I have a problem related to the issue “Glossary unexpected non-wordforming characters”, but I’m moving my original question to this new thread because in my case I’m seeing “unexpected non-wordforming character” errors for text within glossary markups; ie text that is linked to a glossary entry between \w…\w* or +w…+w* markers (style name: Peripheral Ref - Wordlist Entry).
The Markers basic check seems to expect single words between the \w…\w* markers and throws an error when non-wordforming characters are used there. In our case, we have some acceptable keyterm renderings that include commas, and many more cases that include hyphens. The validity of the contexts for these punctuations’ usages are set correctly in the Punctuation Inventory (ie hyphen is only marked as valid word-medial and comma is marked as valid only word-final). They are both listed as valid in the Characters Inventory as well, but we don’t want to list either of them as word-medial punctuation in Project Settings > Language Settings > Other Characters because that would create many different words in the word list for each combination of verb pronoun enclitic that are written together with the hyphen.
Is there a way to allow for these non-wordforming characters as valid in a keyword rendering within the \w…\w* markers without defining them as word-medial punctuation, or is our only other option to deny all the messages for this error?
