Glottocode errors in a few lexicore datasets

[XachaB]

Hi,

I spotted a few glottocodes which seem unknown of glottolog, listed below. I am making a single issue here, rather than 8 issues each on a single dataset bug tracker.

When it was clear which was the correct glottocode (exact same variety name, similar glottocode which makes it look like a type), I provide the correct code too:

dataset	given_glottocode	correct_glottocode	language_name
dunnaslian	sema1250	semn1250	Semnam Malau
dunnaslian	teim1246	temi1246	Temiar Perak
dunnaslian	monn1258	monn1253	Mon
kesslersignificance	nucl1201	nucl1301	Turkish
polyglottaafricana	maka1261	makh1261	Makhuwa-Meetto
saenkoromance	vall1248		Vallader_Romansh
saenkoromance	cagl1238		Campidanese
servamalagasy	meri1291	meri1243	Merina
sidwellbahnaric	kass1248		Kasseng
transnewguineaorg	cent2257		Proto-Central-Sogeram
zgraggenmadang	sali1249	saki1249	maia-saki
zgraggenmadang	para1207	para1307	parawen

I'll make a few PRs when I did find a correction.

Shouldn't cldfbench have checked this automatically and flagged the codes as incorrect ? If not, that would probably be a useful check to add.

[XachaB]

Note: I am only making changes to the languages files, in order to do this quickly on the web interface. If PRs are accepted, the cldf datasets need to be regenerated by someone who has the whole cldfbench environment ready :)

[SimonGreenhill]

Thanks @XachaB -- I've handled dunnaslian, zgraggenmadang, servamalagasy.

[LinguList]

Thanks. All datasets need of course to be re-released later.

[MuffinLinwist]

Thanks, @XachaB, for pointing this out. I'm checking all of the cases listed here and preparing a PR for the new release.

[MuffinLinwist]

I already addressed these wrong Glottocodes and am preparing a PR with the fixes. The only one pending is the Glottocode for Proto-Central-Sogeram (which needs another fix because it's downloading the language metadata from a url in the lexibank script). I wrote an issue about it on the repo and can be access here.

I'll reference this issue once the PR is open and close it once the PR is merged.

[MuffinLinwist]

All of these datasets are fixed now. @chrzyki, perhaps we can close this issue if everything is fit. I can, then, go ahead and proceed with #47.

[chrzyki]

Awesome, thanks for taking care of this @MuffinLinwist. Double-checked the repos again and everything looks good!