Re: [Tuxpaint-devel] Font scoring method

[Shin-ichi T. <dol...@wm...>]

Hi!

On Sun, 18 Dec 2022 02:35:18 -0500, Albert Cahalan wrote:

>Would this chunk of currently unused code help you?
>
>          // translation spares -- design not finalized
>#if 0
>          user_font_styles[num_font_styles]->score +=
>            charset_works(font, gettext("<1>spare-1a"));
>          user_font_styles[num_font_styles]->score +=
>            charset_works(font, gettext("<1>spare-1b"));
>          user_font_styles[num_font_styles]->score +=
>            charset_works(font, gettext("<9>spare-9a")) * 9;
>          user_font_styles[num_font_styles]->score +=
>            charset_works(font, gettext("<9>spare-9b")) * 9;
>#endif
>
>It should not impact English, but that could use documentation!
>The text has repeated digits, either "1" or "9", causing the font
>to be considered bad, and thus the score does not increase.
>
>So you could translate like this:
>
>"<9>spare-9a" becomes a pair of hiragana.
>"<9>spare-9b" becomes a pair of katakana.
>"<1>spare-1a" becomes a pair of kanji.
>"<1>spare-1b" becomes the Yen and Yuan currency symbols.

Oh, I think it will be the one I wanted to have!

Of course I've noticed this unused part, but did not recognize the
difference to the one I proposed.

Yes, the key is that original string has two same characters, is'nt
it! How clever!

Do you mind if I enable this block now ?