more powerful trim_taxa()?

[ctc3d]

Would it be possible for trim_taxa() to also take care of other basic format difficulties like wrong capitalization? Thanks to rando data entry some files I was working on came up with taxa counts like:
Aesculus glabra 1
AESCULUS Glabra 1
Aesculus hippocastanum 14
AESCULUS HIPPOCASTANUM 131

Rationalization of cultivars, etc. is probably too much to ask for -- but maybe some standardization for decently-behaved things like hybrids: Platanus x acerifolia?

[clnsmth]

This seems possible @ctc3d however, count_taxa() addresses this issue (albiet indirectly) by reporting unique formats which can then be used for indexing and replacement.

[ctc3d]

True -- and I was just now doing that in fact -- but for instance here's some more count_taxa() output:
Ginkgo 33
Ginkgo biloba 119
Ginkgo biloba 'Golden Globe' 11
Ginko Biloba 56
Ginko; Biloba 40

Catching 'Ginko' is a lot to ask for, but that semicolon... I'm also getting stray colons and parens and whatnot; insect parts, very small rocks. For any given file, no big deal to do the hand fixes -- but I'll be doing about 300 so any automated junk reduction is a big plus.