This does not address the problem distinguishing
hyphen, long(Em), and short(En) dashes,
nor comma and decimal point in numbers (1.000 = 1,000),
nor underline vs underscore,
nor parenthesis, curly & square brackets,
nor of adjacent character ambiguities in many fonts like
rn = m,
cl = d
vv = w
VV = W
0. = Q
And there are surely others.