While I agree on the variants relationship provided by Kushim in comment #150, I think the fact that there are two independent historical use cases for ⿰衤奐 (Evidence 1 and 3) and one modern annotations usage (Evidence 4) already supported a separate codepoint for ⿰衤奐.
If we encoded ⿰衤奐 as U+2C862 + VS17, then it would be very confusing when we digitalize evidence 3 by U+2C862 + VS17, because here ⿰衤奐 was never a variant for U+2C862 or 襖. On the other hand, as the radicals of 煥 is different to the one of ⿰衤奐, it is not better to encode ⿰衤奐 as an IVD for 煥, either.
If UK were to HE U+2C862, it raises the concern of over-unification because U+2C862 is not intended as an error form of 煥, they have different shape and pronunciations.
The component 奐 is very productive while the 𪥌 component is rarely used (only 𪶓 and 𬡢) in encoded characters. Encoding ⿰衤奐 will offer a better unification "anchor" for any future variants if they are discovered.
Evidence 5 and 6 give ⿰⿱𮫙革成, unifiable with ⿰鞷成. ⿰鞷成 is a variant form of U+29AE8 𩫨. 《元始天王大洞玉經》55a gives the alternative form ⿰⿳一品幸烕 (not submitted). Consider introduce new UCV 鬲/𮫙.
Evidence 3 shows ⿰氵𢑴, evidence 4 shows ⿰氵彛, evidence 5 and evidence 6 show ⿰氵𢑱; they are all potentially unifiable variants of ⿰氵彝 given by evidence 1 and 2
In the fourth evidence, the fanqie 古旦反 suggests that it is a variant of 幹. The U+2D16A 𭅪 is also a variant of 幹. Suggest unify to 𭅪 as long as they are cognate.
If evidence 2 is indeed ⿰氵⿱𫂁马, suggest to change IDS and glyph to ⿰氵⿱𫂁马 so that it is the legit simplified variant of 𤅲 (U+24172), and then we can unify ⿰氵⿱𮅕马 to ⿰氵⿱𫂁马. Otherwise we will end up with the situation where 𤅲 is missing encoded simplified form and ⿰氵⿱𮅕马 is missing encoded traditional form, which may lead to two more encoded characters in the future.
They all look like modern misinterpretation of 䜌󠄂 (U+470C + VS19) (Also the KP glyph for 䜌). Although the glyph is extensively used in books published in the digital era by 中華書局 and other publishers:
It will be great to provide even one historical evidence. I suspect ⿱亠䜌 is a convenient digital artifact for the desired 䜌 + VS19 shape.
According to the original evidence 《［雍正］山東通志》, 朱慶⿰金⿱幺夊 is elected as 崇禎丙子科（1636）舉人 and he is a descendent of 魯荒王朱檀.
According to 《皇明祖訓》：「凡東宮、親王位下、各擬名二十字。日後生子及孫，即以上聞，付宗人府。所立雙名，每一世取一字以為上字；其下一字，臨時隨意選擇，以為雙名，編入玉牒……魯王位下：肇泰陽當健、觀頤壽以弘」, given that his last name is of the 金 radical and he was active in 17th century, his generation name should have been 壽, and therefore his full name should have been 朱壽⿰金⿱幺夊, instead of 朱慶⿰金⿱幺夊.
An IRG expert YUAN Zhiyu has generously provided me a copy of an half folio from a 雍正 era copy of《崇禎丙子科山東鄉試齒錄》, one of the Peking University rare books collection, which indeed gives his name as 朱壽⿰金⿱幺犮 (犮 + VS19).
▲ 《崇禎丙子科山東鄉試錄》（影抄北京大學藏雍正抄本）
Because this evidence predates 《［雍正］山東通志》 and sourced from a Ming dynasty document 《崇禎丙子科山東鄉試錄》, it should be more authoritative than the original evidence. It also makes much more sense than ⿰金⿱幺夊 because both ⿰金⿱幺友 and ⿰金⿱幺犮 are also attested in contemporary dictionaries:
Here is a list of ⿰金⿱幺友 in a loose chronological order.
