They have to be in the database table being used. Obviously "tgl" is there.

I didn't realise that any of the ISO 639-2 codes could be used. Has it always been like this?

@@ -10 +10 @@
-<lsource xml:lang="fil">lawaan</lsource>
+<lsource xml:lang="tgl">lawaan</lsource>

I think we should stick with the standard.

daij, koj
https://en.wiktionary.org/wiki/lauan

This comes from Tagalog but we only have a three-letter code for Filipino (the standardised form of Tagalog). Should we use it? Or could a code be added for Tagalog?

@@ -10 +10,2 @@
-<gloss>lauan</gloss>
+<lsource xml:lang="fil">lawaan</lsource>
+<gloss>lauan (tree, wood)</gloss>

I'm OK with yoji on this entry but I wouldn't mind dropping it either. It's included in 新明解四字熟語辞典.
I don't think the second gloss is needed.

@@ -18 +17,0 @@
-<gloss>era full of happiness and ideals</gloss>

Stray period.

@@ -18 +18 @@
-<gloss>era full of happiness and ideals.</gloss>
+<gloss>era full of happiness and ideals</gloss>

https://www.edrdg.org/projects/yojijukugo.html

This one is in 3 of the 4 sources on the project page, as well as in the yoji.jitenon.jp site. I'll leave the tag there for now, but I think we need to have an agreed way of settling on the use of the tag. I'll raise an issue for it.

I will repeat what I've said many times before on this topic - what does [yoji] here tell us about this word except the very obvious, namely that it's written with 4 characters? In Japanese, anything that consists of 4 characters can be called a 四字熟語, but if we're going to use it a way that's the least bit meaningful to our users, we have to make a  more narrow selection. That's why I once suggested we instead call it [iyoji] for idiomatic 4-ji jukugo to make it clearer we shouldn't just slap it onto anything. Nobody else seemed to favor it and I agreed at the time it was probbly fine as [yoji] but as we keep having this argument as if we're lost sight of what the intended use of it was, I'm not sure that was really the right decision.

I'm also not a fan of the new gloss. Meikyo has 2 senses, btw.

https://yoji.jitenon.jp/yoji/299.html?getdata=黄金時代&search=contain

Maybe we should accept that ref?

@@ -15,0 +16 @@
+<misc>&yoji;</misc>
@@ -16,0 +18 @@
+<gloss>era full of happiness and ideals.</gloss>

It's in ルミナス. This is OK.

GG5
doesn't "expansion of armaments" sound kinda clumsy?

https://books.google.com/ngrams/graph?content=++military+expansion,+arms+buildup,+expansion+of+armaments&year_start=1800&year_end=2019&corpus=26&smoothing=3&direct_url=t1;,military expansion;,c0;.t1;,arms buildup;,c0;.t1;,expansion of armaments;,c0

(軍備拡張 should be aligned)

@@ -20 +20,2 @@
-<gloss>expansion of armaments</gloss>
+<gloss>military expansion</gloss>
+<gloss>arms buildup</gloss>

daij, gg5

The refs say it's an abbreviation of 軍備拡張, not 軍事拡大.

@@ -18 +18 @@
-<xref type="see" seq="1991140">軍事拡大</xref>
+<xref type="see" seq="1248880">軍備拡張</xref>
@@ -20 +19,0 @@
-<gloss>military expansion</gloss>

Align.

@@ -12 +12,2 @@
-<gloss>expansion of armaments</gloss>
+<gloss>military expansion</gloss>
+<gloss>arms buildup</gloss>

Oops

Not sure only. I just checked the ichiman source document and it only has 故郷/こきょう and 古里/ふるさと so I'll drop the ichi1 from 故郷 in this entry.

@@ -5,4 +5 @@
-<keb>故郷</keb>
-<ke_pri>ichi1</ke_pri>
-<ke_pri>news1</ke_pri>
-<ke_pri>nf06</ke_pri>
+<keb>故郷news1,nf06]</keb>

Meikyo has 古里 first and 故郷 last, shinmeikai has 故郷 first and 古里 second.
I’m not sure what would be best here.
Are the freq tags still applicable here?

故郷	2158360
ふるさと	3520691
ふる里	145201
古里	103750

Probably the cleanest thing. In fact, GG5 has no kanji in its ふるさと entry.
There are 68 sentences; one with ふるさと and 67 with 故郷. My guess is that many people seeing 故郷 would think "ふるさと". For ふるさと 中辞典 has 【古里･故郷】. Should we have "ふる里,古里,故郷" or leave at is it is now? Certainly Googits for 故郷+ こきょう are far lower than for 故郷+ふるさと.

@@ -34,5 +33,0 @@
-<reb>こきょう</reb>
-<re_restr>故郷</re_restr>
-<re_pri>ichi1</re_pri>
-</r_ele>
-<r_ele>
@@ -43,0 +39 @@
+<xref type="see" seq="2853884">故郷・こきょう</xref>
@@ -51 +46,0 @@
-<stagr>ふるさと</stagr>

meikyo, shinmeikai, daijs, wisdom,

All of the refs have this split into こきょう and ふるさと
Can we please do the same?
こきょう is marked as a common word in two of the refs and is usually written in kanji.
ふろさと is not marked as common and is usually written in kana.

