Some Unicode glyph mappings suspect

[MattHeffron]

There are 6 pairs of Unicode encodings that each map to the same MCCS value (using (UTOMCODE? ), even though the glyphs are obviously different.
These collisions cause difficulty when constructing a FONTDESCRIPTOR, because they can replace a possibly better glyph.
Note that each pair has one Unicode value that has a single MCCS mapping, and the other has two mappings.

The glyph images below are from UNIFONT size 12 (BDF created from unifont-17.0.03.otf using otf2bdf).

Mapped to MCCS: 165,41 =29985
Unicode1: U+582F
UTOMCODE? = 29985

Unicode2: U+5B57
UTOMCODE? = (15226 29985)

Mapped to MCCS: 165,42 = 29986
Unicode1: U+600E
UTOMCODE? = (22371 29986)

Unicode2: U+69C7
UTOMCODE? = 29986

Mapped to MCCS: 165,43 = 29987
Unicode1: U+5FEB
UTOMCODE? = (12919 29987)

Unicode2: U+9059
UTOMCODE? = 29987

Mapped to MCCS: 165,44 = 29988
Unicode1: U+5E2B
UTOMCODE? = (15189 29988)

Unicode2: U+7464
UTOMCODE? = 29988

Mapped to MCCS: 165,45 = 29989
Unicode1: U+51DC
UTOMCODE? = 29989

Unicode2: U+8655
UTOMCODE? = (20829 29989)

Mapped to MCCS: 165,46 = 29990
Unicode1: U+7199
UTOMCODE? = 29990

Unicode2: U+76F8
UTOMCODE? = (16746 29990)

[rmkaplan]

The MCCS (XCCS) codes are all Kanji, and cset 165 is virtually empty for whatever reason. I would say "wont fix" unless we have someone who can figure out the better choices.

…

On Nov 28, 2025, at 4:24 PM, Matt Heffron ***@***.***> wrote: MattHeffron created an issue (Interlisp/medley#2390) <#2390> There are 6 pairs of Unicode encodings that each map to the same MCCS value (using (UTOMCODE? ), even though the glyphs are obviously different. These collisions cause difficulty when constructing a FONTDESCRIPTOR, because they can replace a possibly better glyph. Note that each pair has one Unicode value that has a single MCCS mapping, and the other has two mappings. The glyph images below are from UNIFONT size 12 (BDF created from unifont-17.0.03.otf using otf2bdf). Mapped to MCCS: 165,41 =29985 Unicode1: U+582F UTOMCODE? = 29985 image.png (view on web) <https://github.com/user-attachments/assets/193b5f84-8d3e-4f4e-9bb3-ec228ef74b90> Unicode2: U+5B57 UTOMCODE? = (15226 29985) image.png (view on web) <https://github.com/user-attachments/assets/28001db7-d95a-4a92-be75-f68e32ca03ef> Mapped to MCCS: 165,42 = 29986 Unicode1: U+600E UTOMCODE? = (22371 29986) image.png (view on web) <https://github.com/user-attachments/assets/d310068c-92d6-4509-a67c-e92a53d3b6f1> Unicode2: U+69C7 UTOMCODE? = 29986 image.png (view on web) <https://github.com/user-attachments/assets/05e81a04-ad79-422c-8c99-ef330ca27293> Mapped to MCCS: 165,43 = 29987 Unicode1: U+5FEB UTOMCODE? = (12919 29987) image.png (view on web) <https://github.com/user-attachments/assets/6085c24e-753c-4ffb-99b9-2372822b1d99> Unicode2: U+9059 UTOMCODE? = 29987 image.png (view on web) <https://github.com/user-attachments/assets/f9b2d547-24e0-4f2b-9f95-91fb30fa4964> Mapped to MCCS: 165,44 = 29988 Unicode1: U+5E2B UTOMCODE? = (15189 29988) image.png (view on web) <https://github.com/user-attachments/assets/b9bc1ae5-8e3d-418b-ac7d-e4dc86f9aeae> Unicode2: U+7464 UTOMCODE? = 29988 image.png (view on web) <https://github.com/user-attachments/assets/ff9a6b88-49a3-46b3-8fd0-8e455deef01e> Mapped to MCCS: 165,45 = 29989 Unicode1: U+51DC UTOMCODE? = 29989 image.png (view on web) <https://github.com/user-attachments/assets/b79f0dfc-5bfe-4407-8728-88bb474b17d0> Unicode2: U+8655 UTOMCODE? = (20829 29989) image.png (view on web) <https://github.com/user-attachments/assets/6e7b161c-5f36-4f41-9124-ae54e2d6510d> Mapped to MCCS: 165,46 = 29990 Unicode1: U+7199 UTOMCODE? = 29990 image.png (view on web) <https://github.com/user-attachments/assets/161a60f5-2279-4b81-b1e6-7a2c80b6bcdf> Unicode2: U+76F8 UTOMCODE? = (16746 29990) image.png (view on web) <https://github.com/user-attachments/assets/ea743bc5-18f9-4540-b105-8ffc96b1684d> — Reply to this email directly, view it on GitHub <#2390>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQSTUJM2K57O6FVJOO53GCL37DRU7AVCNFSM6AAAAACNQMK7YSVHI2DSMVQWIX3LMV43ASLTON2WKOZTGY3TMMJUHAYDEOI>. You are receiving this because you are subscribed to this thread.

[MattHeffron]

So for cset 165 we could ignore it, but are there any other cases where there are duplicate mappings that can happen. I'd like a strategy to handle this, even if just in READ-BDF.
I've been trying to verify that I'm loading the mapped glyphs correctly, by (first) comparing the number of mapped glyphs with the number of bits I set in a bitmap representing used MCCS codes.
For UNIFONT this is off by 6, which led me down this rabbit hole.

[MattHeffron]

All of these Unicode characters are in the "CJK Unified Ideographs" block (4E00-9FFF).
It seems this could be due to a bit of controversy in Unicode mapping.
"The block is the result of Han unification. ..."

[rmkaplan]

None of the glyphs that you pointed to resemble anything in XCCS 165, at least to my eye.

…

On Nov 28, 2025, at 6:28 PM, Matt Heffron ***@***.***> wrote: MattHeffron left a comment (Interlisp/medley#2390) <#2390 (comment)> All of these Unicode characters are in the "CJK Unified Ideographs" block (4E00-9FFF). It seems this could be due to a bit of controversy in Unicode mapping. "The block is the result of Han unification. ..." <https://en.wikipedia.org/wiki/CJK_Unified_Ideographs#CJK_Unified_Ideographs> — Reply to this email directly, view it on GitHub <#2390 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQSTUJOI4RUBFZZ44NSERQ337EAEJAVCNFSM6AAAAACNQMK7YSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTKOJQHEYDOMRRG4>. You are receiving this because you commented.