Given that the ⿰金⿱幺友 evidences predates the ⿰金⿱幺犮 + VS19 ones, and there is only one stroke difference between ⿰金⿱幺夊 and ⿰金⿱幺友, I suggest we encode the ⿰金⿱幺友 shape with both ⿰金⿱幺夊 and ⿰金⿱幺犮 + VS19 being unifiable variant of ⿰金⿱幺友.
But since the current earliest evidence from ［康熙］山西通志 predate this one, we can not determine if ⿰金運 is an error for 鏈 or 鏈 is an error for ⿰金運. So we noted the alternative text 鏈 and hoped that experts can unearth more information on this character.
And here is 𨪵 from 《漢語大字典》p. 4567, which also quotes 《清史稿》:
𨪵：人名用字。《清史稿•世祖紀一》：「乙未，朱聿釗弟聿鐭僭號紹武，據廣州，佟養甲、李成棟率師討之，斬（朱）聿鐭及……鉅野王壽𨪵。」, apparently 朱壽𨪵 in the evidences above and 朱壽⿰金朐 in 《南明史》are the same person.
Since 《南明史》, published by 中華書局 in 2006, is compiled from 錢海岳's manuscript, and 目 is easily confusable with 月 in handwritten text. I think 錢海岳 was meant for U+28AB5 𨪵 but the editors recognized the handwritten shape as ⿰金朐. Therefore, I suspect this is an one-off error and suggest pending more evidences.
In the new evidence, there are ⿳艹冖⿺元⿱⺊又 in Mincho and ⿱艹冦 in Gothic, which are apparently variants of each other as they refer to the same place name. I believe the shape ⿱艹冦 is more convenient for the practise use. I suggest to encode the normalized form ⿱艹冦 instead and update the IDS and glyph.
The right component ⿵戊𡚬 is also similar to U+2AB4F 𪭏, which is also a variant of 威 according to CNS 11643.
If ⿰王⿵戊𡚬 were encoded as-is, given that ⿰王威 is already attested as KC-10302, should we encode ⿰王威 separately in the future? If not, should ⿰王威 be an IVD of ⿰王⿵戊𡚬? If not, should we encode the normalized form ⿰王威 and add an IVD of ⿰王威 to represent ⿰王⿵戊𡚬?
While the first evidence clearly gives ⿸彧虫, the second and third evidences give ⿱彧虫 and they were earlier than the first one. Some other later sources give 𧌒 or 蜮.
So ⿱彧虫 is likely a variant of 𧌒, normalize ⿸彧虫 to ⿱彧虫 to align with the 𧌒 production. I suggest to keep the current IDS ⿱彧虫 and the current glyph as-is.
The original evidence is missing one horizontal stroke in the 春 component. And the annotation gives ⿱夫月.
Other
I don't object the current glyph design & IDS. Like you said both ⿱夫日 and ⿱夫月 are variants of 春. My previous comment is to note the normalization involved here. I think the normalized form ⿰口⿱椿火 is preferred over the exact form ⿰口⿱⿰木⿱夫日火, one can always add an IVD of ⿰口⿱椿火 to present the desired ⿰口⿱⿰木⿱夫日火 shape in 正統道藏.
Aside: In this evidence, the last character in the same column of ⿰口⿱椿火, ⿰口⿱𰟐水 is written as ⿰口⿱⿰火㇯堇一:
The normalization may be inevitable when dealing with ancient text, because they might have different normalization rules: The text here is authored well before the 15th century. As we can see, the shape of 堇 component here is consistent with contemporary dictionary:
▲ 龍龕手鑑（臺北故宮藏宋刊本）卷1 folio 5a
I think we should encode the modern normalized form ⿰口⿱𰟐水 instead of the exact shape ⿰口⿱⿰火㇯堇一, because the standard is for modern audience.
Good catch. We believe the ⿺虎戊 is the desired form because 1) The evidence in 雍正廣西通志 predates the one in 道光潯州府志. 2) 《炎徼紀聞》, an earlier source gives 𧇭, 3) the semi-cursive script of 武 could be similar to 戊, and 4) The text is from 翁萬達《藤峽善後議》. In 《中州音韻》, 武 is 微母魚模合上聲, 戊 is 微母魚模合去聲, so the pronunciation of 武 is also very similar to 戊 in Ming Dynasty.
