# /#{}/e.encoding

**URL:** <https://www.ruby-forum.com/t/e-encoding/127205>\
**Category:** Development\
**Created:** [January 26, 2008, 4:15am UTC](https://www.ruby-forum.com/t/e-encoding/127205 "2008-01-26T04:15:03Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Tanaka\_A](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/c67d28/32.png) [@Tanaka\_A](https://www.ruby-forum.com/u/Tanaka_A)\
**Post date:** [January 26, 2008, 4:15am UTC](https://www.ruby-forum.com/t/e-encoding/127205/1 "2008-01-26T04:15:03Z")

</div>

e$B0J2\<$N$h$&$Ke(B /#{}/e e$B$Ne(B encoding e$B$,e(B US-ASCII  
e$B$K$J$j$^$9!#e(B

% ./ruby -ve ‘p(/#{}/e.encoding)’  
ruby 1.9.0 (2008-01-26 revision 15237) [i686-linux]  
#Encoding:US-ASCII

e$B@55,I=8=$N%=!\<%9$r9=@.$9$kJ8;zNs$NO"7k;~$Ke(B EUC-JP e$B$Je(B  
encoding e$B$,\>C$($F$7$^$&$N$,M}M3$J$h$&$G$9!#e(B

e$B$b$H$b$H$3$NO"7k$KC1$J$kJ8;zNsO"7k$r;H$&$N$O!"%(%9%1!\<%W$5$le(B  
e$B$?CfESH\>C\<$JHse(B ASCII e$BJ8;z$,O"7k$5$l$FCfESH\>C\<$G$J$$$b$N$K$Je(B  
e$B$k$N$r8!=P$G$-$J$$$H$$$&LdBj$,$"$j$^$9!#$J$N$G!"e(B  
Regexp.preprocess e$B$H$$$&%a%=%C%I$r?7@_$7$F!"$=$3$GO"7k$r9T$$!"e(B  
e$B%(%s%3!\<%G%#%s%0$NEAGE$b0JA0F1MM$K$K$5$l$k$h$&$K$7$F$_$^$7$?!#e(B

# Index: re.c

— re.c (revision 15249)  
+++ re.c (working copy)  
@@ -1942,36 +1942,51 @@ rb\_reg\_check\_preprocess(VALUE str)  
return Qnil;  
}

-#if 0  
static VALUE  
-rb\_reg\_preprocess\_obj(VALUE str,

- 

```
   rb_encoding **fixed_enc, onig_errmsg_buffer err)

```

+rb\_reg\_preprocess\_m(int argc, VALUE \*argv, VALUE klass)  
{

- VALUE buf;
- char \*p, \*end;
- rb\_encoding \*enc;

- rb\_encoding \*fixed\_enc = 0;
- onig\_errmsg\_buffer err = “”;
- int i;
- VALUE result = 0;

- StringValue(str);
- p = RSTRING\_PTR(str);
- end = p + RSTRING\_LEN(str);
- enc = rb\_enc\_get(str);

- if (argc == 0) {
- 

```
   rb_raise(rb_eArgError, "no arguments given");

```

- }

- buf = rb\_reg\_preprocess(p, end, enc, fixed\_enc, err);
- RB\_GC\_GUARD(str);
- return buf;  
-}

- for (i = 0; i \< argc; i++) {
- 

```
   VALUE str = argv[i];

```

- 

```
   VALUE buf;

```

- 

```
   char *p, *end;

```

- 

```
   rb_encoding *enc;

```

- 
- 

```
   StringValue(str);

```

- 

```
   p = RSTRING_PTR(str);

```

- 

```
   end = p + RSTRING_LEN(str);

```

- 

```
   enc = rb_enc_get(str);

```

- 
- 

```
   buf = rb_reg_preprocess(p, end, enc, &fixed_enc, err);

```

- 

```
   RB_GC_GUARD(str);

```

- 
- 

```
   if (buf == Qnil)

```

- 

```
       rb_raise(rb_eArgError, "%s", err);

```

- 
- 

```
   if (i == 0) {

```

- 

```
       /* The encoding of the first fragment is the encoding

```

- 

```
        * given by the regexp option or script encoding. */

```

- 

```
       if (fixed_enc == 0) {

```

- 

```
           rb_enc_copy(buf, str);

