No subject

not ignore case when matching the description field in a transaction.
My bank uses all caps for the payee names that I download (e.g.,
WASHINGTON GAS), whereas in my register I use mixed case (e.g.,
Washington Gas), and I noticed that the ranking would increase if I
used all caps in the description.

I think the matching algorithm should do a case-insensitive comparison
because I don't imagine there would ever be any significance to
variation only in case.

(B) Prefix Matching

The algorithm also conveniently tries to ignore some of the ending
part of the description because banks commonly append their own
information. However, I think it does this by doing a compare of the
first half of the downloaded description. However, this doesn't work
if the appended information is more than half of the description. I
think it would be better if instead it just checked to see if the user
entry is a prefix of the downloaded entry. That way short descriptions
are not penalized.

Gregg