typeset model,

[muspene666]

Hi, I've been thinking about a way to change the typeset model,

so that it arranges the bubbles optimally according to the language,

For languages using the Roman alphabet, and even all languages written horizontally,
the typeset is diamond-shaped,

    text 

text text text
text

and on this basis

We could measure the font when it's first used, then record the length in pixels of the font in a csv, json or txt file, bearing the name of the font. This measurement would enable the software to automatically know when to make a line break, and to do so automatically in the trad, displayed on the right,
It would also enable calculations to be made to optimize the form, wouldn't it?
It would also make it easier to calculate the distance to the bubble so that the text isn't pasted or to reduce the font size

I'm still thinking about how to optimize the vertical system,
because with chinese and korean bubbles, the display would be more boxy than diamond-shaped,