Closed maryewal closed 2 years ago
they should be consistent
In cross-checking just a couple, it seems to me that they are not. At least some Malekula languages in the main orthography folder do not reflect Timo's changes in the malekula folder. I suspect this may be why we still are seeing discrepancies in Malekula language counts. Can @antipodite make sure the Malekula folder changes are all reflected in the main?
I copied the malekula profiles into the main etc/orthography/
folder and reran the cldfbench lexibank.makecldf
command. However I'm not seeing any difference in phoneme counts. Could @LinguList tell us at what point the profiles are used to extract the phoneme counts? Is there something obvious I'm missing? When I did this in Feb with an updated profile it did work
@antipodite, did you run the command cldfbench abvdoceanic.inventories
which counts the phonemes, rather than the command that creates the CLDF data?
Yeah I did, should have specified I ran both, once each before moving the profiles, and then once each again after moving them, and then compared the results.
Le us check the diffs then, if you updated the profiles? Could you have a look at them, and see what kinds of changes you have there?
@antipodite, there is another issue. You swapped columns (at least in the first profile). The code reads colummn IPA, but you left IPA as is and called IPAInSource, where you edited, I think. This would explain why nothing happens. If you swap column names, using IPA for the new replacements, this should work, I hope.
@antipodite, if you find this issue to be solved, you can close it.
@antipodite should all of the updates that @tihomirrangelov made in https://github.com/lexibank/abvdoceanic/tree/main/etc/malekula also be reflected for the Malekula languages in the general orthography (https://github.com/lexibank/abvdoceanic/tree/main/etc/orthography) folder, or does it not matter?