Re: Just curious: who is using Pike for what?

On Thu, Dec 15, 2016 at 10:29 PM, Tobias S. Josefowitz
<[email protected]> wrote:
>>>> sz = "ß"
>>>> sz.upper() == sz
> True
>
> (python 2.7 apparently doesn't do this, not for me).

In Python 2.7, what you have is a byte string, so you're probably
trying to uppercase the UTF-8 byte sequence that represents that
character.

> Also while "ß" is generally mapped to "SS" when uppercasing in real
> life (on signage etc), I personally would truly be shocked if a
> programming language did that. Most of the time, in a programming
> language, I would suppose people process strings, not text, and "ß" ->
> "SS" pretty much only ever makes sense in text processing contexts.
> I realize this issue might extend beyond the odd little "ß" character,
> but as a native German speaker I can assure you I so far never had any
> use for a function that would uppercase "ß" to "SS", while I depended
> on upper_case() *not* doing that a lot.

What do you mean by "strings, not text"? What are strings of if not
text? I know the Pike 'string' type can be used to process arbitrary
data - but then you wouldn't be uppercasing or lowercasing that data.

> If there should be a need for such a functionality, it certainly
> should be provided by an extra function, not handled by upper_case().

I can understand wanting an "ASCII-only case conversion" function as
distinct from a "Unicode case conversion", but that's not what we
have. What is it that you expect upper_case to do?

ChrisA