Multigraph characters flagged as errors when capitalized

Question

One of the translation projects that I’m supervising has an issue with Basic Checks errors, and I’m looking for the correct way to resolve them.

The word/name “God” is spelled uNguluvi. For linguistic reasons, sentence medially it is capitalized on the N, not the initialy u.
The alphabetical character settings have the character ng. The help files say that we should not enter capitalized forms of multigraphs, therefore, I have not entered ng/Ng in the alphabetical characters.

With this setup, Paratext is flagging many errors saying: “Invalid or unknown character: g”. We do not have a character g, only ng. Because of this, we have had to deny over 100 errors in a single book. Is there not a better way to handle this? I could put g as a valid character, but that would sort of defeat the intent of the checks, since g is not a valid character elsewhere. Can I enter ng\Ng in the alphabetical characters settings or is there a reason that that is not advised?

Thanks!

Postingan lama - ditampilkan dalam bahasa aslinya

Answer 1

Stephen+Katt,

This points out an issue with the way the alphabetic characters are currently handled. Unfortunately I don’t think there is currently a way to have the Ng in the list. However, you could go ahead and add the “g” to the list. Then to see if there are any "g"s that occur without a preceding “n” you could go to the Wordlist and in the filter window search for:

regex:[^nN]g

This says look for a regular expression that is a “g” proceeded by anything that is not “n” or “N”.

Not ideal, but it would avoid lots of denied messages. As for denying messages - one way to do this is to sort the error list by Message (View > Sort By). This puts all the same error together and you can select all of them to deny at one time. Just to say the obvious - it is better to fix errors rather than deny them!

Blessings,
anon848905

Postingan lama - ditampilkan dalam bahasa aslinya

Comments

Thank you for the reply anon848905. I think I will go ahead and add g as a valid character and we’ll make sure to add that check for lone “g” as a part of our checks.

Postingan lama - ditampilkan dalam bahasa aslinya