[TextKit]Why some special characters could make Chinese characters thinner and smaller

Question

That happens on TextKit. But everything is all right when I use CoreText.

The rich text in the debug window of Xcode also has the same issue.

So, is it a bug of TextKit? Is there any solution in addition to replace TextKit by CoreText?

Some special unicode characters will cause the problem, such as ♬, ☜ — Swain, Aug 28 '17 at 04:06
I don't see a single special character in the picture you post. — El Tomato, Aug 28 '17 at 04:07
@ElTomato ☜ is the special character in the case. The characters "啦啦啦啦" before ☜ are rendered correctly, but the characters after ☜ are rendered abnormally. — Swain, Aug 28 '17 at 08:06
☜ is not a special character. It's a UNICODE symbol. Most likely, the font family you have selected does cover it. Anyway, your topic has little to do with programming. — El Tomato, Aug 28 '17 at 08:32
Why the Chinese characters after ☜ are rendered thinner, but the same Chinese characters before ☜ are rendered correctly? @ElTomato — Swain, Aug 29 '17 at 07:30

score 0 · Accepted Answer · answered Sep 17 '17 at 10:05

I tried it and the difference really exists between CoreText and TextKit. I guess it caused by the font fallback functions.

For example, we have a string "啦☜啦", and set a font. In case the given font don't have all characters, there usually a list of fallback fonts like:

English font
CJK font
Emoji font
Unicode font (support almost all the unicode character)

There are two ways to find a fallback font.

So if TextKit use the second way, two '啦's may use different fonts because of the character '☜'.

To solve this, you can use attributed string and give '☜' a different font size to break the fallback loop, then the second '啦' use CJK font.

1 Answers1