Re: [AD] Proposal to kill non-UTF-8 support

["Eric Botcazou" <ebotcazou@xxxxxxxxxx>]

> Of course, that does make the conversions the problem of the client
> rather than the library. Is that is a good thing or not?

In my mind, that's clearly a Bad Thing:
- the burden is on the client,
- instead of having a consistent encoding format throughout the code (with
little ASCII strings here and there), you end up with two parts that can't
communicate with each other without a conversion layer. (This can be very
funny under Windows: allegro_message() uses MessageBoxW(), so you could end
up with Unicode --> UTF-8 --> Unicode).

But this might be a near-term viewpoint, especially in light of your X
example.

> As for whether a lib should impose a choice, I think it should try
> not to limit what people can do with it, but also, shouldn't go too
> far trying to directly support every one of a million options if
> it would be possible to support just one.

Yes, but what if the lib supports a wide array of options in a previous
version ? Ok, I've again forgotten that we're designing another library ;-)

--
Eric Botcazou
ebotcazou@xxxxxxxxxx