kvg:element="⺍" kvg:original="つ"

[runarberg]

It seems like most instances of components with kvg:element="⺍" have the kvg:original="つ", instead of the expected 小. Is this intended or is this a bug?

Note that this is not 100% consistent, I counted 5 instances with kvg:original="吅" and 2 with kvg:original="巛", none had the expected kvg:original="小".

[benkasminbullock]

This came up before, but I don't remember what the answer was, and I couldn't find where in the mailing list or the old issues it came up. I'll keep searching but I think this is a duplicate.

[runarberg]

The only case which is giving me trouble is 営 (kanji/055b6.svg) which is the only one of these that has ⺍ as a radical. But looking it up, it seems like it may be a mistake. Kanjidic2 only lists number 30 (口) as a radical, while here 口 is a nelson radical while ⺍ (original つ) is a tradit.

<!-- Entry for Kanji: 営 -->
<character>
  <literal>営</literal>
  <codepoint>
    <cp_value cp_type="ucs">55b6</cp_value>
    <cp_value cp_type="jis208">1-17-36</cp_value>
  </codepoint>
  <radical>
    <rad_value rad_type="classical">30</rad_value>
  </radical>

[benkasminbullock]

I seem to remember getting an explanation of what this つ was from someone or another but I really cannot find that, I have searched all the issues here, searched the mailing list, and searched my gmail account without success. It should be documented but I have no records at the moment.

[runarberg]

I did some research and ⺍ is complicated to say the least. Kradfile/Radkfile simply treat ⺍ as ⺌ (denoted as 尚 over there) and as such a variation on radical no. 42 (小). English Wikipedia page on Kangxi radicals agrees with that, but the page on radical 42 doesn’t, and neither does Wiktionary. I couldn’t find anywhere other where ⺍ is a variation of 小. And personally, I don‘t think that is correct, nor even helpful.

I found exactly one character which correctly has ⺍ as its radical (営 has it incorrectly as a radical), and that is 巣 which has classical radical no. 47 巛 (and a nelson radical no. 3 丶). Kanjivg treats this correctly with <g kvg:element="⺍" kvg:original="巛" kvg:radical="tradit">, but this is not true generally. ⺍ seems to be derived from different components in different kanji. For example 営 is simplified from 營 so ⺍ should probably have kvg:original="炏" and 学 is simplified from 學 so original should be 𦥑. Note that in the case of 厳 kvg:original correctly has the value 吅 as the kanji is simplified from 嚴.

Perhaps the correct solution here is to go over all the characters which have kvg:original="つ" and replace it with what the ⺍ part is a simplification of, like is currently the case with 営 and 厳.

[benkasminbullock]

I'm planning to ask on the mailing list about this, if you want to do that instead of me, it would be a good idea. I have completely forgotten what the つ is and as you say it is hard to justify.

[benkasminbullock]

I've sent a message to the mailing list here:

https://groups.google.com/g/kanjivg/c/y_9QKbfQoK0

[benkasminbullock]

Perhaps the correct solution here is to go over all the characters which have kvg:original="つ" and replace it with what the ⺍ part is a simplification of, like is currently the case with 営 and 厳.

Nobody objected on the mailing list so if you would be so kind as to do this work and send a pull request I will add it to the repository.