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

* string.c (rb_str_dump): use \u{ff} escape for UTF-8 encoding

Browse Source 
string.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19897 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This commit is contained in:
matz 2008-10-23 06:53:55 +00:00
parent 472156c236
commit 5a8db29b59
2 changed files with 38 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
ChangeLog
View File

@ -7,6 +7,11 @@ Thu Oct 23 14:08:07 2008 Nobuyoshi Nakada <nobu@ruby-lang.org>
converted before the proto so that the #to_int of the former cannot converted before the proto so that the #to_int of the former cannot
alter the latter. alter the latter.
Thu Oct 23 10:55:08 2008 Yukihiro Matsumoto <matz@ruby-lang.org>
* string.c (rb_str_dump): use \u{ff} escape for UTF-8 encoding
string.
Thu Oct 23 09:26:22 2008 NAKAMURA Usaku <usa@ruby-lang.org> Thu Oct 23 09:26:22 2008 NAKAMURA Usaku <usa@ruby-lang.org>
* ext/socket/socket.c (sock_s_getservbyport): cast to get rid of * ext/socket/socket.c (sock_s_getservbyport): cast to get rid of

40
string.c
View File

@ -4090,11 +4090,12 @@ rb_str_inspect(VALUE str)
VALUE VALUE
rb_str_dump(VALUE str) rb_str_dump(VALUE str)
{ {
rb_encoding *enc0 = rb_enc_get(str); rb_encoding *enc = rb_enc_get(str);
long len; long len;
const char *p, *pend; const char *p, *pend;
char *q, *qend; char *q, *qend;
VALUE result; VALUE result;
int u8 = (enc == rb_utf8_encoding());
len = 2; /* "" */ len = 2; /* "" */
p = RSTRING_PTR(str); pend = p + RSTRING_LEN(str); p = RSTRING_PTR(str); pend = p + RSTRING_LEN(str);
@ -4117,14 +4118,25 @@ rb_str_dump(VALUE str)
len++; len++;
} }
else { else {
if (u8) { /* \u{NN} */
char buf[32];
int n = rb_enc_precise_mbclen(p-1, pend, enc) - 1;
if (MBCLEN_CHARFOUND_P(n)) {
int cc = rb_enc_codepoint(p-1, pend, enc);
sprintf(buf, "%x", cc);
len += strlen(buf)+4;
p += n;
break;
}
}
len += 4; /* \xNN */ len += 4; /* \xNN */
} }
break; break;
} }
} }
if (!rb_enc_asciicompat(enc0)) { if (!rb_enc_asciicompat(enc)) {
len += 19; /* ".force_encoding('')" */ len += 19; /* ".force_encoding('')" */
len += strlen(enc0->name); len += strlen(enc->name);
} }
result = rb_str_new5(str, 0, len); result = rb_str_new5(str, 0, len);
@ -4180,19 +4192,31 @@ rb_str_dump(VALUE str)
} }
else { else {
*q++ = '\\'; *q++ = '\\';
if (u8) {
int n = rb_enc_precise_mbclen(p-1, pend, enc) - 1;
if (MBCLEN_CHARFOUND_P(n)) {
int cc = rb_enc_codepoint(p-1, pend, enc);
p += n;
sprintf(q, "u{%x}", cc);
q += strlen(q);
continue;
}
}
sprintf(q, "x%02X", c); sprintf(q, "x%02X", c);
q += 3; q += 3;
} }
} }
*q++ = '"'; *q++ = '"';
if (!rb_enc_asciicompat(enc0)) { *q = '\0';
sprintf(q, ".force_encoding(\"%s\")", enc0->name); if (!rb_enc_asciicompat(enc)) {
enc0 = rb_ascii8bit_encoding(); sprintf(q, ".force_encoding(\"%s\")", enc->name);
enc = rb_ascii8bit_encoding();
} }
// STR_SET_LEN(result, strlen(RSTRING_PTR(result)));
OBJ_INFECT(result, str); OBJ_INFECT(result, str);
/* result from dump is ASCII */ /* result from dump is ASCII */
rb_enc_associate(result, enc0); rb_enc_associate(result, enc);
ENC_CODERANGE_SET(result, ENC_CODERANGE_7BIT);
return result; return result;
} }