Rethink the ArtiPath pattern matching APIs
The following discussion from !1677 (merged) should be addressed:
-
@Diziet started a discussion: (+1 comment) I don't thnk C Tor even store
K_hs_blind_id
on disk. At least, that's what my reading oftor(1)
from manpages.debian.org suggests.Anyway, I think the need to write this rather strange pattern demonstrates that the pattern API is wrong. You should be allowed to ask for keys by only their
ArtiPath
, and maybe only by their C Tor path.Suggested disposition: add a TODO HSS, or a ticket. Possibly invite me to demonstrate what I think the alternative looks like.