rc-non-ascii

#!/usr/bin/env raku
# ID characters & counts, from file(s), showing anything not pure ASCII (i.e. .ords.any > 127)
## 2019-11-09
#  croaks on malformed UTF-8, so helpful in tracking that down
#  also useful for ID'ing Unicode that won't display properly
#  2022-12-18 segfault? bug in the volume of data piped

#  191.765u 0.711s 3:11.99 100.2% bare junction 'any', ouch
#   22.518u 0.411s 0:22.43 102.1% .sort.tail
#    4.214u 0.284s 0:04.05 110.8% just .ord (but then miss some combining chars)

for @*ARGS -> $f {
   #next unless $f.IO ~~ :f and my %codes = $f.IO.slurp.comb.grep(*.ords.sort.tail > 127).Bag;
    # 'Bag' replaces idiom '%codes{$_}++ for'

    # inefficient kludge for now
    next unless $f.IO ~~ :f;
    my %codes;
    for $f.IO.slurp.comb { %codes{$_}++ if .ords.tail > 127; }

    next unless keys %codes;
    say "\n$f";
    printf "%1s: %5d %s\n", $_, %codes{$_}, .uninames.join(', ') for %codes.keys.sort;

    CATCH { default { say qq{Died in "$f" with "$_" } } }
}