adobe-fonts / adobe-fonts/source-han-sans

Standardize the height of the 曰 components between 曰 (U+66F0), 抇 (U+6287) and 汩 (U+6C69)

未关闭
#466 2 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看
主要语言
Python
星标
17.2k
派生
1.4k
PR 合并指标
30 天内没有已合并 PR

描述

I am opening a separate issue after feedback from @tamcy at #443, because the original issue of remapping deviated into the design issue of the 曰 (U+66F0) component vs 日 (U+65E5), where traditional conventions say that the former must be shorter and wider. In light of this, I will quote the reply (and the suggestion) here:

> The height of the 曰 component in uni6287-HK looks fine to me (both serif and sans). For serif, the design of 抇 also rhymes well with 汩 (U+6C69. not 汨 U+6C68).
>
> ![image](https://github.com/adobe-fonts/source-han-sans/assets/959433/1976caf6-0788-44fb-8aad-11968d736124)
>
> It is actually the 曰 component in Sans version's uni6C69-CN that looks a bit too tall.
>
> ![image](https://github.com/adobe-fonts/source-han-sans/assets/959433/aa6ec66e-d1b6-4c4c-b2ae-a55caaf9ed03)
>
> Also, from the code chart, unlike U+6C68 and U+6C69, it looks like different components are chosen by different sources for U+6287.
>
> ![comparewithlegend](https://github.com/adobe-fonts/source-han-sans/assets/959433/7fa87c04-ca77-4e1a-95a7-2ba3465938ac)
> ```
> VN: 扌日 for U+6287; 扌曰 for U+22A8F.
> KR: 扌日 in U+6287 (i.e. the same as VN); does not include U+22A8F
> TW: 扌曰 for U+6287;扌日 for U+22A8F (i.e. the opposite of VN).
> HK: 扌曰 for U+6287 (i.e. the same as TW but with a different design for 曰); does not include U+22A8F.
> CN: Not sure.
> ```
>
> Thus, if there are really something that needs to be done, I'd suggest the followings be considered for Sans:
> 1. The current uni6287-HK is modified from uni6287-JP. Technically 曰 should be a wider component compared to 日, so make 扌 narrower in width and make 曰 wider.
> 2. Create uni6287-TW from the adjusted uni6287-HK, make the middle horizontal line in 曰 connected, and replace uni6287-JP.
> 3. Modify uni6C69-CN and make the 曰 component shorter.
>
> Sounds a bit complicated here, but the idea is to make Sans follow the setting of Serif.
>
> ![sun](https://github.com/adobe-fonts/source-han-sans/assets/959433/83f61542-547d-489d-8b3e-e7f7e2c350fc)
>
> _Originally posted by @tamcy in https://github.com/adobe-fonts/source-han-sans/issues/443#issuecomment-1756768464_

But I have another idea here. I suggest that we adjust the two glyphs, 抇 (U+6287) and 汩 (U+6C69) to match the height of the 曰 (U+66F0) glyph, which is JP in Sans and CN in Serif ([which I personally think the latter should be replaced with a JP glyph based off Kozuka Mincho](https://github.com/adobe-fonts/source-han-sans/issues/425), explained below).

Screenshot 2023-10-11 at 18 02 44

_From left to right: uni66F0-JP, uni66F0-TW, uni6287-JP, uni6287-HK, uni6C69-TW, uni6C69-CN_

Which means 抇 (U+6287, especially uni6287-HK) should have the 曰 component slightly taller (especially towards the bottom), and also widen the 曰 component as per the quote above, and finally, the corresponding component for 汩 (U+6C69) should be modified to be shorter and wider (applies to both the CN and TW glyphs).

Screenshot 2023-10-11 at 18 08 24

_From left to right: uni66F0-CN, uni66F0-TW, uni6287-TW, uni6287-HK, uni6C69-TW, uni6C69-CN_

And for Serif, I think we might need to either raise the 曰 component in 抇 (U+6287) and 汩 (U+6C69) slightly higher to match the 曰 (U+66F0) glyph, or shorten the 曰 (U+66F0) CN and TW glyphs, or try to balance the height of the 曰 component in the three glyphs somewhere in between.

Which is why for 曰 (U+66F0), as mentioned above, I suggest to replace the CN glyph (marked in red) with an unreleased JP glyph that could possibly be based off Kozuka Mincho (marked in blue), for which the glyph is slightly wider and shorter.

And on another note, the TW glyph for 曰 (U+66F0) can probably be made wider to match either the CN glyph or the Kozuka Mincho glyph.

Screenshot 2023-10-11 at 17 53 40

Here is the overlay comparison between the two glyphs, same colour.
曰 SHSerif vs KozMin

And finally, here's the possible height comparison if I replace uni66F0-CN with the corresponding Kozuka Mincho glyph that could be the uni66F0-JP glyph (same order otherwise).

Screenshot 2023-10-11 at 18 16 43

贡献指南

这个仓库没有索引到贡献指南

评估

这个 Issue 还没有评估数据。

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。