[Cuis-dev] UnicodeString coercion

[Phil B]

Thanks Juan, I'll take a closer look into the cause of my issues.

On Thu, Jul 2, 2026 at 10:49 AM Juan Vuletich <juan at cuis.st> wrote:

> And what you actually want is most likely #asUtf8BytesOrByteString,
> #asUtf8BytesOrByteStringOrByteArray to properly handle Unicode. That of
> course depends on whether the apis you're calling can handle UTF-8.
>
> Besides, 'abc' is an instance of String. You get a UnicodeString when you
> go beyond ASCII. Most apis prefer UTF-8 in that case.
>
> Cheers,
> On 2026-07-01 1:59 PM, Juan Vuletich via Cuis-dev wrote:
>
> Well, that boat sailed a long time ago.
>
> The interoperability between (Byte)String and UnicodeString does exactly
> the same as interoperability between
> SmallInteger/LargeInteger/Fraction/Float. With numbers, this only matters
> when dealing with the external world, where you need to ensure you're using
> the required primitive type, and not what Smalltalk choses for you. Same
> here. #asByteString to the rescue.
>
> See
> https://github.com/Cuis-Smalltalk/Cuis-Smalltalk-Dev/blob/master/Documentation/Papers/2022-11-UnicodeSupportInCuisSmalltalk.pdf
>
> Hope this helps,
> On 2026-07-01 11:59 AM, Phil B via Cuis-dev wrote:
>
> Sorry... forgot to reply all.
>
> To put it more succinctly: rather than assume that anywhere you have a
> string/stream it's just as good to substitute a Unicode string/stream, it
> really seems like it should be opt-in via #asUnicodeString or
> #asUniFileStream for coercion or explicit at instance creation time.  That
> makes the contract clear and you don't get something other than what you
> want/need.
>
> On Wed, Jul 1, 2026 at 11:52 AM Phil B <pbpublist at gmail.com> wrote:
>
>> I hear what you're saying but that assumption breaks a lot of code.  They
>> are *similar* to String, but they are actually very different creatures and
>> only superficially protocol compatible. (i.e. you often cannot
>> substitute a UnicodeString for a String without severely breaking things...
>> even when 'String' is nowhere in the source code that breaks)  I have had
>> to ugly up a lot of code to get out of the implicit opt-in to Unicode that
>> Cuis has done on this front.[1] Please don't rename String... that would
>> make things 100x worse.
>>
>> [1] i.e. 'where the hell did that UnicodeString come from?' when I
>> didn't want it... usually when dealing with a DNU error dialog.  Even when
>> I resolve the DNU, the breakage often continues on down the chain because
>> you can't just substitute UnicodeString for String when you're dealing with
>> FFI, web APIs, file formats etc.
>>
>> On Wed, Jul 1, 2026 at 11:16 AM Juan Vuletich <juan at cuis.st> wrote:
>>
>>> Hi Phil,
>>>
>>> On 2026-07-01 10:27 AM, Phil B via Cuis-dev wrote:
>>> > Is there a reason UnicodeString asString returns self rather than a
>>> > String?
>>> >
>>> Yes. UnicodeStrings are conceptually Strings. Perhaps we'd rename String
>>> as ByteString, and CharacterSequence as String.
>>>
>>> In any case, just use #asByteString that does what you want.
>>>
>>> Cheers,
>>>
>>> --
>>> Juan Vuletich
>>> www.cuis.st
>>> github.com/jvuletich
>>> researchgate.net/profile/Juan-Vuletich
>>> independent.academia.edu/JuanVuletich
>>> patents.justia.com/inventor/juan-manuel-vuletich
>>>
>>>
> --
> Juan Vuletichwww.cuis.stgithub.com/jvuletichresearchgate.net/profile/Juan-Vuletichindependent.academia.edu/JuanVuletichpatents.justia.com/inventor/juan-manuel-vuletich
>
>
> --
> Juan Vuletichwww.cuis.stgithub.com/jvuletichresearchgate.net/profile/Juan-Vuletichindependent.academia.edu/JuanVuletichpatents.justia.com/inventor/juan-manuel-vuletich
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cuis.st/mailman/archives/cuis-dev/attachments/20260702/8bb8943f/attachment.htm>