In .Bl -column, if some of the column width declarations are given

[mandoc.git] / chars.c

diff --git a/chars.c b/chars.c
index 2585484483c4c8487afafbcd837e4dcc5f907034..d758d0ccbd1bd6d0a219b635ffc9e53f2a35e3dc 100644 (file)
--- a/chars.c
+++ b/chars.c
@@ -1,6 +1,7 @@
-/*     $Id: chars.c,v 1.23 2010/07/18 12:10:08 kristaps Exp $ */
+/*     $Id: chars.c,v 1.58 2014/07/23 15:00:08 schwarze Exp $ */

 /*
 /*

- * Copyright (c) 2009 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2009, 2010, 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
  *
  * Permission to use, copy, modify, and distribute this software for any
  * purpose with or without fee is hereby granted, provided that the above


@@ -19,12 +20,13 @@
 #endif
 
 #include <assert.h>
 #endif
 
 #include <assert.h>

-#include <stdio.h>
+#include <ctype.h>

 #include <stdlib.h>
 #include <string.h>
 
 #include "mandoc.h"
 #include <stdlib.h>
 #include <string.h>
 
 #include "mandoc.h"

-#include "chars.h"
+#include "mandoc_aux.h"
+#include "libmandoc.h"

 
 #define        PRINT_HI         126
 #define        PRINT_LO         32
 
 #define        PRINT_HI         126
 #define        PRINT_LO         32


@@ -33,54 +35,39 @@ struct      ln {
        struct ln        *next;
        const char       *code;
        const char       *ascii;
        struct ln        *next;
        const char       *code;
        const char       *ascii;

-       size_t            asciisz;

        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         370
+#define        LINES_MAX         330

 
 

-#define CHAR(in, ch, chsz, code) \
-       { NULL, (in), (ch), (chsz), (code), CHARS_CHAR },
-#define STRING(in, ch, chsz, code) \
-       { NULL, (in), (ch), (chsz), (code), CHARS_STRING },
-#define BOTH(in, ch, chsz, code) \
-       { NULL, (in), (ch), (chsz), (code), CHARS_BOTH },
+#define CHAR(in, ch, code) \
+       { NULL, (in), (ch), (code) },

 
 #define        CHAR_TBL_START    static struct ln lines[LINES_MAX] = {
 #define        CHAR_TBL_END      };
 
 #include "chars.in"
 
 
 #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;
 };
 
        struct ln       **htab;
 };
 

-static inline int        match(const struct ln *,
-                               const char *, size_t, int);
-static const struct ln  *find(struct tbl *, const char *, size_t, int);
+static const struct ln  *find(const struct mchars *,
+                               const char *, size_t);

 
 
 void
 
 
 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;
        struct ln       **htab;
        struct ln        *pp;
        int               i, hash;


@@ -88,21 +75,11 @@ chars_init(enum chars type)
        /*
         * Constructs a very basic chaining hashtable.  The hash routine
         * is simply the integral value of the first character.
        /*
         * 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) {
-               perror(NULL);
-               exit(EXIT_FAILURE);
-       }
-
-       htab = calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **));
-       if (NULL == htab) {
-               perror(NULL);
-               exit(EXIT_FAILURE);
-       }
+       tab = mandoc_malloc(sizeof(struct mchars));
+       htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln *));

 
        for (i = 0; i < LINES_MAX; i++) {
                hash = (int)lines[i].code[0] - PRINT_LO;
 
        for (i = 0; i < LINES_MAX; i++) {
                hash = (int)lines[i].code[0] - PRINT_LO;


@@ -118,127 +95,86 @@ chars_init(enum chars type)
        }
 
        tab->htab = htab;
        }
 
        tab->htab = htab;

-       tab->type = type;

        return(tab);
 }
 
        return(tab);
 }
 

-
-/* 
- * Special character to Unicode codepoint.
- */

 int
 int

-chars_spec2cp(void *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((struct tbl *)arg, p, sz, CHARS_CHAR);
+       ln = find(arg, p, sz);

        if (NULL == ln)
                return(-1);
        return(ln->unicode);
 }
 
        if (NULL == ln)
                return(-1);
        return(ln->unicode);
 }
 

-
-/* 
- * Reserved word to Unicode codepoint.
- */
-int
-chars_res2cp(void *arg, const char *p, size_t sz)
+char
+mchars_num2char(const char *p, size_t sz)

 {
 {

-       const struct ln *ln;
+       int       i;

 
 

-       ln = find((struct tbl *)arg, p, sz, CHARS_STRING);
-       if (NULL == ln)
-               return(-1);
-       return(ln->unicode);
-}
+       if ((i = mandoc_strntoi(p, sz, 10)) < 0)
+               return('\0');

 
 

+       return(i > 0 && i < 256 && isprint(i) ? i : '\0');
+}

 
 

-/* 
- * Special character to string array.
- */
-const char *
-chars_spec2str(void *arg, const char *p, size_t sz, size_t *rsz)
+int
+mchars_num2uc(const char *p, size_t sz)

 {
 {

-       const struct ln *ln;
+       int      i;

 
 

-       ln = find((struct tbl *)arg, p, sz, CHARS_CHAR);
-       if (NULL == ln)
-               return(NULL);
+       if ((i = mandoc_strntoi(p, sz, 16)) < 0)
+               return('\0');

 
 

-       *rsz = ln->asciisz;
-       return(ln->ascii);
-}
+       /*
+        * 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.
+        */

 
 

+       return(i > 0x80 && i <= 0x10FFFF ? i : '\0');
+}

 
 

-/* 
- * Reserved word to string array.
- */

 const char *
 const char *

-chars_res2str(void *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((struct tbl *)arg, p, sz, CHARS_STRING);
-       if (NULL == ln)
+       ln = find(arg, p, sz);
+       if (NULL == ln) {
+               *rsz = 1;

                return(NULL);
                return(NULL);

+       }

 
 

-       *rsz = ln->asciisz;
+       *rsz = strlen(ln->ascii);

        return(ln->ascii);
 }
 
        return(ln->ascii);
 }
 

-

 static const struct ln *
 static const struct ln *

-find(struct tbl *tab, const char *p, size_t sz, int type)
+find(const struct mchars *tab, const char *p, size_t sz)

 {
 {

-       struct ln        *pp, *prev;
-       struct ln       **htab;
+       const struct ln  *pp;

        int               hash;
 
        assert(p);
        int               hash;
 
        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]))
-               return(NULL);
-
-       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);
-       }
+       for (pp = tab->htab[hash]; pp; pp = pp->next)
+               if (0 == strncmp(pp->code, p, sz) &&
+                   '\0' == pp->code[(int)sz])
+                       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]);
-}