```

- 

```
       }

```

- 

```
   }

```

-static VALUE  
-rb\_reg\_preprocess\_m(VALUE klass, VALUE obj)  
-{

- rb\_encoding \*fixed\_enc = 0;
- onig\_errmsg\_buffer err = “”;
- VALUE str = rb\_reg\_preprocess\_obj(obj, &fixed\_enc, err);
- if (str == Qnil)
- 

```
   rb_raise(rb_eArgError, "%s", err);

```

- return rb\_assoc\_new(str, fixed\_enc ? Qtrue : Qfalse);

- 

```
   if (!result)

```

- 

```
       result = buf;

```

- 

```
   else

```

- 

```
       rb_str_buf_append(result, buf);

```

- }
- 
- return result;  
}  
-#endif

static int  
rb\_reg\_initialize(VALUE obj, const char \*s, int len, rb\_encoding \*enc,  
@@ -3035,9 +3050,7 @@ Init\_Regexp(void)  
rb\_define\_singleton\_method(rb\_cRegexp, “last\_match”,  
rb\_reg\_s\_last\_match, -1);  
rb\_define\_singleton\_method(rb\_cRegexp, “try\_convert”,  
rb\_reg\_s\_try\_convert, 1);

-#if 0

- rb\_define\_singleton\_method(rb\_cRegexp, “preprocess”,  
rb\_reg\_preprocess\_m, 1);  
-#endif

- rb\_define\_singleton\_method(rb\_cRegexp, “preprocess”,  
rb\_reg\_preprocess\_m, -1);

static int  
-compile\_dstr(rb\_iseq\_t \*iseq, LINK\_ANCHOR \*ret, NODE \* node)  
+compile\_dstr\_fragments(rb\_iseq\_t \*iseq, LINK\_ANCHOR \*ret, NODE \* node,  
int \*cntp)  
{  
NODE \*list = node-\>nd\_next;  
VALUE lit = node-\>nd\_lit;  
@@ -1862,12 +1862,31 @@ compile\_dstr(rb\_iseq\_t \*iseq, LINK\_ANCHO  
cnt++;  
list = list-\>nd\_next;  
}

- \*cntp = cnt;
- 
- return COMPILE\_OK;  
+}

+static int  
+compile\_dstr(rb\_iseq\_t \*iseq, LINK\_ANCHOR \*ret, NODE \* node)  
+{

- int cnt;
- compile\_dstr\_fragments(iseq, ret, node, &cnt);  
ADD\_INSN1(ret, nd\_line(node), concatstrings, INT2FIX(cnt));  
return COMPILE\_OK;  
}

static int  
+compile\_dregx(rb\_iseq\_t \*iseq, LINK\_ANCHOR \*ret, NODE \* node)  
+{

- int cnt;
- ADD\_INSN1(ret, nd\_line(node), putobject, rb\_cRegexp);
- compile\_dstr\_fragments(iseq, ret, node, &cnt);
- ADD\_SEND(ret, nd\_line(node), ID2SYM(rb\_intern(“preprocess”)),  
INT2FIX(cnt));
- return COMPILE\_OK;  
+}
- 

+static int  
compile\_branch\_condition(rb\_iseq\_t \*iseq, LINK\_ANCHOR \*ret, NODE \*  
cond,  
LABEL \*then\_label, LABEL \*else\_label)  
{  
@@ -4077,7 +4096,7 @@ iseq\_compile\_each(rb\_iseq\_t \*iseq, LINK\_  
break;  
}  
case NODE\_DREGX:{

- compile\_dstr(iseq, ret, node);

- compile\_dregx(iseq, ret, node);  
ADD\_INSN1(ret, nd\_line(node), toregexp, INT2FIX(node-\>nd\_cflag));

- compile\_dstr(iseq, ret, node);

- compile\_dregx(iseq, ret, node);  
ADD\_INSN1(ret, nd\_line(node), toregexp, INT2FIX(node-\>nd\_cflag));

- #assert\_raise(ArgumentError) { eval(s("/\xc2\xa1#{r}/s")) }
- #assert\_raise(ArgumentError) { eval(s("/#{r}\xc2\xa1/s")) }

