--- Day changed Sat Nov 28 2020
08:50 -!- shesek [~shesek@164.90.217.137] has joined ##miniscript     
08:50 -!- shesek [~shesek@164.90.217.137] has quit [Changing host]     
08:50 -!- shesek [~shesek@unaffiliated/shesek] has joined ##miniscript     
09:25 -!- afilini [~user@gateway/tor-sasl/afilini] has quit [Ping timeout: 240 seconds]     
09:27 -!- shesek [~shesek@unaffiliated/shesek] has quit [Ping timeout: 240 seconds]     
09:27 -!- afilini [~user@gateway/tor-sasl/afilini] has joined ##miniscript     
10:13 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has quit [Ping timeout: 240 seconds]     
10:13 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has joined ##miniscript     
15:02 < sanket1729_> darosior: I think we are not some miscommunication, maybe I am not able to convey properly or I am misunderstanding something
15:12 < darosior> sanket1729_: i think too :)
15:12 < darosior> Reading your last comment now
15:19 < darosior> sanket1729_: ok, so i think i got it, sorry for being sloppy: you propose to remove the check as it's wrong because checking the *witness* size against the max *witness script* size ?
15:21 < darosior> That makes sense, but i wonder that we should either do no witness standardness checks at satisfaction time, or if we really want to do some for for Scripts parsed with from_insane() we should be more comprehensive (max size of each stack item, max witness script size, etc..). Here we only check the number of stakc elements
15:35 < darosior> andytoshi, afilini: shoudl i strip the MIT license commented at the top of the file for the descriptor checksum implem ?.. Here is what i did https://github.com/darosior/rust-miniscript/commit/15c756d0df372d1778ff6240f69f2bd686aacf8b
15:41 < andytoshi> darosior: parse_insane shouldn't allow things whose witness scripts exceed the max witness size
15:41 < andytoshi> darosior: as for the copyright, looks fine to me if it's fine with afilini 
15:42 < darosior> Hmm so checking it in satisfaction is effectively redundant ?
15:43 < andytoshi> yes     
15:44 < andytoshi> the witness script is a property of the descriptor, it's a global property
15:44 < darosior> Also, do we want Debug to print the descriptor's checksum too or we could keep it as a way to print it without the checksum part?
15:44 < andytoshi> i think debug should print the checksum
17:45 < darosior> Hmm now i'm in doubt with regard to parsing. Why do we have a parse_descriptor() method in addition to FromStr impl ? Intuitively it seems that parse_descriptor() is ""higher level"" and FromStr is just about the tree, so i implemented the checksum check in parse_descriptor but maybe it should actually be part of the FromStr impl as well ?
18:28 < andytoshi> iirc parse_descriptor was only there to return a key map
18:29 < andytoshi> i think from_str should definitely handle checksums
19:54 < sanket1729_> darosior: indeed. So, for that PR we can remove the check from Segwitv0, but we still need the check_witness function
19:54 < sanket1729_> because it checks whether the actual constructed witness follows the rules
19:55 < sanket1729_> for sane scripts, satisfier would always work, but for insane scripts, it can produce witness that is nonstandard which is why we need that function. 
23:35 -!- ghost43_ [~daer@gateway/tor-sasl/daer] has joined ##miniscript     
23:36 -!- ghost43 [~daer@gateway/tor-sasl/daer] has quit [Ping timeout: 240 seconds]