gnustep/libs-base
0
0
Fork 0
mirror of https://github.com/gnustep/libs-base.git synced 2025-05-29 16:01:38 +00:00
Code Issues Projects Releases Packages Wiki Activity

implement -hash for constant strings

Browse source 
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33979 72102866-910b-0410-8b05-ffd578937521
This commit is contained in:
rfm 2011-10-14 07:07:39 +00:00
parent 2756763d0f
commit 400f0b9e4d
2 changed files with 43 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ChangeLog
View file

@ -1,3 +1,7 @@
2011-10-14 Richard Frith-Macdonald <rfm@gnu.org>
* Source/GSString.m: Implement -hash for literal UTF-8 performance.
2011-10-12 Richard Frith-Macdonald <rfm@gnu.org>
* Source/GSPrivate.h:

42
Source/GSString.m
View file

@ -4833,6 +4833,9 @@ lengthUTF8(const uint8_t *p, unsigned l, BOOL *ascii)
* l = length (bytes) of the utf-8 data
* o = pointer to current offset within the data
* n = pointer to either zero or the next pre-read part of a surrogate pair.
* The condition for having read the entire string is that the offset (*o)
* is the number of bytes in the string, and the unichar pointed to by *n
* is zero (meaning there is no second part of a surrogate pair remaining).
*/
static inline unichar
nextUTF8(const uint8_t *p, unsigned l, unsigned *o, unichar *n)
@ -4970,7 +4973,7 @@ nextUTF8(const uint8_t *p, unsigned l, unsigned *o, unichar *n)
unichar n = 0;
unsigned i = 0;
while (i < nxcslen)
while (i < nxcslen || n > 0)
{
u = nextUTF8((const uint8_t *)nxcsptr, nxcslen, &i, &n);
if (l++ == index)
@ -5018,14 +5021,14 @@ nextUTF8(const uint8_t *p, unsigned l, unsigned *o, unichar *n)
format: @"in %s, range { %u, %u } extends beyond string",
GSNameFromSelector(_cmd), aRange.location, aRange.length];
while (index < aRange.location && i < nxcslen)
while (index < aRange.location && (i < nxcslen || n > 0))
{
nextUTF8((const uint8_t *)nxcsptr, nxcslen, &i, &n);
index++;
}
if (index == aRange.location)
{
while (index < max && i < nxcslen)
while (index < max && (i < nxcslen || n > 0))
{
*buffer++ = nextUTF8((const uint8_t *)nxcsptr, nxcslen, &i, &n);
index++;
@ -5039,6 +5042,39 @@ nextUTF8(const uint8_t *p, unsigned l, unsigned *o, unichar *n)
}
}
/* Must match the implementation in NSString
*/
- (NSUInteger) hash
{
if (nxcslen > 0)
{
unsigned ret = 0;
unichar n = 0;
unsigned i = 0;
unichar c;
while (i < nxcslen)
{
c = nextUTF8((const uint8_t *)nxcsptr, nxcslen, &i, &n);
ret = (ret << 5) + ret + c;
}
if (0 != n)
{
ret = (ret << 5) + ret + n; // Add final character
}
ret &= 0x0fffffff;
if (ret == 0)
{
ret = 0x0fffffff;
}
return ret;
}
else
{
return 0x0ffffffe; /* Hash for an empty string. */
}
}
- (id) initWithBytes: (const void*)bytes
length: (NSUInteger)length
encoding: (NSStringEncoding)encoding