Quiet. Closing.

Not sure only. I just checked the ichiman source document and it only has 故郷/こきょう and 古里/ふるさと so I'll drop the ichi1 from 故郷 in this entry.

@@ -6 +5,0 @@
-<ke_pri>ichi1</ke_pri>

Meikyo has 古里 first and 故郷 last, shinmeikai has 故郷 first and 古里 second.
I’m not sure what would be best here.
Are the freq tags still applicable here?

故郷	2158360
ふるさと	3520691
ふる里	145201
古里	103750

Probably the cleanest thing. In fact, GG5 has no kanji in its ふるさと entry.
There are 68 sentences; one with ふるさと and 67 with 故郷. My guess is that many people seeing 故郷 would think "ふるさと". For ふるさと 中辞典 has 【古里･故郷】. Should we have "ふる里,古里,故郷" or leave at is it is now? Certainly Googits for 故郷+ こきょう are far lower than for 故郷+ふるさと.

@@ -34,5 +33,0 @@
-<reb>こきょう</reb>
-<re_restr>故郷</re_restr>
-<re_pri>ichi1</re_pri>
-</r_ele>
-<r_ele>
@@ -43,0 +39 @@
+<xref type="see" seq="2853884">故郷・こきょう</xref>
@@ -51 +46,0 @@
-<stagr>ふるさと</stagr>

meikyo, shinmeikai, daijs, wisdom,

All of the refs have this split into こきょう and ふるさと
Can we please do the same?
こきょう is marked as a common word in two of the refs and is usually written in kanji.
ふろさと is not marked as common and is usually written in kana.

@@ -12,0 +13 @@
+<pos>&vi;</pos>

Entry created

I don't think it's obscure. Plenty of hits online for "柑橘類 アルベド".

@@ -9 +9 @@
-<xref type="see" seq="2153840">反射能・はんしゃのう</xref>
+<xref type="see" seq="2153840">反射能</xref>
@@ -17 +17 @@
-<xref type="see" seq="2179560">中果皮・ちゅうかひ</xref>
+<xref type="see" seq="2179560">中果皮</xref>
@@ -20 +19,0 @@
-<gloss>mesocarp of a fruit</gloss>
@@ -21,0 +21 @@
+<gloss>mesocarp</gloss>

daijs has that 2nd sense too, but I eonder if it should maybe be tagged as [obsc]

ＫＯＤ追加語彙
https://en.wikipedia.org/wiki/Albedo_(disambiguation)

@@ -10,0 +11 @@
+<field>&physics;</field>
@@ -11,0 +13,9 @@
+<gloss>reflectivity</gloss>
+</sense>
+<sense>
+<pos>&n;</pos>
+<xref type="see" seq="2179560">中果皮・ちゅうかひ</xref>
+<field>&bot;</field>
+<gloss>albedo</gloss>
+<gloss>mesocarp of a fruit</gloss>
+<gloss>pith</gloss>

「モノのなまえ辞典」 (ISBN: 9784591164211) pg. 13

Regarding the existing entry for アルベド, it also seems to mean 'pith' or 'medulla' (or 'mesocarp'). See also: 中果皮, 髄, etc.

Note further the similar word フルベド that I submitted.

@@ -9,0 +9,2 @@
+<xref type="see" seq="2153840">反射能・はんしゃのう</xref>
+<field>&astron;</field>

enwiki
https://en.wikipedia.org/wiki/Direct_reduced_iron

@@ -12,0 +13,2 @@
+<gloss>direct reduced iron</gloss>
+<gloss>DRI</gloss>

Entry created

I agree with Marcus's suggestion to delete this entry and add 三鞭酒 as an ateji form to シャンパン. I'm not seeing any evidence that it's read さんべんしゅ when referring to champagne.

世界大百科事典 薬酒 entry
。蛤蚧酒はトカゲを漬け込んだもので，3種の動物の陽根をつけるという三鞭酒（サンピエンチユウ）などとともに，強精剤としての効果を信ずる人も多いという。

I 'm really not seeing a good case for splitting here. No source has さんべんしゅ. Several do have シャンパン (or rather, mention this kanji in their シャンパン entries) The rakuten/amazon listings aren't reliable. The drink referred to is an (at least in Japan) very obscure Chinese drink which doesn't seem to have anything to do with champagne, and is maybe actually read サンピエンチュウ. Now that we have rK, why not use it?

We've been thrashing this around for years. We settled on the two-entry approach in 2017, and I still think it's the cleanest. I'd like to keep さんべんしゅ there somewhere, which would make consolidation messy. I think the  しゃんぱん reading can go.