For the new ⿰氵兠 evidence mentioned in comment #704, 《［康熙］德清縣志》gives ⿰土兠, which is a variant of U+2EC2A 𮰪, and will be unifiable to 𮰪 if we agree upon the proposed UCV.
IRG Working Set 2024v1.0
Source: HUANG Junliang
Date: Generated on 2026-10-04
Unification
Showing 13 comments.
Current cognate disunified examples:
U+2361B 𣘛 = U+6A77 橷
I agree on the unification comment #258. There are only one 丿 stroke difference between ⿰土⿱亠屰 and 𡍩. In WS2017, we have already unified ⿰犭⿳⿱⺊乛七巾 to 𤟝.
[ {{WS2017-02527}} ]
𤆛 is also a variant of 与. In fact, the third evidence 一切経音義高麗版 clearly gives 𤆛.
If we encoded ⿰衤奐 as U+2C862 + VS17, then it would be very confusing when we digitalize evidence 3 by U+2C862 + VS17, because here ⿰衤奐 was never a variant for U+2C862 or 襖. On the other hand, as the radicals of 煥 is different to the one of ⿰衤奐, it is not better to encode ⿰衤奐 as an IVD for 煥, either.
If UK were to HE U+2C862, it raises the concern of over-unification because U+2C862 is not intended as an error form of 煥, they have different shape and pronunciations.
The component 奐 is very productive while the 𪥌 component is rarely used (only 𪶓 and 𬡢) in encoded characters. Encoding ⿰衤奐 will offer a better unification "anchor" for any future variants if they are discovered.
Copied from the submission note:
Evidence 5 and 6 give ⿰⿱𮫙革成, unifiable with ⿰鞷成. ⿰鞷成 is a variant form of U+29AE8 𩫨. 《元始天王大洞玉經》55a gives the alternative form ⿰⿳一品幸烕 (not submitted). Consider introduce new UCV 鬲/𮫙.
Copied from the submission note:
Evidence 3 shows ⿰氵𢑴, evidence 4 shows ⿰氵彛, evidence 5 and evidence 6 show ⿰氵𢑱; they are all potentially unifiable variants of ⿰氵彝 given by evidence 1 and 2
Unencoded cognate pairs:
U+26266 𦉦 = U+24AE1 𤫡 = ⿰王𦉍 B03502-006
Evidence 1 gives ⿰片𭺛; evidence 2 gives ⿰片𭺜, normalized to ⿰片瓦. Consider extending UCV #439 to include 𭺛 and 𭺜.
鼎/𣇓/𪔂/⿶㇯𪔂目日.
Cognate but currently disunified pairs:
U+85A1 薡 = U+26F4D 𦽍
In the fourth evidence, the fanqie 古旦反 suggests that it is a variant of 幹. The U+2D16A 𭅪 is also a variant of 幹. Suggest unify to 𭅪 as long as they are cognate.
Suggest unification with 𬜻.
There is only one stroke difference between ⿳艹从干 and 𬜻 + VS17:
We can add a new IVD for 𬜻.
The proposed UCV 廛/𠪨 was also discussed in the link above. For reference here is Kushim's comment in that thread on the application rate of this UCV:
<blockquote>
For 𠪨~廛:
𧓋 (C12501) ~ 𧔊 (C12501-001)
𧸪 (C13655-002) ~ 𧸳 (C13655)
𨞬 ~ 鄽 (GHZR: 4054.04)
⿰金𠪨 (C15168-002) ~ 𨮻 (C15168)
⿰𧾷𠪨 (B04971-005) ~ 躔 (B04971)
⿵門𠪨 (C15366-001) ~ 𨷠 (TB-2215) 𨷭 (TB-2219)
⿺走𠪨 (C13847-002) ~ 𧾡 (C13847)
So it's okay to add UCV 𠪨~廛.
</blockquote>
In IRG #61, there were different opinions about this UCV. It will be great if we can reach consensus this time.
Agree. Unify to 𡅎 (U+2114E) by UCV #213a.
Attributes
Showing 7 comments.
If both of them are accepted for encoding, the IDS of this character should be updated to ⿰扌WS2024-03272.
Note that Tsukada Masaki, the author of L2/23-130 also believes that the left component should be 山:
It should be safe to normalize the IDS to ⿱艹冦.
Evidence
Showing 27 comments.
