Fix to auto-closing of LINK tag in -Txhtml (thanks to Daniel Friesel).

[mandoc.git] / man_hash.c

diff --git a/man_hash.c b/man_hash.c
index 115a99812189a857e6de8e88a83edad198e8f3ed..587baeda22299cbe8a9f7387a47603fd839084d0 100644 (file)
--- a/man_hash.c
+++ b/man_hash.c
@@ -1,4 +1,4 @@
-/*     $Id: man_hash.c,v 1.12 2009/09/16 14:40:56 kristaps Exp $ */
+/*     $Id: man_hash.c,v 1.18 2010/03/27 10:14:32 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -14,17 +14,45 @@
  * 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 <sys/types.h>
+
 #include <assert.h>
+#include <ctype.h>
 #include <limits.h>
 #include <stdlib.h>
 #include <string.h>
 
 #include "libman.h"
 
+#define        HASH_DEPTH       6
 
-static unsigned char   table[26 * 6];
+#define        HASH_ROW(x) do { \
+               if ('.' == (x)) \
+                       (x) = 26; \
+               else if (isupper((u_char)(x))) \
+                       (x) -= 65; \
+               else \
+                       (x) -= 97; \
+               (x) *= HASH_DEPTH; \
+       } while (/* CONSTCOND */ 0)
 
+/*
+ * Lookup table is indexed first by lower-case first letter (plus one
+ * for the period, which is stored in the last row), then by lower or
+ * uppercase second letter.  Buckets correspond to the index of the
+ * macro (the integer value of the enum stored as a char to save a bit
+ * of space).
+ */
+static u_char           table[27 * HASH_DEPTH];
 
+/*
+ * XXX - this hash has global scope, so if intended for use as a library
+ * with multiple callers, it will need re-invocation protection.
+ */
 void
 man_hash_init(void)
 {
@@ -32,40 +60,45 @@ man_hash_init(void)
 
        memset(table, UCHAR_MAX, sizeof(table));
 
+       assert(/* LINTED */ 
+                       MAN_MAX < UCHAR_MAX);
+
        for (i = 0; i < MAN_MAX; i++) {
                x = man_macronames[i][0];
-               assert((x >= 65 && x <= 90) ||
-                               (x >= 97 && x <= 122));
 
-               x -= (x <= 90) ? 65 : 97;
-               x *= 6;
+               assert(isalpha((u_char)x) || '.' == x);
+
+               HASH_ROW(x);
 
-               for (j = 0; j < 6; j++)
+               for (j = 0; j < HASH_DEPTH; j++)
                        if (UCHAR_MAX == table[x + j]) {
-                               table[x + j] = i;
+                               table[x + j] = (u_char)i;
                                break;
                        }
-               assert(j < 6);
+
+               assert(j < HASH_DEPTH);
        }
 }
 
 
-int
+enum mant
 man_hash_find(const char *tmp)
 {
-       int              x, i, tok;
+       int              x, y, i;
+       enum mant        tok;
 
-       if (0 == (x = tmp[0]))
+       if ('\0' == (x = tmp[0]))
                return(MAN_MAX);
-       if ( ! ((x >= 65 && x <= 90) || (x >= 97 && x <= 122)))
+       if ( ! (isalpha((u_char)x) || '.' == x))
                return(MAN_MAX);
 
-       x -= (x <= 90) ? 65 : 97;
-       x *= 6;
+       HASH_ROW(x);
 
-       for (i = 0; i < 6; i++) {
-               if (UCHAR_MAX == (tok = table[x + i]))
+       for (i = 0; i < HASH_DEPTH; i++) {
+               if (UCHAR_MAX == (y = table[x + i]))
                        return(MAN_MAX);
+
+               tok = (enum mant)y;
                if (0 == strcmp(tmp, man_macronames[tok]))
                        return(tok);
        }