(-1); + return(ln->unicode); +} + + +/* + * Reserved word to Unicode codepoint. + */ +int +chars_res2cp(void *arg, const char *p, size_t sz) +{ + const struct ln *ln; + + ln = find((struct tbl *)arg, p, sz, CHARS_STRING); + if (NULL == ln) + return(-1); + return(ln->unicode); +} + + +/* + * Special character to string array. + */ const char * -chars_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz) +chars_spec2str(void *arg, const char *p, size_t sz, size_t *rsz) { + const struct ln *ln; - return(find((struct tbl *)arg, p, sz, rsz, CHARS_CHAR)); + ln = find((struct tbl *)arg, p, sz, CHARS_CHAR); + if (NULL == ln) + return(NULL); + + *rsz = strlen(ln->ascii); + return(ln->ascii); } +/* + * Reserved word to string array. + */ const char * -chars_a2res(void *arg, const char *p, size_t sz, size_t *rsz) +chars_res2str(void *arg, const char *p, size_t sz, size_t *rsz) { + const struct ln *ln; - return(find((struct tbl *)arg, p, sz, rsz, CHARS_STRING)); + ln = find((struct tbl *)arg, p, sz, CHARS_STRING); + if (NULL == ln) + return(NULL); + + *rsz = strlen(ln->ascii); + return(ln->ascii); } -static const char * -find(struct tbl *tab, const char *p, size_t sz, size_t *rsz, int type) +static const struct ln * +find(struct tbl *tab, const char *p, size_t sz, int type) { struct ln *pp, *prev; struct ln **htab; int hash; assert(p); - assert(sz > 0); + if (0 == sz) + return(NULL); - if (p[0] < ASCII_PRINT_LO || p[0] > ASCII_PRINT_HI) + if (p[0] < PRINT_LO || p[0] > PRINT_HI) return(NULL); /* @@ -154,35 +206,25 @@ find(struct tbl *tab, const char *p, size_t sz, size_t * to optimise for repeat hits. */ - hash = (int)p[0] - ASCII_PRINT_LO; + hash = (int)p[0] - PRINT_LO; htab = tab->htab; if (NULL == (pp = htab[hash])) return(NULL); - if (NULL == pp->next) { - if ( ! match(pp, p, sz, type)) - return(NULL); - *rsz = pp->outsz; - return(pp->out); - } - for (prev = NULL; pp; pp = pp->next) { if ( ! match(pp, p, sz, type)) { prev = pp; continue; } - /* Re-order the hash chain. */ - if (prev) { prev->next = pp->next; pp->next = htab[hash]; htab[hash] = pp; } - *rsz = pp->outsz; - return(pp->out); + return(pp); } return(NULL); @@ -195,7 +237,7 @@ match(const struct ln *ln, const char *p, size_t sz, i if ( ! (ln->type & type)) return(0); - if (ln->codesz != sz) + if (strncmp(ln->code, p, sz)) return(0); - return(0 == strncmp(ln->code, p, sz)); + return('\0' == ln->code[(int)sz]); }