adobe-fonts / source-han-serif

Source Han Serif | 思源宋体 | 思源宋體 | 思源宋體 香港 | 源ノ明朝 | 본명조
https://adobe.ly/SourceHanSerif
Other
8.02k stars 643 forks source link

The 尤 and 冘 components may need unification #148

Open Marcus98T opened 2 years ago

Marcus98T commented 2 years ago

I suggest that we should unify the 尤 and 冘 components (with my preferred JP form where 丿 and 乚 touches) for all regions whenever possible, saving a bit of glyph space. There's nothing that says the 丿 and 乚 should not touch for CN/TW/HK locales. This is for Serif, the Sans version is here to avoid confusion.

Screenshot 2022-01-14 at 04 05 41

Here are the characters to review: 冘抌帎沈忱妉枕眈衴耽紞酖躭訦鈂馾黆魫黕邥瓭鸩鴆髧沊醓 莐霃萙 尢尤优扰犹沋忧肬蚘紌訧鱿魷就駀疣嵆鑙諬嵇稽 莸就僦噈憱殧蹴鹫鷲蹵

Screenshot 2022-01-14 at 02 52 02

The v1 reference is from a Source Han Serif fork called GenYoMincho (源樣明體 TW version) as I cannot install v1 and v2 at the same time. Glyphs are largely the same as v1 Serif, especially those which are removed v1 JP glyphs.

Screenshot 2022-01-14 at 02 52 47 Screenshot 2022-01-14 at 02 53 03

LEGEND Green - the v1 glyph can be restored to replace all locales. Blue - the v1 glyph can be restored to replace only the JP/KR/CN locales. Purple - the v1 glyph cannot be restored whatsoever (has traditional print forms unsuitable for CN/TW/HK locales). Red - the glyphs to be replaced by the v1 JP glyphs. Orange - the glyphs to be adjusted. Yellow - remap the TW form to the HK form. Magenta/Pink - Replace with v2 JP/KR forms and remove existing glyph. Cyan - Replace with v2 JP/KR forms. Light green - replace with HK form and adjust glyph, then remove the CN glyph. White - no change needed.

As for 黕 (marked dark grey), special case here: Rename the CN glyph to the JP glyph and adjust the 冘 portion such that it matches the JP aesthetics, then the old JP glyph can be renamed to TW/HK or just simply delete it in favour of the v1 TW glyph. My quote from earlier:

Sorry if I commented quite late, but I kinda felt the CN glyph for 黕 should have been the JP glyph, but adjust that 冘 component such that the last stroke 乚 is closer to 丿, like the other "JP" glyph, then afterwards can remap to JP, KR and CN. Anyway, unless I am wrong, I think there's nothing that says the 乚 must not touch the 丿 for CN/TW/HK.

Screenshot 2021-12-07 at 01 24 54

Then either no need to restore the v1 TW and map that current "JP" to TW/HK, or restore the other v1 TW form and drop the old "JP" glyph.

For 尤, I did put the character through my font list and it appears that maybe it should be treated separately for the CN region, but how the 丿 and 乚 strokes are applied is still inconsistent, especially between Sans and Serif. I probably suggest we review first and whether conforming all these characters containing 尤 to JP forms is acceptable for the CN region, and please comment with concrete evidence to prove otherwise.

Marcus98T commented 2 years ago

As of 2.001, while this left stroke thing in the 黑 component was made more consistent in HK/TW, I suggest CID 46941 be renamed to "uni9ED5-TW", CID 46942 be renamed to "uni9ED5-JP", and minor adjustments to the 冘 component for 46942 will still have to be done for consistency sake as shown below.

Screenshot 2022-01-27 at 13 50 14

CID 46942 (uni9ED5-CN) will be mapped to JP and KR once adjustment is done.

EDIT: Maybe keep the CN name for uni9ED5-CN.