ruby-changes:3943
From: ko1@a...
Date: Tue, 12 Feb 2008 02:47:14 +0900 (JST)
Subject: [ruby-changes:3943] matz - Ruby:r15433 (trunk): * range.c (range_include): specialize single character string
matz 2008-02-12 02:46:52 +0900 (Tue, 12 Feb 2008)
New Revision: 15433
Modified files:
trunk/ChangeLog
trunk/object.c
trunk/parse.y
trunk/range.c
trunk/string.c
trunk/version.h
Log:
* range.c (range_include): specialize single character string
case (e.g. (?a ..?z).include(?x)) for performance.
[ruby-core:15481]
* string.c (rb_str_upto): specialize single character case.
* string.c (rb_str_hash): omit coderange scan for performance.
* object.c (rb_check_to_integer): check Fixnum first.
* object.c (rb_to_integer): ditto.
* string.c (rb_str_equal): inline memcmp to avoid unnecessary
rb_str_comparable().
* parse.y (rb_intern2): use US-ASCII encoding.
* parse.y (rb_intern_str): ditto.
http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/version.h?r1=15433&r2=15432&diff_format=u
http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/string.c?r1=15433&r2=15432&diff_format=u
http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/parse.y?r1=15433&r2=15432&diff_format=u
http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/ChangeLog?r1=15433&r2=15432&diff_format=u
http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/range.c?r1=15433&r2=15432&diff_format=u
http://svn.ruby-lang.org/cgi-bin/viewvc.cgi/trunk/object.c?r1=15433&r2=15432&diff_format=u
Index: ChangeLog
===================================================================
--- ChangeLog (revision 15432)
+++ ChangeLog (revision 15433)
@@ -1,3 +1,24 @@
+Tue Feb 12 02:42:27 2008 Yukihiro Matsumoto <matz@r...>
+
+ * range.c (range_include): specialize single character string
+ case (e.g. (?a ..?z).include(?x)) for performance.
+ [ruby-core:15481]
+
+ * string.c (rb_str_upto): specialize single character case.
+
+ * string.c (rb_str_hash): omit coderange scan for performance.
+
+ * object.c (rb_check_to_integer): check Fixnum first.
+
+ * object.c (rb_to_integer): ditto.
+
+ * string.c (rb_str_equal): inline memcmp to avoid unnecessary
+ rb_str_comparable().
+
+ * parse.y (rb_intern2): use US-ASCII encoding.
+
+ * parse.y (rb_intern_str): ditto.
+
Mon Feb 11 17:21:18 2008 Kouhei Sutou <kou@c...>
* lib/rss/rss.rb (RSS::VERSION), test/rss/test_version.rb:
Index: string.c
===================================================================
--- string.c (revision 15432)
+++ string.c (revision 15433)
@@ -1509,11 +1509,7 @@
int
rb_str_hash(VALUE str)
{
- int e = ENCODING_GET(str);
- if (e && is_ascii_string(str)) {
- e = 0;
- }
- return hash((const void *)RSTRING_PTR(str), RSTRING_LEN(str), e);
+ return hash((const void *)RSTRING_PTR(str), RSTRING_LEN(str), 0);
}
/*
@@ -1593,6 +1589,8 @@
VALUE
rb_str_equal(VALUE str1, VALUE str2)
{
+ int len;
+
if (str1 == str2) return Qtrue;
if (TYPE(str2) != T_STRING) {
if (!rb_respond_to(str2, rb_intern("to_str"))) {
@@ -1601,8 +1599,8 @@
return rb_equal(str2, str1);
}
if (!rb_str_comparable(str1, str2)) return Qfalse;
- if (RSTRING_LEN(str1) == RSTRING_LEN(str2) &&
- rb_str_cmp(str1, str2) == 0) {
+ if (RSTRING_LEN(str1) == (len = RSTRING_LEN(str2)) &&
+ memcmp(RSTRING_PTR(str1), RSTRING_PTR(str2), len) == 0) {
return Qtrue;
}
return Qfalse;
@@ -2295,14 +2293,30 @@
VALUE current, after_end;
ID succ;
int n, excl;
+ rb_encoding *enc;
rb_scan_args(argc, argv, "11", &end, &exclusive);
- rb_enc_check(beg, end);
excl = RTEST(exclusive);
succ = rb_intern("succ");
StringValue(end);
+ enc = rb_enc_check(beg, end);
+ if (RSTRING_LEN(beg) == 1 && RSTRING_LEN(end) == 1 &&
+ is_ascii_string(beg) && is_ascii_string(end)) {
+ char c = RSTRING_PTR(beg)[0];
+ char e = RSTRING_PTR(end)[0];
+
+ if (c > e || (excl && c == e)) return beg;
+ for (;;) {
+ rb_yield(rb_enc_str_new(&c, 1, enc));
+ if (!excl && c == e) break;
+ c++;
+ if (excl && c == e) break;
+ }
+ return beg;
+ }
n = rb_str_cmp(beg, end);
if (n > 0 || (excl && n == 0)) return beg;
+
after_end = rb_funcall(end, succ, 0, 0);
current = beg;
while (!rb_str_equal(current, after_end)) {
@@ -2311,7 +2325,6 @@
