summaryrefslogtreecommitdiff
path: root/lib/unistr/u8-strchr.c
diff options
context:
space:
mode:
authorManuel A. Fernandez Montecelo <manuel.montezelo@gmail.com>2016-05-27 14:35:40 +0100
committerManuel A. Fernandez Montecelo <manuel.montezelo@gmail.com>2016-05-27 14:35:40 +0100
commitb1de003dac299705a7f01c997d2b866bafe39926 (patch)
tree1cc16a3877e945116387a380f7f3023f81fa36e4 /lib/unistr/u8-strchr.c
parent752fd7247bc223bcea35bd89cf56d1c08ead9ba6 (diff)
parent3590c846d4c2febbc05b4ad6b14a06edc549e453 (diff)
Merge tag 'upstream/0.9.6+really0.9.6'
Upstream version 0.9.6+really0.9.6
Diffstat (limited to 'lib/unistr/u8-strchr.c')
-rw-r--r--lib/unistr/u8-strchr.c191
1 files changed, 165 insertions, 26 deletions
diff --git a/lib/unistr/u8-strchr.c b/lib/unistr/u8-strchr.c
index 3be14c7..dd1cb84 100644
--- a/lib/unistr/u8-strchr.c
+++ b/lib/unistr/u8-strchr.c
@@ -1,5 +1,5 @@
/* Search character in UTF-8 string.
- Copyright (C) 1999, 2002, 2006-2007, 2009-2010 Free Software Foundation,
+ Copyright (C) 1999, 2002, 2006-2007, 2009-2015 Free Software Foundation,
Inc.
Written by Bruno Haible <bruno@clisp.org>, 2002.
@@ -21,6 +21,8 @@
/* Specification. */
#include "unistr.h"
+#include <string.h>
+
uint8_t *
u8_strchr (const uint8_t *s, ucs4_t uc)
{
@@ -30,72 +32,209 @@ u8_strchr (const uint8_t *s, ucs4_t uc)
{
uint8_t c0 = uc;
- for (;; s++)
+ if (false)
+ {
+ /* Unoptimized code. */
+ for (;;)
+ {
+ uint8_t s0 = *s;
+ if (s0 == c0)
+ return (uint8_t *) s;
+ s++;
+ if (s0 == 0)
+ break;
+ }
+ }
+ else
{
- if (*s == c0)
- break;
- if (*s == 0)
- goto notfound;
+ /* Optimized code.
+ strchr() is often so well optimized, that it's worth the
+ added function call. */
+ return (uint8_t *) strchr ((const char *) s, c0);
}
- return (uint8_t *) s;
}
else
+ /* Loops equivalent to strstr, optimized for a specific length (2, 3, 4)
+ of the needle. We use an algorithm similar to Boyer-Moore which
+ is documented in lib/unistr/u8-chr.c. There is additional
+ complication because we need to check after every byte for
+ a NUL byte, but the idea is the same. */
switch (u8_uctomb_aux (c, uc, 6))
{
case 2:
- if (*s == 0)
- goto notfound;
+ if (*s == 0 || s[1] == 0)
+ break;
{
uint8_t c0 = c[0];
uint8_t c1 = c[1];
+ /* Search for { c0, c1 }. */
+ uint8_t s1 = s[1];
- for (;; s++)
+ for (;;)
{
+ /* Here s[0] != 0, s[1] != 0.
+ Test whether s[0..1] == { c0, c1 }. */
+ if (s1 == c1)
+ {
+ if (*s == c0)
+ return (uint8_t *) s;
+ else
+ /* Skip the search at s + 1, because s[1] = c1 < c0. */
+ goto case2_skip2;
+ }
+ else
+ {
+ if (s1 == c0)
+ goto case2_skip1;
+ else
+ /* Skip the search at s + 1, because s[1] != c0. */
+ goto case2_skip2;
+ }
+ case2_skip2:
+ s++;
+ s1 = s[1];
+ if (s[1] == 0)
+ break;
+ case2_skip1:
+ s++;
+ s1 = s[1];
if (s[1] == 0)
- goto notfound;
- if (*s == c0 && s[1] == c1)
break;
}
- return (uint8_t *) s;
}
+ break;
case 3:
- if (*s == 0 || s[1] == 0)
- goto notfound;
+ if (*s == 0 || s[1] == 0 || s[2] == 0)
+ break;
{
uint8_t c0 = c[0];
uint8_t c1 = c[1];
uint8_t c2 = c[2];
+ /* Search for { c0, c1, c2 }. */
+ uint8_t s2 = s[2];
- for (;; s++)
+ for (;;)
{
+ /* Here s[0] != 0, s[1] != 0, s[2] != 0.
