What's the best way to translate MS generated translations of user input to what the user actually typed prior to insert or update into PG backend table ?

[David Gauthier <davegauthierpg@xxxxxxxxx>]

psql (11.5, server 11.3) on linux

I'm using MS-Access as a Windows front-end to a PG DB table through ODBC (PostgreSQL Unicode ODBC Driver).  Seems to be working fine except for when users enter "..." as part of a string, MS (in it's infinite wisdom) decides to translate that to what emacs is describing as...

character: … (displayed as …) (codepoint 8230, #o20046, #x2026)
preferred charset: unicode (Unicode (ISO10646))
code point in charset: 0x2026

It makes this translation for some users but not others, so I'm assuming it has something to do with how individuals set up their Windows env.  No matter, assume we can't have them change their environment in whatever soln we come up with. 

I want to change the 0x2026 to the 3 periods which the user entered.

I was thinking of a pre-insert and pre-update trigger which could make the translation.  But I'd rather not try to do this one char at a time... translate "..." today to fix today's issue, then "--" tomorrow when that pops up, then the way MS translates double quotes the next day, etc... .  Is there an elegant way to do this ?

Thanks In Advance for any help !