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

Add str_enc_copy_direct

Browse Source 
This commit adds str_enc_copy_direct, which is like str_enc_copy but
does not check the frozen status of str1 and does not check the validity
of the encoding of str2. This makes certain string operations ~5% faster.

```ruby
puts(Benchmark.measure do
  100_000_000.times do
    "a".downcase
  end
end)
```

Before this patch:

```
  7.587598   0.040858   7.628456 (  7.669022)
```

After this patch:

```
  7.133128   0.039809   7.172937 (  7.183124)
```
This commit is contained in:
Peter Zhu 2023-01-11 14:25:34 -05:00
parent 207f8d0027
commit bfc887f391
Notes: git 2023-01-12 14:06:43 +00:00 
Merged: https://github.com/ruby/ruby/pull/7106
1 changed files with 19 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
string.c
View File

@ -768,6 +768,20 @@ str_enc_copy(VALUE str1, VALUE str2)
rb_enc_set_index(str1, ENCODING_GET(str2)); rb_enc_set_index(str1, ENCODING_GET(str2));
} }
/* Like str_enc_copy, but does not check frozen status of str1.
* You should use this only if you're certain that str1 is not frozen. */
static inline void
str_enc_copy_direct(VALUE str1, VALUE str2)
{
int inlined_encoding = RB_ENCODING_GET_INLINED(str2);
if (inlined_encoding == ENCODING_INLINE_MAX) {
rb_enc_set_index(str1, rb_enc_get_index(str2));
}
else {
ENCODING_SET_INLINED(str1, inlined_encoding);
}
}
static void static void
rb_enc_cr_str_copy_for_substr(VALUE dest, VALUE src) rb_enc_cr_str_copy_for_substr(VALUE dest, VALUE src)
{ {
@ -6156,7 +6170,7 @@ str_byte_substr(VALUE str, long beg, long len, int empty)
VALUE str2 = str_subseq(str, beg, len); VALUE str2 = str_subseq(str, beg, len);
str_enc_copy(str2, str); str_enc_copy_direct(str2, str);
if (RSTRING_LEN(str2) == 0) { if (RSTRING_LEN(str2) == 0) {
if (!rb_enc_asciicompat(STR_ENC_GET(str))) if (!rb_enc_asciicompat(STR_ENC_GET(str)))
@ -6382,7 +6396,7 @@ rb_str_reverse(VALUE str)
} }
} }
STR_SET_LEN(rev, RSTRING_LEN(str)); STR_SET_LEN(rev, RSTRING_LEN(str));
str_enc_copy(rev, str); str_enc_copy_direct(rev, str);
ENC_CODERANGE_SET(rev, cr); ENC_CODERANGE_SET(rev, cr);
return rev; return rev;
@ -7337,7 +7351,7 @@ rb_str_casemap(VALUE source, OnigCaseFoldType *flags, rb_encoding *enc)
RB_GC_GUARD(buffer_anchor); RB_GC_GUARD(buffer_anchor);
/* TODO: check about string terminator character */ /* TODO: check about string terminator character */
str_enc_copy(target, source); str_enc_copy_direct(target, source);
/*ENC_CODERANGE_SET(mapped, cr);*/ /*ENC_CODERANGE_SET(mapped, cr);*/
return target; return target;
@ -7468,7 +7482,7 @@ rb_str_upcase(int argc, VALUE *argv, VALUE str)
enc = str_true_enc(str); enc = str_true_enc(str);
if (case_option_single_p(flags, enc, str)) { if (case_option_single_p(flags, enc, str)) {
ret = rb_str_new(RSTRING_PTR(str), RSTRING_LEN(str)); ret = rb_str_new(RSTRING_PTR(str), RSTRING_LEN(str));
str_enc_copy(ret, str); str_enc_copy_direct(ret, str);
upcase_single(ret); upcase_single(ret);
} }
else if (flags&ONIGENC_CASE_ASCII_ONLY) { else if (flags&ONIGENC_CASE_ASCII_ONLY) {
@ -7570,7 +7584,7 @@ rb_str_downcase(int argc, VALUE *argv, VALUE str)
enc = str_true_enc(str); enc = str_true_enc(str);
if (case_option_single_p(flags, enc, str)) { if (case_option_single_p(flags, enc, str)) {
ret = rb_str_new(RSTRING_PTR(str), RSTRING_LEN(str)); ret = rb_str_new(RSTRING_PTR(str), RSTRING_LEN(str));
str_enc_copy(ret, str); str_enc_copy_direct(ret, str);
downcase_single(ret); downcase_single(ret);
} }
else if (flags&ONIGENC_CASE_ASCII_ONLY) { else if (flags&ONIGENC_CASE_ASCII_ONLY) {