Potentially unifiable glyphs in Source Han Serif v2.001

[tamcy]

I noticed a number of glyphs in Source Han Serif 2.001 that are potentially mergeable and here is a summary of them. I hope it can help freeing up a few more CIDs for use.

The list contains visually identical glyphs or those with only minor differences such that merging them should not impose any redesigning effort nor further inconsistency. I tried to skip those that may cause a debate (like whether a regional standard should be followed or not; I think it's better to leave this to another issue).

Remarks are subjective, they are written from a Hong Kong user's perspective.

I am only listing the relevant codepoints in this comment. For a listing of codepoints with glyph shapes, please refer to this PDF file. The PDF is produced from an HTML document which I am also attaching the source files here in case you may find it useful. (The HTML file unifiable-codepoints.html is a template. All items are defined in items.js, and app.js is responsible for visualizing the data to the HTML.)

Codepoint list

A.1. Visually identical, glyph accessible by base glyph

ID	Codepoint	Character
A.1.1	U+406F	䁯
A.1.2	U+4D77	䵷
A.1.3	U+4E0F	丏
A.1.4	U+4E10	丐
A.1.5	U+4F72	佲
A.1.6	U+503B	倻
A.1.9	U+53E1	叡
A.1.8	U+537A	卺
A.1.10	U+5413	吓
A.1.11	U+55A9	喩
A.1.12	U+55E1	嗡
A.1.13	U+564F	噏
A.1.14	U+57E6	埦
A.1.15	U+5ADD	嫝
A.1.16	U+58D1	壑
A.1.17	U+5BF3	寳
A.1.18	U+5EB7	康
A.1.19	U+5F31	弱
A.1.20	U+5F54	彔
A.1.21	U+6177	慷
A.1.22	U+6240	所
A.1.23	U+63C7	揇
A.1.24	U+6960	楠
A.1.25	U+6A59	橙
A.1.26	U+6B9F	殟
A.1.27	U+6E71	湱
A.1.28	U+7023	瀣
A.1.29	U+706B	火
A.1.30	U+708E	炎
A.1.31	U+72C4	狄
A.1.32	U+73E1	珡
A.1.33	U+73E4	珤
A.1.34	U+76FE	盾
A.1.35	U+7CD1	糑
A.1.36	U+7C72	籲
A.1.37	U+7F71	罱
A.1.38	U+7FDD	翝
A.1.39	U+7FE6	翦
A.1.40	U+8020	耠
A.1.41	U+82E4	苤
A.1.42	U+877B	蝻
A.1.43	U+8789	螉
A.1.44	U+88D8	裘
A.1.45	U+8EB6	躶
A.1.46	U+96B6	隶
A.1.47	U+92F3	鋳
A.1.48	U+93CB	鏋
A.1.49	U+9ACA	髊
A.1.50	U+9C2F	鰯
A.1.51	U+9D02	鴂
A.1.52	U+21336	𡌶
A.1.53	U+26951	𦥑

A.2. Visually identical, glyph only accessible via variation sequence

ID	Codepoint	Character
A.2.1	U+4E30	丰
A.2.2	U+50CF	像
A.2.3	U+5300	匀
A.2.4	U+74CA	瓊
A.2.5	U+8987	覇
A.2.6	U+9873	顳

B.1. Minor differences (全)

The width of the top horizontal line should be a typeface design decision, not worth dedicating new glyphs for TW. See also U+9293 銓, U+8DE7 跧, U+8F07 輇, U+6D24 洤, U+5168 全 and U+4F7A 佺 - JP glyphs are used for TW. This is unlike the component 分 or 八 where a separate glyph set is provided for TW.

ID	Codepoint	Character
B.1.1	U+4F7A	佺
B.1.2	U+62F4	拴
B.1.3	U+6813	栓
B.1.4	U+7277	牷

B.2. Minor differences (癸)

The only difference is that the last stroke of 天 touches the second last stroke in uni9368-HK. Probably not worth a separate glyph. See also U+4906 䤆 (dedicated HK glyph), U+63C6 揆 (uses JP glyph), U+8067 聧 (uses TW glyph) - they are all designed with the strokes touched.

ID	Codepoint	Character
B.2.1	U+9368	鍨

B.3. Minor differences (求)

The only difference is whether the 4th stroke of 求 touches the vertical stroke with hook. After checking a few fonts from type foundries that produce Simplified Chinese typefaces, I believe this should be a pure design choice and a separate design for CN is not necessary. Also see U+6344 捄, U+6882 梂, U+6B8F 殏, U+76DA 盚 and U+86F7 蛷 - JP glyphs are used for CN.

ID	Codepoint	Character
B.3.1	U+4FC5	俅
B.3.2	U+6551	救
B.3.3	U+7403	球

[CoolMarvel43]

Also of note, 県 (U+770C) in CN (red) and HK (blue) can be unified. It falls under your category of A.1. Visually identical, glyph accessible by base glyph.