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

KR glyph of 晩 U+6669: Inconsistency with Adobe-KR-9/Unicode representative glyph #89

Open spencer246 opened 3 years ago

spencer246 commented 3 years ago

Although the example font of Adobe-KR1, Source Han Serif AK9 (labeled as SHSrf-AK9 in the figure below) is derived from Source Han Serif, the AK9 version and the ordinary Korean version of Source Han Serif differ in how they render 晩 U+6669 (KS X 1001 repertoire, Adobe-KR Supplement 7):

U+6669 晩 6669

In the ordinary Source Han Serif (v1.001), the KR glyph of 晩 U+6669 looks almost the same as the KR glyph of 晚 U+665A (Adobe-KR Supplement 8; outside of KS X 1001):

U+665A 晚 665A

(For the record, in KR version of Source Han Sans v2, 晩 U+6669 and 晚 U+665A are also identical, and in fact their glyphs are completely merged into a single CID.)

The glyph of 晩 U+6669 in Adobe-KR version of the Source Han Serif is consistent with the representative glyphs of Unicode v13.0 representative glyphs (below), while that of 晩 U+6669 in the Source Han Serif / Source Han Sans is not.

6669 Unicode

665A Unicode

For the record, Malgun Gothic (Default KR font in Windows) renders 晩 U+6669 like the Unicode representative glyph, while Apple SD Gothic Neo (Default KR font in MacOS) renders it like Source Han Serif but both fonts do not have glyphs for 晚 U+665A.


As far as I know, when there are multiple z-variants in the Unicode repertoire that are not unified due to the source separation rule, Source Han Sans and Source Han Serif in general follow the representative glyphs displayed in the Unicode code chart, rather than render them as the same glyph. (EDIT: This may not be true, because other glyphs like TW variants of 郞 U+90DE and 郎 U+90CE are also rendered the same. EDIT2: On the second thought, it seems that the case of 郞 U+90DE and 郎 U+90CE is not relevant here because the TW glyph of 郎 U+90CE is simply wrong in Source Han Sans v2.001 and Source Han Serif v1.001; it deserves a separate issue.)

In this regard, is the difference of the glyph of 晩 U+6669 in the Adobe-KR-9 (i.e., in the Unicode code chart) and the KR version of the Source Han Serif v1.001 intended?