+ Test whether s[0..2] == { c0, c1, c2 }. */
+ if (s2 == c2)
+ {
+ if (s[1] == c1 && *s == c0)
+ return (uint8_t *) s;
+ else
+ /* If c2 != c1:
+ Skip the search at s + 1, because s[2] == c2 != c1.
+ Skip the search at s + 2, because s[2] == c2 < c0. */
+ if (c2 == c1)
+ goto case3_skip1;
+ else
+ goto case3_skip3;
+ }
+ else
+ {
+ if (s2 == c1)
+ goto case3_skip1;
+ else if (s2 == c0)
+ /* Skip the search at s + 1, because s[2] != c1. */
+ goto case3_skip2;
+ else
+ /* Skip the search at s + 1, because s[2] != c1.
+ Skip the search at s + 2, because s[2] != c0. */
+ goto case3_skip3;
+ }
+ case3_skip3:
+ s++;
+ s2 = s[2];
+ if (s[2] == 0)
+ break;
+ case3_skip2:
+ s++;
+ s2 = s[2];
+ if (s[2] == 0)
+ break;
+ case3_skip1:
+ s++;
+ s2 = s[2];
if (s[2] == 0)
- goto notfound;
- if (*s == c0 && s[1] == c1 && s[2] == c2)
break;
}
- return (uint8_t *) s;
}
+ break;
case 4:
- if (*s == 0 || s[1] == 0 || s[2] == 0)
- goto notfound;
+ if (*s == 0 || s[1] == 0 || s[2] == 0 || s[3] == 0)
+ break;
{
uint8_t c0 = c[0];
uint8_t c1 = c[1];
uint8_t c2 = c[2];
uint8_t c3 = c[3];
+ /* Search for { c0, c1, c2, c3 }. */
+ uint8_t s3 = s[3];
- for (;; s++)
+ for (;;)
{
+ /* Here s[0] != 0, s[1] != 0, s[2] != 0, s[3] != 0.
+ Test whether s[0..3] == { c0, c1, c2, c3 }. */
+ if (s3 == c3)
+ {
+ if (s[2] == c2 && s[1] == c1 && *s == c0)
+ return (uint8_t *) s;
+ else
+ /* If c3 != c2:
+ Skip the search at s + 1, because s[3] == c3 != c2.
+ If c3 != c1:
+ Skip the search at s + 2, because s[3] == c3 != c1.
+ Skip the search at s + 3, because s[3] == c3 < c0. */
+ if (c3 == c2)
+ goto case4_skip1;
+ else if (c3 == c1)
+ goto case4_skip2;
+ else
+ goto case4_skip4;
+ }
+ else
+ {
+ if (s3 == c2)
+ goto case4_skip1;
+ else if (s3 == c1)
+ /* Skip the search at s + 1, because s[3] != c2. */
+ goto case4_skip2;
+ else if (s3 == c0)
+ /* Skip the search at s + 1, because s[3] != c2.
+ Skip the search at s + 2, because s[3] != c1. */
+ goto case4_skip3;
+ else
+ /* Skip the search at s + 1, because s[3] != c2.
+ Skip the search at s + 2, because s[3] != c1.
+ Skip the search at s + 3, because s[3] != c0. */
+ goto case4_skip4;
+ }
+ case4_skip4:
+ s++;
+ s3 = s[3];
+ if (s[3] == 0)
+ break;
+ case4_skip3:
+ s++;
+ s3 = s[3];
+ if (s[3] == 0)
+ break;
+ case4_skip2:
+ s++;
+ s3 = s[3];
+ if (s[3] == 0)
+ break;
+ case4_skip1:
+ s++;
+ s3 = s[3];
if (s[3] == 0)
- goto notfound;
- if (*s == c0 && s[1] == c1 && s[2] == c2 && s[3] == c3)
break;
}
- return (uint8_t *) s;
}
+ break;
}
-notfound:
+
return NULL;
}