Hyperion Entertainment Message Boards

WAS: Dynamic Languages selection for preferred language in Locale

I'd like to see the list remaining unchanged with codeset selection seperate/after Language selection,

If there is any possibility of merging multiple codeset routines into a single language library...would this be plausible?

Allow conversion based on a systemwide UTF-8 encoding and allow language libraries to have a flag for "additional encoding" options?

or would this be overloading the locale subsystem with extra complexity?

the current "select the codeset with the first preferred language" seems a bit "off"... why not an extra step of LanguageOptions AFTER the selection of "Language".

(Within my own Perception-IME project I am looking at usage of UTF-8 everywhere... with additional encoding options present as well)

Belxjander wrote:I'd like to see the list remaining unchanged with codeset selection seperate/after Language selection,

That's not possible since there is only a single codeset used for everything (keyboard input, the 8 bit fonts used in IGraphics->Text() output, default codeset for loading and converting catalogs, etc.), it's ILocale->OpenLocale(NULL)->loc_CodeSet, the codeset of the first language.

Everything, incl. the codesets of all languages, will be changed to codeset 106 (UTF-8) sometime, but before that can be done there is still a lot to do in several AmigaOS parts.

So there will remain a "preferred languages" listing but the "default codeset" is being migrated to UTF-8?

That is good news for me, which means any Languages can provide one or more modules with conversion to or from UTF-8 into other encodings as a sane standard basis for using locale.library If my understanding is good about that subsystem.

I'll focus on provision of UTF-8 encoded materials within the "chinese.language", "japanese.language" and "korean.language" libraries I have started, with a focus towards completion of Japanese using perception.library as an Input to UTF-8 wedge.

I've defaulted to pass-through for existing Applications and just need to decide on what modifier key for an initial "non-breaking space" selection along with conversion of existing character encodings into UTF-8 codepoints.

After that I am just focusing on a LanguageTranslation() and LanguageSort() function pair being plugged into the InputMethod for preferred language sections of UTF-8 codepoints being sorted before anything else.

@Belxjander
Please do try to remain focused. Stick to our plan and produce a functional IME. Don't worry about the rest of the system. We will get to it.