Consolidation of Post-V2 Character/Glyph Addition Suggestions

[kenlunde]

The following table shows the 45 glyphs that were added in the Version 2.001 update, and unless otherwise noted, the additions are from my own notes:

Character	Glyph Name	Notes
U+3A52 㩒	uni3A52-HK
U+3BF2 㯲	uni3BF2-HK
U+465B 䙛	uni465B-HK	See Issue #215
U+50F8 僸	uni50F8-HK
U+53B4 厴	uni53B4-HK	See Issue #215
U+5662 噢	uni5662-HK	See Issue #212
U+58BA 墺	uni58BA-HK	Use existing TW glyph, uni58BA-TW, as-is (see Issues #204 & #212)
U+5967 奧	uni5967-HK	See Issue #212
U+5AFA 嫺	uni5AFA-TW	ETen addition; remove TW mapping override to uni5AFA-HK
U+5DB4 嶴	uni5DB4-HK	Use existing TW glyph, uni5DB4-TW, as-is (see Issues #204 & #212)
U+5DDF 巟	uni5DDF-HK	See Issue #204
U+5EE8 廨	uni5EE8-HK	See Issue #215
U+614C 慌	uni614C-HK	See Issue #204
U+6428 搨	uni6428-HK	See Issue #215
U+64D9 擙	uni64D9-HK	Use existing TW glyph, uni64D9-TW, as-is (see Issues #204 & #212)
U+64DB 擛	uni64DB-HK	See Issue #215; remove HK mapping override to uni64DB-JP
U+6983 榃	uni6983-HK
U+69A7 榧	uni69A7-HK	See Issue #215
U+69BB 榻	uni69BB-HK	See Issue #215
U+6A07 樇	uni6A07-HK	See Issue #215; remove HK mapping override to uni6A07-CN
U+6BCC 毌	uni6BCC-HK	See Issue #215; remove HK mapping override to uni6BCC-CN
U+6BFE 毾	uni6BFE-HK	See Issue #215
U+6F16 漖	uni6F16-HK	See Issue #215
U+6FB3 澳	uni6FB3-HK	See Issue #212
U+71CA 燊	uni71CA-JP	Restore from Version 1.000 data
U+71E0 燠	uni71E0-HK	See Issue #212
U+72C5 狅	uni72C5-HK	See Issue #215
U+7981 禁	uni7981-HK
U+7B9A 箚	uni7B9A-TW	CNS 11643 Plane 2 addition (was U+5284 劄)
U+7CA7 粧	uni7CA7-TW	ETen addition
U+7DFB 緻	uni7DFB-HK	See Issue #215; remove HK mapping override to uni7DFB-CN
U+819A 膚	uni819A-HK	See Issue #204; remove HK mapping override to uni819A-JP
U+8581 薁	uni8581-HK	See Issue #212
U+8669 虩	uni8669-HK
U+8943 襃	uni8943-HK	See Issue #215
U+8956 襖	uni8956-HK	See Issue #212
U+905C 遜	uni905C-HK	See Issue #204
U+95B9 閹	uni95B9-HK	See Issue #215; remove HK mapping override to uni95B9-JP
U+96A9 隩	uni96A9-HK	See Issue #212
U+9834 頴	uni9834-HK	See Issue #215
U+9B0D 鬍	uni9B0D-HK	See Issue #204; remove HK mapping override to uni9B0D-CN
U+9B2F 鬯	uni9B2F-HK	See Issue #215; remove HK mapping override to uni9B2F-CN
U+9EFB 黻	uni9EFB-HK	See Issue #204
U+1F16C	u1F16C	Unicode 12.0 RAISED MR SIGN
U+2C081 𬂁	u2C081-CN	See Issue #228

[tamcy]

As the final stroke of the 木 component on the top-right is different, glyphs of the following characters cannot be shared with other regions:

Currently shared with CN:
U+3A52 㩒
U+3BF2 㯲
U+7981 禁
U+6983 榃

Currently shared with JP:
U+50F8 僸

[KrasnayaPloshchad]

Is it possible to add italic glyphs and making them available via itlc feature?

[kenlunde]

@KrasnayaPloshchad Negative. The number of required glyphs is too high, and the glyph set is already full. BTW, the feature tag to which you're referring is 'ital'.

[stone-zeng]

Similar issues have been raised for several times (#191 and adobe-fonts/source-han-serif#75).

Now, the italic shape of Source Serif Pro has been made, while the HK support of Source Han Sans has been added as well. So do you think it's time to support italics (in an independent font file)? (perhaps I should open a new issue)

[kenlunde]

@stone-zeng Why do you think it would be a good idea to open a new issue about adding italics? Adding HK support to Source Han Serif will result in an overflowing glyph set that will require a special touch to fit within the 65,535-glyph limit, meaning that adding glyphs for italics is an effective non-starter.

[stone-zeng]

I don't know whether ital feature is necessary. Why not separate roman and italics into different otf/otc files, so that the 65,535-limit is no longer a limit? Anyway, I think many people will be happy if a properly-adjusted italic shape could be directly used altogether with Source Han Sans/Serif.

[frankrolf]

I agree with @stone-zeng in thinking that a separate Italic OTC would be a nice idea, however I don’t know how common that division is/how frequently the ital feature is used. It may well be that ital is expected.

One remark:
When you say “the Italic shape of Source Serif Pro has been made”, you need to consider that Source Serif and Source Han Serif are quite different in terms of their setup – Source Han Serif uses four masters for Latin, while Source Serif only uses three.
Also, Source Han Serif includes a big number of additional Western glyphs that don’t ship with Source Serif – which presumably also would need Italics.
In short: it’s not that simple, and creating Source Han Serif Italic would mean a considerable amount of work.

[Buernia]

CN, JP and KR glyph of 燊 (U+71CA) should be reverted to pre-V2 glyph.

Below is the glyph in 通用规范汉字表：

[tamcy]

The top-left component of HK's 虩 (U+8669) is different from TW, so the glyph cannot be shared.

[tamcy]

(This entry is about U+961D. U+2ED6 and U+2ECF are included in the screenshot for reference.)

Currently there's only one glyph for U+961D, which is uni961D-CN. It points to CID 43066, the same as U+2ED6.

However, U+961D is also a HKSCS character (P.252 of the reference glyph document), and its shape is centered to the em-box:

The decision to center the character in the em box is the same for other regions except for CN (Unicode code chart P.463):

So a new glyph would be needed for HK, which can be further shared to all other non CN regions.

[Buernia]

CN, JP and KR glyph of 胣 (U+80E3) should be reverted to pre-V2 glyph.

[byyxx128]

How about "𫟷" as the simplified Chinese name of the 116th element?

I did not find the latest progress about it, since it was still unavailable.

[punchcutter]

@byyxx128 What's the question? 𫟷 is in Source Han Sans.

[byyxx128]

What's the question? 𫟷 is in Source Han Sans.

@punchcutter I tried different Font Types, 𫟷 is only missing in SubsetOTF.
https://github.com/adobe-fonts/source-han-sans/tree/release/SubsetOTF/CN

In addition, some characters'(e.g. 𬭛𫟼) style also looks different from other characters' (e.g. 𬬻𬭊𬭳𬭶鿏𬬭鿔鿭𫓧镆𫟷).


Weight of font: Light 👆

Not every weight has the problem.

Weight of font: Heavy 👆

You may need to test all weights of the fonts.

[punchcutter]

@byyxx128 I see. We can add u2B7F7-CN to the CN Subset and check the balance on those other glyphs.