- assert\_raise(ArgumentError) { eval(s("/\xc2\xa1#{r}/s")) }

- assert\_raise(ArgumentError) { eval(s("/#{r}\xc2\xa1/s")) }

- #assert\_raise(ArgumentError) { /\xc2\xa1#{r}/s }

- assert\_raise(ArgumentError) { /\xc2\xa1#{r}/s }

- #assert\_raise(ArgumentError) { /\xc2\xa1#{r}/s }

- assert\_raise(ArgumentError) { /\xc2\xa1#{r}/s }  
end

def test\_begin\_end\_offset  
@@ -560,7 +560,7 @@ class TestM17N \< Test::Unit::TestCase  
}  
assert\_regexp\_fixed\_ascii8bit(/#{}\xc2\xa1/n)  
assert\_regexp\_fixed\_ascii8bit(/\xc2\xa1#{}/n)

- #assert\_raise(SyntaxError) { s1, s2 = s(’\xc2’), s(’\xa1’);  
/#{s1}#{s2}/ }

- assert\_nothing\_raised { s1, s2 = a(’\xc2’), a(’\xa1’); /#{s1}#{s2}/  
}  
end

def test\_dynamic\_eucjp\_regexp  
@@ -570,7 +570,7 @@ class TestM17N \< Test::Unit::TestCase  
assert\_raise(SyntaxError) { eval(’/\xc2#{}/e’) }  
assert\_raise(SyntaxError) { eval(’/#{}\xc2/e’) }  
assert\_raise(SyntaxError) { eval(’/\xc2#{}\xa1/e’) }

- #assert\_raise(SyntaxError) { s1, s2 = e(’\xc2’), e(’\xa1’);  
/#{s1}#{s2}/ }

- assert\_raise(ArgumentError) { s1, s2 = e(’\xc2’), e(’\xa1’);  
/#{s1}#{s2}/ }  
end

def test\_dynamic\_sjis\_regexp  
@@ -580,7 +580,7 @@ class TestM17N \< Test::Unit::TestCase  
assert\_raise(SyntaxError) { eval(’/\x81#{}/s’) }  
assert\_raise(SyntaxError) { eval(’/#{}\x81/s’) }  
assert\_raise(SyntaxError) { eval(’/\x81#{}\xa1/s’) }

- #assert\_raise(SyntaxError) { s1, s2 = s(’\x81’), s(’\xa1’);  
/#{s1}#{s2}/ }

- assert\_raise(ArgumentError) { s1, s2 = s(’\x81’), s(’\xa1’);  
/#{s1}#{s2}/ }  
end

def test\_dynamic\_utf8\_regexp  
@@ -590,7 +590,7 @@ class TestM17N \< Test::Unit::TestCase  
assert\_raise(SyntaxError) { eval(’/\xc2#{}/u’) }  
assert\_raise(SyntaxError) { eval(’/#{}\xc2/u’) }  
assert\_raise(SyntaxError) { eval(’/\xc2#{}\xa1/u’) }

- #assert\_raise(SyntaxError) { s1, s2 = u(’\xc2’), u(’\xa1’);  
/#{s1}#{s2}/ }

- assert\_raise(ArgumentError) { s1, s2 = u(’\xc2’), u(’\xa1’);  
/#{s1}#{s2}/ }  
end

def test\_regexp\_unicode

---

<div class="post-metadata">

**Author:** ![Tanaka\_A](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/c67d28/32.png) [@Tanaka\_A](https://www.ruby-forum.com/u/Tanaka_A)\
**Post date:** [January 29, 2008, 7:28am UTC](https://www.ruby-forum.com/t/e-encoding/127205/2 "2008-01-29T07:28:51Z")

</div>

e$B$^$D$b$He(B e$B$f$-$R$m$G$9e(B

[ruby-dev:33484]e$B$G:EB%$5$l$?$N$G!#e(B

In message “Re: [ruby-dev:33400] /#{}/e.encoding”  
on Sat, 26 Jan 2008 12:14:00 +0900, Tanaka A. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)  
writes:  
|  
|e$B0J2\<$N$h$&$Ke(B /#{}/e e$B$Ne(B encoding e$B$,e(B US-ASCII e$B$K$J$j$^$9!#e(B  
|  
| % ./ruby -ve ‘p(/#{}/e.encoding)’  
| ruby 1.9.0 (2008-01-26 revision 15237) [i686-linux]  
| #Encoding:US-ASCII  
|  
|e$B@55,I=8=$N%=!\<%9$r9=@.$9$kJ8;zNs$NO"7k;~$Ke(B EUC-JP e$B$Je(B  
|encoding e$B$,\>C$($F$7$^$&$N$,M}M3$J$h$&$G$9!#e(B

