Subselects would be a pretty awesome feature.

"select name from foo where name not in (select name from ../bar where date < ...)"

I'm usually fine with `find`, but when doing things more interesting than just "find files in this directory that are not in that directory", while uncommon, tend to make me think about my pipeline a bit.

Out of curiosity, I'm interested in how folks would do the "in this not that" folder query. At a gut shot, I'd assume that diff would be used. I'm about to dig through the find man page to see if it has something directly to help.

One tool for this is the 'comm' utility: given two files containing sorted lines, it can output one or more of (1) lines only in file 1, (2) lines only in file 2, and (3) lines common to both files.

Assuming no dups in file1, this outputs lines in file1 that aren't in file2:

    sort file1 file2 file2 | uniq -u

(double file2 is not a typo :)

Or (a little faster):

  comm -23 <(sort file1) <(sort file2)

Great idea! Got the issue up [0], feel free to add to it if you'd like.

[0] https://github.com/kshvmdn/fsql/issues/4

SQL is familiar-ish to a lot of people.

Contrariwise, I rarely see a non-trivial `find` invocation that does not have a few bugs. (including my own!)