R help - Dec 2012 - Writing escaped unicode

I'd like to write unicode strings using the "\u" escape syntax. 
According to the documentation, print.default or encodeString will escape
unicode using the \u convention.  In practice, I can't make it work.
> b="Unicode character: \ufffd"
> print.default(b)
[1] "Unicode character: ?"
> encodeString(b)
[1] "Unicode character: ?"

I want to write the string back out in the same escape formatting as I read it
in.  This is because I'm interfacing with some Ruby code that requires
unicode to be in this escaped format.

Thanks in advance!

On Mon, Dec 10, 2012 at 11:46:40PM -0500, David Kulp
wrote:
> I'd like to write unicode strings using the "\u" escape
syntax.  According to the documentation, print.default or encodeString will
escape unicode using the \u convention.  In practice, I can't make it work.
> 
> > b="Unicode character: \ufffd"
> > print.default(b)
> [1] "Unicode character: ???"
> > encodeString(b)
> [1] "Unicode character: ???"
> 
> I want to write the string back out in the same escape formatting as I read
it in.  This is because I'm interfacing with some Ruby code that requires
unicode to be in this escaped format.
as I read the documentation, encodeString escapes control characters,
but not "unicode characters". The notion of a "unicode
character" is
not entirely well defined, considering that the very mission of the
unicode consortium is to make sure that there are no non-unicode
characters...  ;-)
>From this it follows that replacing all characters with their \uxxxx
representation, e.g. by

    paste(sprintf("\\u%04x", utf8ToInt(b)), collapse = "");

should work with the Ruby client you try to talk to. Obviously, this
bloats the string rather more than necessary (particularly if most of
the characters are in the ASCII range), but if the volume you're
piping into the client is small, this may be good enough.

Best regards, Jan
-- 
 +- Jan T. Kim -------------------------------------------------------+
 |             email: jttkim at gmail.com                                |
 |             WWW:   http://www.jtkim.dreamhosters.com/              |
 *-----=<  hierarchical systems are for files, not for humans  >=-----*

Thanks, Jan.  Unfortunately I have huge streams of data to transmit and it
should be mostly human readable, too, so escape encoding the entire string
isn't an option.
A workaround is that I found I can use the rjson package as in
> cat(toJSON(fromJSON("\"Unicode char: \ufffd\"")))
"Unicode char: \ufffd"

but this seems like an awful hack and suggests that there really is a better way
to escape extended characters.

> From: Jan T Kim <jttkim@googlemail.com>
> Subject: Re: [R] Writing escaped unicode
> Date: December 11, 2012 5:49:18 AM EST
> To: r-help@r-project.org
> 
> 
> On Mon, Dec 10, 2012 at 11:46:40PM -0500, David Kulp wrote:
>> I'd like to write unicode strings using the "\u" escape
syntax.  According to the documentation, print.default or encodeString will
escape unicode using the \u convention.  In practice, I can't make it work.
>> 
>>> b="Unicode character: \ufffd"
>>> print.default(b)
>> [1] "Unicode character: ???"
>>> encodeString(b)
>> [1] "Unicode character: ???"
>> 
>> I want to write the string back out in the same escape formatting as I
read it in.  This is because I'm interfacing with some Ruby code that
requires unicode to be in this escaped format.
> 
> as I read the documentation, encodeString escapes control characters,
> but not "unicode characters". The notion of a "unicode
character" is
> not entirely well defined, considering that the very mission of the
> unicode consortium is to make sure that there are no non-unicode
> characters...  ;-)
> 
>> From this it follows that replacing all characters with their \uxxxx
> representation, e.g. by
> 
>    paste(sprintf("\\u%04x", utf8ToInt(b)), collapse =
"");
> 
> should work with the Ruby client you try to talk to. Obviously, this
> bloats the string rather more than necessary (particularly if most of
> the characters are in the ASCII range), but if the volume you're
> piping into the client is small, this may be good enough.
> 
> Best regards, Jan
> -- 
> +- Jan T. Kim -------------------------------------------------------+
> |             email: jttkim@gmail.com                                |
> |             WWW:   http://www.jtkim.dreamhosters.com/              |
> *-----=<  hierarchical systems are for files, not for humans 
>=-----*
> 

	[[alternative HTML version deleted]]