e$B$J$k$[$I!#e(B

|e$B$b$H$b$H$3$NO"7k$KC1$J$kJ8;zNsO"7k$r;H$&$N$O!“%(%9%1!\<%W$5$le(B  
|e$B$?CfESH\>C\<$JHse(B ASCII e$BJ8;z$,O"7k$5$l$FCfESH\>C\<$G$J$$$b$N$K$Je(B  
|e$B$k$N$r8!=P$G$-$J$$$H$$$&LdBj$,$”$j$^$9!#$J$N$G!"e(B  
|Regexp.preprocess e$B$H$$$&%a%=%C%I$r?7@_$7$F!"$=$3$GO"7k$r9T$$!"e(B  
|e$B%(%s%3!\<%G%#%s%0$NEAGE$b0JA0F1MM$K$K$5$l$k$h$&$K$7$F$_$^$7$?!#e(B

e$B$d$O$j%3%s%Q%$%kFbIt$N;v\>p$r%a%=%C%I$H$7$F8+$;$A$c$&$N$K$Oe(B  
e$B$A$g$C$HDq93$,$"$j$^$9!#$?$H$($Pe(Bto\_stre$B$N$h$&$J%i%s%?%$%`$J$ie(B  
e$B5$$K$J$i$J$$$N$K$J!#e(B

e$B$G!“L?Na$rA}$d$9$N$OK\>$^$7$/$J$$$H$$$&$3$H$J$s$G$9$,!”$J$i$Pe(B  
toregexpe$BL?Na$,e(Bcnte$B%\*%Z%i%s%I$r\<u$1IU$1$k$h$&$K$7$F!"$=$NCf$Ge(B  
preprocesse$B4X?t$r8F$V$h$&$K$7$F$O$I$&$G$7$g$&!)e(B

```
                            e$B$^$D$b$He(B e$B$f$-$R$me(B /:|)

```

---

<div class="post-metadata">

**Author:** ![Tanaka\_A](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/c67d28/32.png) [@Tanaka\_A](https://www.ruby-forum.com/u/Tanaka_A)\
**Post date:** [January 29, 2008, 9:06am UTC](https://www.ruby-forum.com/t/e-encoding/127205/3 "2008-01-29T09:06:55Z")

</div>

In article [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid),  
Yukihiro M. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) writes:

> e$B$d$O$j%3%s%Q%$%kFbIt$N;v\>p$r%a%=%C%I$H$7$F8+$;$A$c$&$N$K$Oe(B  
> e$B$A$g$C$HDq93$,$"$j$^$9!#$?$H$($Pe(Bto\_stre$B$N$h$&$J%i%s%?%$%`$J$ie(B  
> e$B5$$K$J$i$J$$$N$K$J!#e(B
> 
> e$B$G!“L?Na$rA}$d$9$N$OK\>$^$7$/$J$$$H$$$&$3$H$J$s$G$9$,!”$J$i$Pe(B  
> toregexpe$BL?Na$,e(Bcnte$B%\*%Z%i%s%I$r\<u$1IU$1$k$h$&$K$7$F!"$=$NCf$Ge(B  
> preprocesse$B4X?t$r8F$V$h$&$K$7$F$O$I$&$G$7$g$&!)e(B

e$B$“$!!”$J$k$[$I!#$=$&$9$l$PM\>7W$J$b$N$rI=$K=P$5$:$K$G$-$^$9$M!#e(B

e$B$=$&$$$&$U$&$K$7$F%3%\_%C%H$7$^$7$?!#e(B

---

<div class="post-metadata">

**Author:** ![Tanaka\_A](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/c67d28/32.png) [@Tanaka\_A](https://www.ruby-forum.com/u/Tanaka_A)\
**Post date:** [January 29, 2008, 9:32am UTC](https://www.ruby-forum.com/t/e-encoding/127205/4 "2008-01-29T09:32:13Z")

</div>

e$B!!$5$5$@$G$9!#e(B

Yukihiro M. wrote:

> e$B$G!“L?Na$rA}$d$9$N$OK\>$^$7$/$J$$$H$$$&$3$H$J$s$G$9$,!”$J$i$Pe(B  
> toregexpe$BL?Na$,e(Bcnte$B%\*%Z%i%s%I$r\<u$1IU$1$k$h$&$K$7$F!"$=$NCf$Ge(B  
> preprocesse$B4X?t$r8F$V$h$&$K$7$F$O$I$&$G$7$g$&!)e(B

e$B!!8D?ME\*$K$O!“L?Na$O2DG=$J$i%a%=%C%I$KDI$$=P$7$?$$$J$!!”$H;W$C$F$$$^e(B  
e$B$9!#$=$&$$$&FCJL$J5!G=$O!“e(Bfreeze  
e$B$7$?%_%V%8%’%/%H$NFC0[%a%=%C%I$K$7e(B  
e$B$F$_$/$H$+!#Nc$($P!”%/%i%9Dj5A!"%a%=%C%IDj5A!"e(Balias, undef  
e$B$J$s$+$Oe(B  
e$BA4It30$KDI$$=P$7$?$$$J$!$H!#e(B

