Post #3038237
2026-05-04 14:02 UTC
@_the_cloud@mastodon.social my system (Debian) had a much smaller dictionary installed. I've replaced it with a huge one with 663373 entries, many dubious.
I can do it in one shot with:
rot13 < /usr/share/dict/words | sort | comm -12 <(sort /usr/share/dict/words) - | grep -P '^[a-z]{2,}$' | grep '^[a-m]'
but I don't think that MacOS's grep supports '-P', PCRE matching.
abjurer/nowhere is still the longest pair
Replies (1)
-
@_the_cloud@mastodon.social 2026-05-04 14:35
@scruss@xoxo.zone My dictionary has all 26 one-letter entries as words, and since A and I don't pair up, I just filter them all out. It stops iterating after M, since all pairs have already been made at that point. But it's still n-squared and stupidly slow; I'm sure there's a better approach. Will see if doing rot13 on the whole dictionary at once is faster.