▲过增元,梁新刚,朱宏晔.⿰火积——描述物体传递热量能力的物理量[J].自然科学进展,2006(10):1288-1296.
Guo et al. also created a new English word for ⿰火积: Entransy, a blend of en[ergy] + trans[fer] + [abilit]y, in their 2007 paper Entransy—A physical quantity describing heat transfer ability.
▲ 金臺集（毛氏汲古閣明末刊本）卷2 folio 3a
⿱入疌 also shares the same meaning with 𤴘 according to MOE dictionary:
▲ ［民國］鄰水縣志四十五卷（鄰水縣檔案館藏民國三十七年稿本）卷17農林志 page 3
In this evidence 樟⿰木𡿺 is a variant of 樟腦.
Credits to renyi who provided the evidence.
▲ 《［嘉靖］常熟縣志》（明嘉靖刊本）卷1 folio 16a
▲ 《吳中水利全書》（明崇禎刊本）卷5 folio 61a
The second evidence gives ⿰氵𨃱, potentially unifiable with ⿰氵𨂝.
▲ 渝北童氏族譜（清宣統二年刊本）四房卷下 folio 33b
Credits to renyi who provided the evidence.
▲ 《［萬曆］湖州府志（1963年上海古籍書店影印本）》卷13 folio 5a
Both new evidence and the submitted evidence 2 give ⿰氵兠.
It will be great to provide even one historical evidence. I suspect ⿱亠䜌 is a convenient digital artifact for the desired 䜌 + VS19 shape.
▲ 《文字蒙求廣義》（清光緒江楚書局刊本）卷3 folio 11b
The first evidence is from 《漢語大字典》, which quotes 龍龕手鑑. 《龍龕手鑑》（南宋刊本）gives 𱧅, as is provided as the original evidence for WS2017-01992:
▲ 《龍龕手鑑》（南宋刊本）卷2 folio 12a 水部平聲
The second original evidence is from 《宋史·卷218》, 《宋史》（朝鮮刊本）gives 𱧅.
▲ 《宋史》（朝鮮刊本）卷218 folio 33b
The third original evidence is from 《宋史·卷239》, 《宋史》（朝鮮刊本）gives 𱧅
▲ 《宋史》（朝鮮刊本）卷239 folio 9a
[ {{WS2017-01992}} ]
▲ folio 8a
If SAT wants to separately code ⿰亻⿱乖一, consider normalize it to ⿰亻埀 and add ⿰亻⿱乖一 as an IVD of ⿰亻埀.
▲ 《［雍正］山西通志》（清雍正刊本）卷26 folio 18a
▲ 蓝德康, 松冈荣志:《汉字海》(北京: 华语教学出版社; 北京中易中标电子信息技术有限公司, 2018) [ISBN 978-7-5138-1500-0] p. 1521
Here are two new evidences:
▲ 新修絫音引證群籍玉篇（中華再造善本影印金刊本）卷22 folio 30b
▲ 精刻海若湯先生校訂音釋五侯鯖字海（明末刊本）卷17 folio 3a
In both new evidences, ⿰石𢒈 is in 石部七畫, which also matches our SC data here. Should it indeed be ⿰石𰐨, it would have been placed in 石部九畫 instead.
▲ 宋史（朝鮮刊本）卷218 folio 33b
▲ 《龍龕手鑑（南宋刊本）卷4 日部入聲
▲ 《欽定續通志》(文淵閣四庫全書本) 卷61 folio 16a
▲ 貴州通志（民國37年鉛印本）土司志六 folio 18a
The new evidence gives ⿰王𩶷, unifiable to ⿲王魚亍 by UCV #322.
According to 《皇明祖訓》：「凡東宮、親王位下、各擬名二十字。日後生子及孫，即以上聞，付宗人府。所立雙名，每一世取一字以為上字；其下一字，臨時隨意選擇，以為雙名，編入玉牒……魯王位下：肇泰陽當健、觀頤壽以弘」, given that his last name is of the 金 radical and he was active in 17th century, his generation name should have been 壽, and therefore his full name should have been 朱壽⿰金⿱幺夊, instead of 朱慶⿰金⿱幺夊.