current = rb_funcall(current, succ, 0, 0);
StringValue(current);
if (excl && rb_str_equal(current, end)) break;
- StringValue(current);
if (RSTRING_LEN(current) > RSTRING_LEN(end) || RSTRING_LEN(current) == 0)
break;
}
Index: object.c
===================================================================
--- object.c (revision 15432)
+++ object.c (revision 15433)
@@ -1940,7 +1940,10 @@
static VALUE
rb_to_integer(VALUE val, const char *method)
{
- VALUE v = convert_type(val, "Integer", method, Qtrue);
+ VALUE v;
+
+ if (FIXNUM_P(val)) return val;
+ v = convert_type(val, "Integer", method, Qtrue);
if (!rb_obj_is_kind_of(v, rb_cInteger)) {
char *cname = rb_obj_classname(val);
rb_raise(rb_eTypeError, "can't convert %s to Integer (%s#%s gives %s)",
@@ -1952,7 +1955,10 @@
VALUE
rb_check_to_integer(VALUE val, const char *method)
{
- VALUE v = convert_type(val, "Integer", method, Qfalse);
+ VALUE v;
+
+ if (FIXNUM_P(val)) return val;
+ v = convert_type(val, "Integer", method, Qfalse);
if (!rb_obj_is_kind_of(v, rb_cInteger)) {
return Qnil;
}
Index: range.c
===================================================================
--- range.c (revision 15432)
+++ range.c (revision 15433)
@@ -10,6 +10,7 @@
**********************************************************************/
#include "ruby/ruby.h"
+#include "ruby/encoding.h"
VALUE rb_cRange;
static ID id_cmp, id_succ, id_beg, id_end, id_excl;
@@ -748,6 +749,27 @@
}
return Qfalse;
}
+ else if (TYPE(beg) == T_STRING && TYPE(end) == T_STRING &&
+ RSTRING_LEN(beg) == 1 && RSTRING_LEN(end) == 1) {
+ rb_encoding *enc = rb_enc_check(beg, end);
+
+ if (NIL_P(val)) return Qfalse;
+ if (TYPE(val) == T_STRING) {
+ if (RSTRING_LEN(val) == 0 || RSTRING_LEN(val) > 1)
+ return Qfalse;
+ else {
+ char b = RSTRING_PTR(beg)[0];
+ char e = RSTRING_PTR(end)[0];
+ char v = RSTRING_PTR(val)[0];
+
+ if (ISASCII(b) && ISASCII(e) && ISASCII(v)) {
+ if (b <= v && v < e) return Qtrue;
+ if (!EXCL(range) && v == e) return Qtrue;
+ return Qfalse;
+ }
+ }
+ }
+ }
/* TODO: ruby_frame->this_func = rb_intern("include?"); */
return rb_call_super(1, &val);
}
Index: parse.y
===================================================================
--- parse.y (revision 15432)
+++ parse.y (revision 15433)
@@ -271,7 +271,7 @@
#define STR_NEW0() rb_enc_str_new(0,0,rb_usascii_encoding())
#define STR_NEW2(p) rb_enc_str_new((p),strlen(p),parser->enc)
#define STR_NEW3(p,n,e,func) parser_str_new((p),(n),(e),(func),parser->enc)
-#define STR_ENC(m) ((m)?parser->enc:rb_ascii8bit_encoding())
+#define STR_ENC(m) ((m)?parser->enc:rb_usascii_encoding())
#define ENC_SINGLE(cr) ((cr)==ENC_CODERANGE_7BIT)
#define TOK_INTERN(mb) rb_intern3(tok(), toklen(), STR_ENC(mb))
@@ -9043,7 +9043,7 @@
ID
rb_intern2(const char *name, long len)
{
- return rb_intern3(name, len, rb_ascii8bit_encoding());
+ return rb_intern3(name, len, rb_usascii_encoding());
}
#undef rb_intern
@@ -9056,14 +9056,16 @@
ID
rb_intern_str(VALUE str)
{
- int idx = 0;
+ rb_encoding *enc;
ID id;
- if (rb_enc_str_coderange(str) != ENC_CODERANGE_7BIT) {
- idx = rb_enc_get_index(str);
+ if (rb_enc_str_coderange(str) == ENC_CODERANGE_7BIT) {
+ enc = rb_usascii_encoding();
}
- id = rb_intern3(RSTRING_PTR(str), RSTRING_LEN(str),
- rb_enc_from_index(idx));
+ else {
+ enc = rb_enc_get(str);
+ }
+ id = rb_intern3(RSTRING_PTR(str), RSTRING_LEN(str), enc);
RB_GC_GUARD(str);
return id;
}
Index: version.h
===================================================================
--- version.h (revision 15432)
+++ version.h (revision 15433)
@@ -1,7 +1,7 @@
#define RUBY_VERSION "1.9.0"
-#define RUBY_RELEASE_DATE "2008-02-11"
+#define RUBY_RELEASE_DATE "2008-02-12"
#define RUBY_VERSION_CODE 190
-#define RUBY_RELEASE_CODE 20080211
+#define RUBY_RELEASE_CODE 20080212
#define RUBY_PATCHLEVEL 0
#define RUBY_VERSION_MAJOR 1
@@ -9,7 +9,7 @@
#define RUBY_VERSION_TEENY 0
#define RUBY_RELEASE_YEAR 2008
#define RUBY_RELEASE_MONTH 2
-#define RUBY_RELEASE_DAY 11
+#define RUBY_RELEASE_DAY 12
#ifdef RUBY_EXTERN
RUBY_EXTERN const char ruby_version[];
--
ML: ruby-changes@q...
Info: http://www.atdot.net/~ko1/quickml/