---

<div class="post-metadata">

**Author:** ![Tanaka\_A](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/c67d28/32.png) [@Tanaka\_A](https://www.ruby-forum.com/u/Tanaka_A)\
**Post date:** [January 29, 2008, 10:05am UTC](https://www.ruby-forum.com/t/e-encoding/127205/5 "2008-01-29T10:05:10Z")

</div>

e$B$^$D$b$He(B e$B$f$-$R$m$G$9e(B

In message “Re: [ruby-dev:33496] Re: /#{}/e.encoding”  
on Tue, 29 Jan 2008 17:31:32 +0900, SASADA Koichi [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)  
writes:

|\> e$B$G!“L?Na$rA}$d$9$N$OK\>$^$7$/$J$$$H$$$&$3$H$J$s$G$9$,!”$J$i$Pe(B  
|\> toregexpe$BL?Na$,e(Bcnte$B%_%Z%i%s%I$r\<u$1IU$1$k$h$&$K$7$F!"$=$NCf$Ge(B  
|\> preprocesse$B4X?t$r8F$V$h$&$K$7$F$O$I$&$G$7$g$&!)e(B  
|  
|e$B!!8D?ME_$K$O!“L?Na$O2DG=$J$i%a%=%C%I$KDI$$=P$7$?$$$J$!!”$H;W$C$F$$$^e(B  
|e$B$9!#e(B

e$B$U!\<$`!#e(BVMe$B\<BAu\<T$H$7$F$=$NMWK\>$OM}2r$G$-$^$9!#e(B

|e$B$=$&$$$&FCJL$J5!G=$O!“e(Bfreeze e$B$7$?%_%V%8%'%/%H$NFC0[%a%=%C%I$K$7e(B  
|e$B$F$_$/$H$+!#Nc$($P!”%/%i%9Dj5A!"%a%=%C%IDj5A!"e(Balias, undef e$B$J$s$+$Oe(B  
|e$BA4It30$KDI$$=P$7$?$$$J$!$H!#e(B

e$B%/%i%9Dj5A!“%a%=%C%IDj5A!“e(Baliase$B!“e(Bundefe$B$O$=$l$\>$lAjEv$9$k%a%=%Ce(B  
e$B%I$b$”$j$^$9$h$M!#$@$+$i!”$=$l$[$ILdBj$b$J$$$N$GFC$K\<h$j\>e$2e(B  
e$B$F$3$J$+$C$?$N$G$9$,!”:#2s$N$b$N$\_$?$$$K8=\>u$G$OBP1~$9$k%a%=%Ce(B  
e$B%I$b$J$$$N$K%a%=%C%I$KDI$$=P$7$?$$$H$J$k$H!"$^$?$A$g$C$HJL$Ne(B  
e$BOC$G$9$M!#e(B

e$BMW$9$k$K!"$=$N%a%=%C%I$r:FDj5A$7$?$j!"JQ$J%3%s%F%-%9%H$G8F$Se(B  
e$B=P$7$?$j$9$k$3$H$,7y$J$o$1$G$9!#$J$s$+%^%/%m$K;w$?!V7y$5!W!#e(B  
e$B:FDj5A$O$$$A$\*$&e(Bfreezee$B$G6X;\_$G$-$^$9$1$I!"8e\<T$O$M$(!#e(B

e$B$^$“!”;EAH$_$H$7$F$O%a%=%C%I8F$S=P$7$H$7$F$$$$F!"$=$N%%V%8%'e(B  
e$B%/%H!W$re(BRubye$B%l%Y%k$+$i4JC1$K$_$($J$/$9$k$H$+$GBP1~$G$-$J$$$3e(B  
e$B$H$O$J$$$H;W$$$^$9$1$I!#$^$?!"AjCL$7$^$7$g$&!#e(B

```
                            e$B$^$D$b$He(B e$B$f$-$R$me(B /:|)

```