Writing huge Sets() to disk

[Tim Peters]

[Martin MOKREJŠ]
> This comm(1) approach doesn't work for me. It somehow fails to
> detect common entries when the offset is too big.
>
> file 1:
>
> A
> F
> G
> I
> K
> M
> N
> R
> V
> AA
> AI
> FG
> FR
> GF
> GI
> GR
> IG
> IK
> IN
> IV
> KI
> MA
> NG
> RA
> RI
> VF
> AIK
> FGR
> FRA
> GFG
> GIN
> GRI
> IGI
> IGR
> IKI
> ING
> IVF
> KIG
> MAI
> NGF
> RAA
> RIG
> 
> file 2:
> 
> W
> W
> W
> W
> W
> W
> W
> W
> W
> W
> AA
> AI
> FG
> FR
> GF
> GI
> GR
> IG
> IK
> IN
> IV
> KI
> MA
> NG
> RA
> RI
> VF
> AAAAA
> AAAAA
> AAAAA
> AAAAA
> AAAAA
> AAAAA
> AAAAA
> AAAAA
> AAAAA
> AAAAA
> AAAAA
> AAAAA

I'll repeat:

>> As I mentioned before, if you store keys in sorted text files ...

Those files aren't in sorted order, so of course `comm` can't do
anything useful with them.  Do `man sort`; sorting is not optional
here.