Punctuation treated as word forming characters

Question

I have created a new transliteration project, and am trying to apply some consistent changes to supplement the one-to-one transliteration already accomplished via TecKit map. The main project is RTL, the transliteration is LTR.

In this project (but not in other projects of either directionality) Wordlist and Search are treating punctuation marks as word-forming characters. e.g.:

image871×114 16 KB

If I do a “Whole Word Only” search on “amadeye” I will only get the three hits that are not followed by punctuation.

I have checked and rechecked the Language, Number, and Quotation settings, as well as Character Inventory and Punctuation Inventory, and don’t see any reason for this behavior.

I am stumped. Any suggestions?

Answer 1

Make sure that in the Alphabetic Character list of the language settings you do not have any punctuation listed. Also check the word medial punctuation list (Other Characters tab) to verify that these punctuation marks are not listed.

Comments

Ah!! That was it. I had copied the “Other Characters” from the parent project, which had all punctuation listed in the “word medial” spot (though I changed the RTL ones to LTR). What punctuation should be listed there?

---

LTR projects commonly have apostrophe and hyphen. My RTL projects don’t usually have anything. But you might get a complex script project with a special type of word joiner in that location.

---

Shegnada has given a good answer. The short answer is any punctuation that can occur in the middle of a word. This allows the Word list tool to identify those words.

---

Very helpful! Many thanks to you both.