the completion appears to be stripping diacritics

[ultrablue]

Hi:

Here's what I'm seeing.

I'm using a custom word list. It contains several words that include diacritics. I've added a test string to the custom word list:
āṣīṛū

With "Minimum word trigger length" set to 3, completr offers the following completion:

asiṛū
^^^

And that's what's inserted into the document.

With "Minimum word trigger length" set to 2, completr offers the following completion:

asīṛū
^^

And that's what's inserted into the document.

With "Minimum word trigger length" set to 1, completr offers the following completion:

aṣīṛū
^

And that's what's inserted into the document.

Expected behavior in all three cases:
The actual string from my custom word list should be inserted:
āṣīṛū

In other words, the number of characters that are matched are being replaced by their non-diacritic cousins.

What I've done:
Toggled "Ignore diacritics when filtering" several times.

Added the following to the Word Character Regex setting:
"Āā"

Neither seemed to make any difference.

Please let me know if there's anything I can do to help. This problem is adding a fair amount of friction to my obsidian flow.

With respect,
Greg

---

obsidian-completr verision: 3.2.0

Here's my current data.json:

{
  "characterRegex": "a-zA-ZöäüÖÄÜßĀā",
  "maxLookBackDistance": 50,
  "autoFocus": true,
  "autoTrigger": false,
  "minWordLength": 4,
  "minWordTriggerLength": 3,
  "wordInsertionMode": "Ignore-Case & Append",
  "ignoreDiacriticsWhenFiltering": true,
  "latexProviderEnabled": false,
  "latexTriggerInCodeBlocks": false,
  "latexMinWordTriggerLength": 2,
  "latexIgnoreCase": true,
  "fileScannerProviderEnabled": false,
  "fileScannerScanCurrent": true,
  "wordListProviderEnabled": true,
  "frontMatterProviderEnabled": true,
  "frontMatterTagAppendSuffix": true,
  "frontMatterIgnoreCase": true,
  "calloutProviderEnabled": true,
  "calloutProviderSource": "Completr"
}