1berry/ruby
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

* string.c (rb_enc_strlen_cr): get length with coderange scan.

Browse Source 
* string.c (str_strlen): use rb_enc_strlen_cr. [ruby-dev:33849]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@15550 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This commit is contained in:
naruse 2008-02-19 12:18:03 +00:00
parent e4021207fb
commit f1c975b87a
2 changed files with 65 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
ChangeLog
View File

@ -1,3 +1,9 @@
Tue Feb 19 21:11:49 2008 NARUSE, Yui <naruse@ruby-lang.org>
* string.c (rb_enc_strlen_cr): get length with coderange scan.
* string.c (str_strlen): use rb_enc_strlen_cr. [ruby-dev:33849]
Tue Feb 19 20:49:49 2008 Nobuyoshi Nakada <nobu@ruby-lang.org> Tue Feb 19 20:49:49 2008 Nobuyoshi Nakada <nobu@ruby-lang.org>
* eval.c (rb_raise_jump): moved adjustment for control frame. * eval.c (rb_raise_jump): moved adjustment for control frame.

61
string.c
View File

@ -128,12 +128,12 @@ search_nonascii(const char *p, const char *e)
const unsigned long *s, *t; const unsigned long *s, *t;
const VALUE lowbits = sizeof(unsigned long) - 1; const VALUE lowbits = sizeof(unsigned long) - 1;
s = (const unsigned long*)(~lowbits & ((VALUE)p + lowbits)); s = (const unsigned long*)(~lowbits & ((VALUE)p + lowbits));
t = (const unsigned long*)(~lowbits & (VALUE)e);
while (p < (const char *)s) { while (p < (const char *)s) {
if (!ISASCII(*p)) if (!ISASCII(*p))
return p; return p;
p++; p++;
} }
t = (const unsigned long*)(~lowbits & (VALUE)e);
while (s < t) { while (s < t) {
if (*s & NONASCII_MASK) { if (*s & NONASCII_MASK) {
t = s; t = s;
@ -619,10 +619,63 @@ rb_enc_strlen(const char *p, const char *e, rb_encoding *enc)
return c; return c;
} }
long
rb_enc_strlen_cr(const char *p, const char *e, rb_encoding *enc, int *cr)
{
long c;
const char *q;
int ret;
*cr = 0;
if (rb_enc_mbmaxlen(enc) == rb_enc_mbminlen(enc)) {
return (e - p) / rb_enc_mbminlen(enc);
}
else if (rb_enc_asciicompat(enc)) {
c = 0;
while (p < e) {
if (ISASCII(*p)) {
q = search_nonascii(p, e);
if (!q) {
return c + (e - p);
}
c += q - p;
p = q;
}
ret = rb_enc_precise_mbclen(p, e, enc);
if (MBCLEN_CHARFOUND_P(ret)) {
*cr |= ENC_CODERANGE_VALID;
p += MBCLEN_CHARFOUND_LEN(ret);
}
else {
*cr = ENC_CODERANGE_BROKEN;
p++;
}
c++;
}
if (!*cr) *cr = ENC_CODERANGE_7BIT;
return c;
}
for (c=0; p<e; c++) {
ret = rb_enc_precise_mbclen(p, e, enc);
if (MBCLEN_CHARFOUND_P(ret)) {
*cr |= ENC_CODERANGE_VALID;
p += MBCLEN_CHARFOUND_LEN(ret);
}
else {
*cr = ENC_CODERANGE_BROKEN;
p++;
}
}
if (!*cr) *cr = ENC_CODERANGE_7BIT;
return c;
}
static long static long
str_strlen(VALUE str, rb_encoding *enc) str_strlen(VALUE str, rb_encoding *enc)
{ {
const char *p, *e; const char *p, *e;
int n, cr;
if (single_byte_optimizable(str)) return RSTRING_LEN(str); if (single_byte_optimizable(str)) return RSTRING_LEN(str);
if (!enc) enc = STR_ENC_GET(str); if (!enc) enc = STR_ENC_GET(str);
@ -661,7 +714,11 @@ str_strlen(VALUE str, rb_encoding *enc)
return len; return len;
} }
#endif #endif
return rb_enc_strlen(p, e, enc); n = rb_enc_strlen_cr(p, e, enc, &cr);
if (cr) {
ENC_CODERANGE_SET(str, cr);
}
return n;
} }
/* /*