@@ -5,0 +6 @@
+<ke_inf>&ateji;</ke_inf>
@@ -9,3 +9,0 @@
-</r_ele>
-<r_ele>
-<reb>しゃんぱん</reb>
@@ -17 +15 @@
-<gloss>Champagne</gloss>
+<gloss>champagne</gloss>

That specifically is a Chinese product though, doesn't look to be related to actual CHampagne. (the listings don't mention シャンパン etc.)
I'd drop the reading or actually delete this entry and include it as a [rK,ateji] on the main シャンパン entry.

Yes, we should split on source language. I'll create a new entry.

@@ -6,3 +5,0 @@
-</r_ele>
-<r_ele>
-<reb>キーウ</reb>

https://languagelog.ldc.upenn.edu/nll/?p=53786

キーウ is being used by the BBC's Japanese operation as it more closely matches Kyiv. "Kiev" is more the Soviet era name. Possibly the entry should be split.

@@ -5,0 +6,3 @@
+</r_ele>
+<r_ele>
+<reb>キーウ</reb>

wiki

@@ -9,0 +10 @@
+<gloss>Kyiv</gloss>

I will remove it for now. 
The issue can be raised on Github

@@ -6,3 +5,0 @@
-</k_ele>
-<k_ele>
-<keb>2.5次元</keb>

This definitely isn't something that is in any way exclusive to this entry. If we're going to change how we do this, it should be across all entries.

It's been our standard forever to use zenkaku alphanumerics but I see your point. Google returns many more hits for 2.5次元 than ２．５次元. For terms like ＭＯＸ燃料 and ２進数 the Google hits are the same regardless of the character versions used.
We *can* use the plain ISO/Unicode alphanumerics as well, so I'll include them too.

@@ -5,0 +6,3 @@
+</k_ele>
+<k_ele>
+<keb>2.5次元</keb>

https://ja.wikipedia.org/wiki/2.5次元

The sources don't use fullwidth for the numerical portion of the word. Is that just a requirement on our end? Not a huge deal, but the spacing between the decimal point looks strange.

Robin is probably right. I'm OK either way.

新メニュー	316841
新しいメニュー	48679

I let it in as 新+loanword is rather the exception. I think it could be handy to be able to confirm it's in use.

I don't think we need an entry for a very simple collocation like this, especially when 新しいルール is also common. "new rule" is not an idiom. Nor is it something people would do a reverse lookup for.

Interesting.

新ルール	53025
新しいルール	21100


新フライト	87
新しいフライト	385


新ファッション   	2439
新しいファッション	12446


新相手   	51
新しい相手	4693

A+b but あたらしい is usu. more common than 新-

@@ -12 +12 @@
-<xref type="see" seq="1584695">面・めん・2</xref>
+<xref type="see" seq="1584695">面・2</xref>
@@ -17 +17 @@
-<xref type="see" seq="1584695">面・めん・3</xref>
+<xref type="see" seq="1584695">面・3</xref>

GG5

Not sure that "esp" is valid.

@@ -12 +12,8 @@
-<gloss>mask, esp. animal or human face, for use in plays or by children</gloss>
+<xref type="see" seq="1584695">面・めん・2</xref>
+<gloss>mask</gloss>
+</sense>
+<sense>
+<pos>&n;</pos>
+<xref type="see" seq="1584695">面・めん・3</xref>
+<field>&MA;</field>
+<gloss>striking the head (kendo)</gloss>

https://ejje.weblio.jp/content/お面

Similar in a way to 日本製 etc.

@@ -12 +12 @@
-<gloss>embassy of Japan</gloss>
+<gloss>Embassy of Japan</gloss>

日本大使館	141000
A+B, but I think the ngrams justify its inclusion.

GG5
州庁	1573

From 2834182.

中辞典

@@ -12 +12,3 @@
-<gloss>fishing method in which a band-shaped net is stretched in the sea like a fence to catch fish</gloss>
+<xref type="see" seq="1306460">刺し網</xref>
+<gloss>gill-net fishing</gloss>
+<gloss>tangle-net fishing</gloss>

https://kotobank.jp/word/刺し網漁-786967

帯状の網を海の中に垣根のように張って魚をとる漁法

KOD: 中央語 【言語】 a central language; a metropolitan language; metropolitan 《French》.

"central language" works but it's a bit literal.

daijs
その国の政治・文化の中心地で使われる言葉。
[補説]標準語・共通語と完全に一致はしない。

overly long and literal definition, probably, but it's really hard to pin down the difference between this word and 標準語 and 共通語 otherwise
"language of the elite"?
"central language"?
"main language"?

"normal use" is maybe not great, on second thought

@@ -13 +12,0 @@
-<gloss>normal use</gloss>

gg5
saw in amazon review: α7Ⅲを二年使っての買い増しです。とにかく小さくて軽い。レンズキットのレンズは小さいけれど28-60で普段使いでは問題ありません。

Quiet. Closing.

Splitting, as suggested in 1603050.