An IRG expert YUAN Zhiyu has generously provided me a copy of an half folio from a 雍正 era copy of《崇禎丙子科山東鄉試齒錄》, one of the Peking University rare books collection, which indeed gives his name as 朱壽⿰金⿱幺犮 (犮 + VS19).
▲ 《崇禎丙子科山東鄉試錄》（影抄北京大學藏雍正抄本）
Because this evidence predates 《［雍正］山東通志》 and sourced from a Ming dynasty document 《崇禎丙子科山東鄉試錄》, it should be more authoritative than the original evidence. It also makes much more sense than ⿰金⿱幺夊 because both ⿰金⿱幺友 and ⿰金⿱幺犮 are also attested in contemporary dictionaries:
Here is a list of ⿰金⿱幺友 in a loose chronological order.
▲ 《泰和五音新改併類聚四聲篇》（金崇慶荊國器刊元修補本）卷2 folio 4a (⿰金⿱⿰𠂈丶友、七𭭕切, this character is from 《搜真玉鏡》)
▲ 《泰和五音新改併類聚四聲篇》（元張仁刊本）卷2 folio 4a
▲ 《成化丁亥重刊改併五音類聚四聲篇》（明成化丁亥(三年)至庚寅(七年)金臺大隆福寺集貲刊本，故善012372-012376，國立故宮博物院）卷2 folio 4b (Note that from this version, the fanqie is modified from 七歡切 to 七勸切)
▲ 《新校經史海篇直音》（明萬曆刊本）卷1 folio 4a (Note that 音茜 is very likely to be derived from the fanqie 七勸切 in 《成化丁亥重刊改併五音類聚四聲篇》)
▲ 《刻太古遺踪海篇集韻大全》（明萬曆17年刊本）卷23 folio 4a
▲ 《重校全補海篇直音》（明萬曆23年刊本）卷1 folio 5a
▲ 《重校古本五音類聚四声切韻直音海篇大全》（明萬曆30年刊本）卷1 folio 5b
▲ 《翰林重攷字義韻律大板海篇》（明刊本）卷15 folio 16b
Here is a list of ⿰金⿱幺友 in a loose chronological order.
▲ 《翰林筆削字義韻律鰲頭海篇心鏡》（明萬曆10年序刊本）卷15 folio 16b
▲ 《翰林重攷字義韻律大板海篇心鏡》（明萬曆24年刊本）卷15 folio 18a
▲ 《遵古本正韻石齋海篇》（明崇禎刊本）卷14 folio 18a (a.k.a 新刻洪武元韻勘正切字海篇群玉)
▲ 《精刻海若湯先生校訂音釋五侯鯖字海》（明末刊本）卷1 folio 4b
Given that the ⿰金⿱幺友 evidences predates the ⿰金⿱幺犮 + VS19 ones, and there is only one stroke difference between ⿰金⿱幺夊 and ⿰金⿱幺友, I suggest we encode the ⿰金⿱幺友 shape with both ⿰金⿱幺夊 and ⿰金⿱幺犮 + VS19 being unifiable variant of ⿰金⿱幺友.
▲ ［嘉慶］沁水縣志（清嘉慶刊本）卷3 folio 3b
But since the current earliest evidence from ［康熙］山西通志 predate this one, we can not determine if ⿰金運 is an error for 鏈 or 鏈 is an error for ⿰金運. So we noted the alternative text 鏈 and hoped that experts can unearth more information on this character.
▲ 《清世祖實錄》juan 30 folio 24b gives ⿲金目勾, which can be considered as a variant of 𨪵
▲ 《八旗通志初集》（清乾隆刊本）juan 138 gives ⿲金目勽, corrupted form of ⿲金目勾.
▲ 《東華錄》（清光緒石印本）順治 juan 2 gives 𨪵
▲ 《清史稿（1927）》juan 4 folio 14b also gives 𨪵.
And here is 𨪵 from 《漢語大字典》p. 4567, which also quotes 《清史稿》:
𨪵：人名用字。《清史稿•世祖紀一》：「乙未，朱聿釗弟聿鐭僭號紹武，據廣州，佟養甲、李成棟率師討之，斬（朱）聿鐭及……鉅野王壽𨪵。」, apparently 朱壽𨪵 in the evidences above and 朱壽⿰金朐 in 《南明史》are the same person.
