Forcing two token vords near stems and leaves

[RobGea]

iirc,  MarcoP and EMS did something about glyphs and text intrusions.

I made a quick pass (not thorough) of words that occur between text-intrusion markers ('<->')in EVA ZL3-n transliteration.
Excluded are any texts from  Start-of-line (SOL) to marker  and text from  marker to End-of-line (EOL).
As attachment.
  filer_words3_edit.txt (Size: 3.29 KB / Downloads: 9)

I have not worked my way through this list entirely but some noteworthy things.
<->sals<->    **all words between markers are single occurrences
<->sky<->
<->saro<->
<->ydey<->
<->ytorory<->
<->chkeeod<->
<->odaiil<->
<->ykoda<->
<->ykeeb<->
<->chshy<->

yk  --almost all are '.yk,' or 'yk['  and 1 standalone
sy  --almost always at EOL or before a text intrusion
olk  --rare
dchd --rare

I was thinking if they were filler words then those words may be only found between text intrusions but there are only a few of those.

You could make a complete collection of these words find what word comes before or after and see if there is a difference between normal text and text-intrusion text.

EVA-'sy' is quite interesting because  of where it occurs
sy before markers  9
sy between markers  5
sy as eol    19
sy in normal text  4