file-utf8filenamescanner

#!/usr/bin/python3
import os, sys, re 

count_ascii , count_nonascii, count_validutf8, count_invalidutf8  =  0, 0, 0, 0

# in py3, the data type of paths that come out are str/bytes, following what you handed in,
#    so it's easiest to start by forcing a bytes path from a str argv
for dirname in sys.argv[1:]: 
    for r, ds, fs in os.walk( bytes(dirname,'utf8') ):  #  (assuming that the starting point isn't broken)
        for fn in fs:
            ffn = os.path.join(r, fn)
            match = re.search(br'([\x00-\x1f\x7f-\xff])',ffn)
            has_nonascii = (match != None)
            if has_nonascii:
                #print match.groups()
                count_nonascii += 1
                try:
                    dec = ffn.decode('utf8')
                    count_validutf8 += 1
                    #print( "OKAYUTF8 '%s'"%ffn)
                except UnicodeDecodeError:
                    print( 'BADUTF8 in %r'%ffn)
                    #os.rename(ffn, ffn.decode('latin1').encode('utf8') )
                    count_invalidutf8 += 1
            else:
                count_ascii += 1
                #print( "ASCII %r"%ffn)
                
print( "Pure-ASCII paths: %d"%count_ascii)
print( "Non-ASCII paths: %d, of which"%count_nonascii)
print( "  valid UTF8: %d"%count_validutf8)
print( "  invalid UTF8: %d"%count_invalidutf8)