Since 《南明史》, published by 中華書局 in 2006, is compiled from 錢海岳's manuscript, and 目 is easily confusable with 月 in handwritten text. I think 錢海岳 was meant for U+28AB5 𨪵 but the editors recognized the handwritten shape as ⿰金朐. Therefore, I suspect this is an one-off error and suggest pending more evidences.
▲ 《□雲岩碑記》
▲ 《三晉石刻大全·大同市廣靈縣卷》三晉出版社 2013.12 pp. 59
It would be great if there are multiple land registration record evidences. One evidence seems not enough to back this character with unusual shape.
▲ 高豆⿱艹冦町報かわにし 2018年2月号（15ページ）pp. 29
In the new evidence, there are ⿳艹冖⿺元⿱⺊又 in Mincho and ⿱艹冦 in Gothic, which are apparently variants of each other as they refer to the same place name. I believe the shape ⿱艹冦 is more convenient for the practise use. I suggest to encode the normalized form ⿱艹冦 instead and update the IDS and glyph.
▲ 《新集古文四聲韻》（中華再造善本影印宋刊本）卷3 folio 28a 上聲厚四十六
Glyph Design & Normalization
Showing 4 comments.
In evidence 1, the dot is very subtle and it may well be overlooked.
If ⿰王⿵戊𡚬 were encoded as-is, given that ⿰王威 is already attested as KC-10302, should we encode ⿰王威 separately in the future? If not, should ⿰王威 be an IVD of ⿰王⿵戊𡚬? If not, should we encode the normalized form ⿰王威 and add an IVD of ⿰王威 to represent ⿰王⿵戊𡚬?
▲ 《皇明文徵》（明崇禎刊本）卷1 folio 12b
▲ 《御定歷代賦彙》（清康熙刊本）卷19 folio 13b.
So ⿱彧虫 is likely a variant of 𧌒, normalize ⿸彧虫 to ⿱彧虫 to align with the 𧌒 production. I suggest to keep the current IDS ⿱彧虫 and the current glyph as-is.
Other
Showing 10 comments.
Yes the quote is from 《史記·司馬相如列傳·上林賦》:「滭浡滵汨，湢測泌瀄」.
In evidence 2, 滭 is changed to 澤, which does not quite make sense since 滭浡 is a 雙聲詞. The evidence 1 further changed 汨 to 洞.
⿰氵郣 is likely a variant of 渤. Here is 郣 in Kangxi dictionary: 「郣……【正字通】通作渤」.
?
?
Aside: In this evidence, the last character in the same column of ⿰口⿱椿火, ⿰口⿱𰟐水 is written as ⿰口⿱⿰火㇯堇一:
The normalization may be inevitable when dealing with ancient text, because they might have different normalization rules: The text here is authored well before the 15th century. As we can see, the shape of 堇 component here is consistent with contemporary dictionary:
▲ 龍龕手鑑（臺北故宮藏宋刊本）卷1 folio 5a
I think we should encode the modern normalized form ⿰口⿱𰟐水 instead of the exact shape ⿰口⿱⿰火㇯堇一, because the standard is for modern audience.
Is this character a variant of 鸞?
▲ 《炎徼紀聞》（明嘉靖刊本）卷2 folio 16a.
Data for Unihan
Showing 5 comments.
For the new ⿰氵兠 evidence mentioned in comment #704, 《［康熙］德清縣志》gives ⿰土兠, which is a variant of U+2EC2A 𮰪, and will be unifiable to 𮰪 if we agree upon the proposed UCV.
▲ 《［康熙］德清縣志》（清康熙刊本） 卷2 folio 17b
Therefore ⿰氵兜 is a semantic variant of 𮰪.
▲ 四川總志（清康熙刊本）卷30 folio 10b 張德地〈題爲恭報鹽課鹽稅銀兩等事〉：「梁（山）、萬、新（寧）、開四縣人民俱各於忠州𡍼渰二井背負食用」
▲ 蜀水攷（清光緒刊本）卷4 folio 13b：「涂渰二溪皆有鹽井，唐所置塗㽏監也，㽏即渰」
《新唐書·食貨志》：「順宗時……增雲安渙陽塗𣉯三監」
Related characters:
The fanqie implies that ⿰金⿱幺友 is a variant of 鋑.