Ironruby core - Apr 2011 - Encoding::InvalidByteSequenceError: character U+7453 can't be encoded in IBM437

Hello,

I''m getting the following error while running one of the examples
(Enumerate
Top Level Windows <https://github.com/ffi/ffi/wiki/Windows-Examples>) from
the FFI wiki:

*Encoding::InvalidByteSequenceError: character U+7453 can''t be encoded
in
IBM437*
*        from (irb):26:in `puts''*

The example works fine in MRI 1.9.2, so I''m curious if this might be a
bug,
or if I''m screwing something up in my FFI impl. Official FFI uses
*str_new<http://www.ruby-doc.org/doxygen/1.8.4/string_8c-source.html#l00078>
*, which in turn performs a simple *memcopy* given a *char**, which is more
or less what I''m doing to construct my string...

At any rate, the example, which features callbacks, makes it this far (which
is exciting):

*[001] Found ''''*
*[002] Found ''''*
*[003] Found ''''*
*[004] Found ''''*
*[005] Found ''''*
*[006] Found ''''*
*[007] Found ''''*
*[008] Found ''''*
*Encoding::InvalidByteSequenceError: character U+7453 can''t be encoded
in
IBM437*
*        from (irb):26:in `puts''*


-Charles
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://rubyforge.org/pipermail/ironruby-core/attachments/20110408/a06b609d/attachment.html>

On second thought (and with a little more rest), I''m pretty positive
this my
problem :).

-Charles

On Fri, Apr 8, 2011 at 10:42 PM, Charles Strahan <
charles.c.strahan at gmail.com> wrote:
> Hello,
>
> I''m getting the following error while running one of the examples
(Enumerate
> Top Level Windows <https://github.com/ffi/ffi/wiki/Windows-Examples>)
from
> the FFI wiki:
>
> *Encoding::InvalidByteSequenceError: character U+7453 can''t be
encoded in
> IBM437*
> *        from (irb):26:in `puts''*
>
> The example works fine in MRI 1.9.2, so I''m curious if this might
be a bug,
> or if I''m screwing something up in my FFI impl. Official FFI uses
*str_new<http://www.ruby-doc.org/doxygen/1.8.4/string_8c-source.html#l00078>
> *, which in turn performs a simple *memcopy* given a *char**, which is
> more or less what I''m doing to construct my string...
>
> At any rate, the example, which features callbacks, makes it this far
> (which is exciting):
>
> *[001] Found ''''*
> *[002] Found ''''*
> *[003] Found ''''*
> *[004] Found ''''*
> *[005] Found ''''*
> *[006] Found ''''*
> *[007] Found ''''*
> *[008] Found ''''*
> *Encoding::InvalidByteSequenceError: character U+7453 can''t be
encoded in
> IBM437*
> *        from (irb):26:in `puts''*
>
>
> -Charles
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://rubyforge.org/pipermail/ironruby-core/attachments/20110409/24510d0a/attachment-0001.html>

Nope - must be a bug in IronRuby:

*MRI v1.9.2:*
*
irb(main):001:0> str = "%s" % "\u00c2"
=> "\u00C2"
irb(main):002:0> str.encoding
=> #<Encoding:UTF-8>
irb(main):003:0> puts str
??
*

*IronRuby v1.1.3:*
*
irb(main):001:0> str = "%s" % "\u00c2"
=> "\xC2"
irb(main):002:0> str.encoding
=> #<Encoding:IBM437>
irb(main):003:0> puts str
Encoding::InvalidByteSequenceError: character U+00C2 can''t be encoded
in
IBM437
        from (irb):3:in `puts''
        from (irb):3
        from C:/IronRuby/bin/irb:13
*



*String#%* seems to handle encoding differently between the two Rubies.  If
I can figure it out, I''ll send a pull request.


Cheers,
-Charles


On Sat, Apr 9, 2011 at 7:18 PM, Charles Strahan <charles.c.strahan at
gmail.com
> wrote:
> On second thought (and with a little more rest), I''m pretty
positive this
> my problem :).
>
> -Charles
>
>
> On Fri, Apr 8, 2011 at 10:42 PM, Charles Strahan <
> charles.c.strahan at gmail.com> wrote:
>
>> Hello,
>>
>> I''m getting the following error while running one of the
examples (Enumerate
>> Top Level Windows
<https://github.com/ffi/ffi/wiki/Windows-Examples>)
>> from the FFI wiki:
>>
>> *Encoding::InvalidByteSequenceError: character U+7453 can''t be
encoded in
>> IBM437*
>> *        from (irb):26:in `puts''*
>>
>> The example works fine in MRI 1.9.2, so I''m curious if this
might be a
>> bug, or if I''m screwing something up in my FFI impl. Official
FFI uses *
>>
str_new<http://www.ruby-doc.org/doxygen/1.8.4/string_8c-source.html#l00078>
>> *, which in turn performs a simple *memcopy* given a *char**, which is
>> more or less what I''m doing to construct my string...
>>
>> At any rate, the example, which features callbacks, makes it this far
>> (which is exciting):
>>
>> *[001] Found ''''*
>> *[002] Found ''''*
>> *[003] Found ''''*
>> *[004] Found ''''*
>> *[005] Found ''''*
>> *[006] Found ''''*
>> *[007] Found ''''*
>> *[008] Found ''''*
>> *Encoding::InvalidByteSequenceError: character U+7453 can''t be
encoded in
>> IBM437*
>> *        from (irb):26:in `puts''*
>>
>>
>> -Charles
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://rubyforge.org/pipermail/ironruby-core/attachments/20110410/7322b78e/attachment.html>