=================================================================== RCS file: /cvs/mandoc/chars.c,v retrieving revision 1.10 retrieving revision 1.58 diff -u -p -r1.10 -r1.58 --- mandoc/chars.c 2009/10/30 18:43:24 1.10 +++ mandoc/chars.c 2014/07/23 15:00:08 1.58 @@ -1,6 +1,7 @@ -/* $Id: chars.c,v 1.10 2009/10/30 18:43:24 kristaps Exp $ */ +/* $Id: chars.c,v 1.58 2014/07/23 15:00:08 schwarze Exp $ */ /* - * Copyright (c) 2009 Kristaps Dzonsons + * Copyright (c) 2009, 2010, 2011 Kristaps Dzonsons + * Copyright (c) 2011 Ingo Schwarze * * Permission to use, copy, modify, and distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -14,12 +15,18 @@ * 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 "chars.h" +#include "mandoc.h" +#include "mandoc_aux.h" +#include "libmandoc.h" #define PRINT_HI 126 #define PRINT_LO 32 @@ -28,56 +35,39 @@ struct ln { struct ln *next; const char *code; const char *ascii; - const char *html; - size_t codesz; - size_t asciisz; - size_t htmlsz; - int type; -#define CHARS_CHAR (1 << 0) -#define CHARS_STRING (1 << 1) -#define CHARS_BOTH (0x03) + int unicode; }; -#define LINES_MAX 351 +#define LINES_MAX 330 -#define CHAR(w, x, y, z, a, b) \ - { NULL, (w), (y), (a), (x), (z), (b), CHARS_CHAR }, -#define STRING(w, x, y, z, a, b) \ - { NULL, (w), (y), (a), (x), (z), (b), CHARS_STRING }, -#define BOTH(w, x, y, z, a, b) \ - { NULL, (w), (y), (a), (x), (z), (b), CHARS_BOTH }, +#define CHAR(in, ch, code) \ + { NULL, (in), (ch), (code) }, -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 mchars { 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(const struct mchars *, + const char *, size_t); void -chars_free(void *arg) +mchars_free(struct mchars *arg) { - struct tbl *tab; - tab = (struct tbl *)arg; - - free(tab->htab); - free(tab); + free(arg->htab); + free(arg); } - -void * -chars_init(enum chars type) +struct mchars * +mchars_alloc(void) { - struct tbl *tab; + struct mchars *tab; struct ln **htab; struct ln *pp; int i, hash; @@ -85,22 +75,12 @@ chars_init(enum chars type) /* * Constructs a very basic chaining hashtable. The hash routine * is simply the integral value of the first character. - * Subsequent entries are chained in the order they're processed - * (they're in-line re-ordered during lookup). + * Subsequent entries are chained in the order they're processed. */ - tab = malloc(sizeof(struct tbl)); - if (NULL == tab) { - fprintf(stderr, "memory exhausted\n"); - exit(EXIT_FAILURE); - } + tab = mandoc_malloc(sizeof(struct mchars)); + htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln *)); - htab = calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **)); - if (NULL == htab) { - fprintf(stderr, "memory exhausted\n"); - exit(EXIT_FAILURE); - } - for (i = 0; i < LINES_MAX; i++) { hash = (int)lines[i].code[0] - PRINT_LO; @@ -115,95 +95,86 @@ chars_init(enum chars type) } tab->htab = htab; - tab->type = type; return(tab); } - -const char * -chars_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz) +int +mchars_spec2cp(const struct mchars *arg, const char *p, size_t sz) { + const struct ln *ln; - return(find((struct tbl *)arg, p, sz, rsz, CHARS_CHAR)); + ln = find(arg, p, sz); + if (NULL == ln) + return(-1); + return(ln->unicode); } - -const char * -chars_a2res(void *arg, const char *p, size_t sz, size_t *rsz) +char +mchars_num2char(const char *p, size_t sz) { + int i; - return(find((struct tbl *)arg, p, sz, rsz, CHARS_STRING)); + if ((i = mandoc_strntoi(p, sz, 10)) < 0) + return('\0'); + + return(i > 0 && i < 256 && isprint(i) ? i : '\0'); } - -static const char * -find(struct tbl *tab, const char *p, size_t sz, size_t *rsz, int type) +int +mchars_num2uc(const char *p, size_t sz) { - struct ln *pp, *prev; - struct ln **htab; - int hash; + int i; - assert(p); - assert(sz > 0); + if ((i = mandoc_strntoi(p, sz, 16)) < 0) + return('\0'); - if (p[0] < PRINT_LO || p[0] > PRINT_HI) - 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. + * Security warning: + * Never extend the range of accepted characters + * to overlap with the ASCII range, 0x00-0x7F + * without re-auditing the callers of this function. + * Some callers might relay on the fact that we never + * return ASCII characters for their escaping decisions. + * + * XXX Code is missing here to exclude bogus ranges. */ - hash = (int)p[0] - PRINT_LO; - htab = tab->htab; + return(i > 0x80 && i <= 0x10FFFF ? i : '\0'); +} - if (NULL == (pp = htab[hash])) - return(NULL); +const char * +mchars_spec2str(const struct mchars *arg, + const char *p, size_t sz, size_t *rsz) +{ + const struct ln *ln; - if (NULL == pp->next) { - if ( ! match(pp, p, sz, type)) - return(NULL); - - if (CHARS_HTML == tab->type) { - *rsz = pp->htmlsz; - return(pp->html); - } - *rsz = pp->asciisz; - return(pp->ascii); + ln = find(arg, p, sz); + if (NULL == ln) { + *rsz = 1; + return(NULL); } - for (prev = NULL; pp; pp = pp->next) { - if ( ! match(pp, p, sz, type)) { - prev = pp; - continue; - } + *rsz = strlen(ln->ascii); + return(ln->ascii); +} - if (prev) { - prev->next = pp->next; - pp->next = htab[hash]; - htab[hash] = pp; - } +static const struct ln * +find(const struct mchars *tab, const char *p, size_t sz) +{ + const struct ln *pp; + int hash; - if (CHARS_HTML == tab->type) { - *rsz = pp->htmlsz; - return(pp->html); - } - *rsz = pp->asciisz; - return(pp->ascii); - } + assert(p); - return(NULL); -} + if (0 == sz || p[0] < PRINT_LO || p[0] > PRINT_HI) + return(NULL); + hash = (int)p[0] - PRINT_LO; -static inline int -match(const struct ln *ln, const char *p, size_t sz, int type) -{ + for (pp = tab->htab[hash]; pp; pp = pp->next) + if (0 == strncmp(pp->code, p, sz) && + '\0' == pp->code[(int)sz]) + return(pp); - if ( ! (ln->type & type)) - return(0); - if (ln->codesz != sz) - return(0); - return(0 == strncmp(ln->code, p, sz)); + return(NULL); }