# \[Bug #981\] String#encode() keeps invalid bytes when converting to US-ASCII

**URL:** <https://www.ruby-forum.com/t/bug-981-string-encode-keeps-invalid-bytes-when-converting-to-us-ascii/156896>\
**Category:** Development\
**Created:** [January 5, 2009, 2:05am UTC](https://www.ruby-forum.com/t/bug-981-string-encode-keeps-invalid-bytes-when-converting-to-us-ascii/156896 "2009-01-05T02:05:12Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Martin\_DSSS\_rst](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/e9c0ed/32.png) [@Martin\_DSSS\_rst](https://www.ruby-forum.com/u/Martin_DSSS_rst)\
**Post date:** [January 5, 2009, 2:05am UTC](https://www.ruby-forum.com/t/bug-981-string-encode-keeps-invalid-bytes-when-converting-to-us-ascii/156896/1 "2009-01-05T02:05:12Z")

</div>

Bug #981: String#encode() keeps invalid bytes when converting to  
US-ASCII  
[http://redmine.ruby-lang.org/issues/show/981](http://redmine.ruby-lang.org/issues/show/981)

e$B5/I\<\<Te(B: Martin De$(D+de(Brst  
e$B%9%F!\<%?%9e(B: Open, e$BM%@hEYe(B: Normal  
e$BC4Ev\<Te(B: Akira T.

String#encode() (e$BJQ49$N;XDj$J$7!"e(Bdefault\_internal e$B$KJQ49e(B)  
e$B$Oe(B  
default\_internal e$B$,e(B US-ASCII e$B$N\>l9g$@$1e(B invalid  
e$B$J%P%$%H$de(B  
unknown e$B$JJ8;z$r$=$N$^$^%3%T!\<$7$^$9!#B\>$N%(%s%3!\<%G%#%s%0e(B  
e$B$G$Oe(B ‘?’ e$B$J$I$K%l%W%l!\<%9$5$l$^$9!#e(B

e$B\<B9TNce(B:

> ruby -E:ascii-8bit -e’puts "\uabcd’.encode.inspect’  
> “?”

> ruby -E:euc-jp -e’puts "\uabcd’.encode.inspect’  
> “?”

> ruby -E:us-ascii -e’puts "\uabcd’.encode.inspect’  
> “\xEA\xAF\x8D”

e$B0U?^E\*$J$b$N$G$“$k$H$O$I$&$7$F$b;W$$$^$;$s$,!”$b$7$=$&e(B  
e$B$G$7$?$ie(B rdoc e$B$G=q$$$?J}$,$$$$$+$H;W$$$^$9!#e(B

([ruby-dev:37662] e$B;2\>He(B)

e$B$h$m$7$/$\*4j$$$7$^$9!#e(B Martin.

---

<div class="post-metadata">

**Author:** ![Martin\_DSSS\_rst](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/e9c0ed/32.png) [@Martin\_DSSS\_rst](https://www.ruby-forum.com/u/Martin_DSSS_rst)\
**Post date:** [January 15, 2009, 6:16am UTC](https://www.ruby-forum.com/t/bug-981-string-encode-keeps-invalid-bytes-when-converting-to-us-ascii/156896/2 "2009-01-15T06:16:35Z")

</div>

e$B%A%1%C%He(B #981 e$B$,99?7$5$l$^$7$?!#e(B (by Yui NARUSE)

e$BC4Ev\<Te(B Akira T.e$B$+$ie(BYukihiro M.e$B$KJQ99e(B

e$B$3$N5sF0$Oe(B String#encode  
e$B$,D\>@$N860x$G$O$J$/!"e(BEncoding.default\_internal  
e$BB&$,860x$G$9!#e(B  
% ruby19 -E:us-ascii -e’p Encoding.default\_internal’  
#Encoding:UTF-8

enc\_set\_default\_encoding e$B$r8+$k$H!"e(B  
if (def-\>index == ENCINDEX\_US\_ASCII)  
def-\>index = defindex;  
e$B$H!"e(BUS-ASCII e$B$N\>l9g$K%G%U%)%k%H$N%(%s%3!\<%G%#%s%0e(B  
(default\_internal e$B$G$Oe(B UTF-8) e$B$,%;%C%H$5$l$F$$$^$9!#e(B  
e$B8=:\_$N;EMM$G$O$3$NJU$N=hM}$OITMW$J5$$,$9$k$N$G!"0J2\<$NDL$j$H$+!#e(B

# Index: encoding.c

— encoding.c (revision 21519)  
+++ encoding.c (working copy)  
@@ -1051,7 +1051,7 @@ struct default\_encoding {

static int  
enc\_set\_default\_encoding(struct default\_encoding \*def, VALUE encoding,

- 

```
                   const char *name, int defindex)

```

- 

```
                   const char *name)

```

{  
int overridden = Qfalse;  
if (def-\>index != -2)  
@@ -1066,8 +1066,6 @@ enc\_set\_default\_encoding(struct default\_encoding \*  
}  
else {  
def-\>index = rb\_enc\_to\_index(rb\_to\_encoding(encoding));

- 

```
  if (def->index == ENCINDEX_US_ASCII)

```

- 

```
      def->index = defindex;
  def->enc = 0;
  enc_alias_internal(name, def->index);

```

}  
@@ -1118,7 +1116,7 @@ rb\_enc\_set\_default\_external(VALUE encoding)  
rb\_raise(rb\_eArgError, “default external can not be nil”);  
}  
enc\_set\_default\_encoding(&default\_external, encoding,
- 

```
                       "external", ENCINDEX_US_ASCII);

```

- 

```
                       "external");

```

}

/\*  
@@ -1171,7 +1169,7 @@ void  
rb\_enc\_set\_default\_internal(VALUE encoding)  
{  
enc\_set\_default\_encoding(&default\_internal, encoding,

- 

```
                       "internal", ENCINDEX_UTF_8);

```

- 

```
                       "internal");

```

}

/\*

* * *

[http://redmine.ruby-lang.org/issues/show/981](http://redmine.ruby-lang.org/issues/show/981)

---

<div class="post-metadata">

**Author:** ![Martin\_DSSS\_rst](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/e9c0ed/32.png) [@Martin\_DSSS\_rst](https://www.ruby-forum.com/u/Martin_DSSS_rst)\
**Post date:** [January 15, 2009, 8:07am UTC](https://www.ruby-forum.com/t/bug-981-string-encode-keeps-invalid-bytes-when-converting-to-us-ascii/156896/3 "2009-01-15T08:07:47Z")

</div>

e$B%A%1%C%He(B #981 e$B$,99?7$5$l$^$7$?!#e(B (by Yukihiro M.)

e$B%9%F!\<%?%9e(B Opene$B$+$ie(BClosede$B$KJQ99e(B  
e$B?JD=e(B % 0e$B$+$ie(B100e$B$KJQ99e(B

## Applied in changeset r21528.

[http://redmine.ruby-lang.org/issues/show/981](http://redmine.ruby-lang.org/issues/show/981)